experiment.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x800 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Experiment 1: Distribution of full-precision base hypervectors\n",
    "\n",
    "import dna\n",
    "import hdgim\n",
    "import torch\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "torch.set_printoptions(sci_mode=False)\n",
    "\n",
    "# [begin] hyperparameters\n",
    "HYPERVECTOR_DIMENSION = 1000  # denoted D in paper\n",
    "DNA_SEQUENCE_LENGTH = 1000\n",
    "DNA_SUBSEQUENCE_LENGTH = 20  # denoted n in paper\n",
    "BIT_PRECISION = 2  # denoted B in paper\n",
    "NOISE_PROBABILTY = 0.1  # denoted p in paper, probability: 0 ~ 1, simulate FeFET noise\n",
    "# [end] hyperparameters\n",
    "\n",
    "# create dataset\n",
    "dataset = dna.DNADataset(DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH)\n",
    "\n",
    "# create model\n",
    "model = hdgim.HDGIM(HYPERVECTOR_DIMENSION, DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH, BIT_PRECISION, NOISE_PROBABILTY)\n",
    "model.set_dataset(dataset)\n",
    "model.create_voltage_matrix()\n",
    "model.create_base_hypervectors()\n",
    "model.create_dna_sequence()\n",
    "model.create_dna_subsequences()\n",
    "model.bind()\n",
    "\n",
    "# draw plot\n",
    "# Create a 2x2 grid of subplots\n",
    "fig, axes = plt.subplots(2, 2, figsize=(10, 8))\n",
    "\n",
    "# Plot the histograms in each subplot\n",
    "axes[0, 0].hist(model.base_hypervectors[dna.DNA.A].numpy(), bins=30, density=True, alpha=0.7, color='blue', edgecolor='black')\n",
    "axes[0, 0].set_title('Value of Base Hypervector A')\n",
    "axes[0, 0].set_xlabel('Value')\n",
    "axes[0, 0].set_ylabel('Density')\n",
    "\n",
    "axes[0, 1].hist(model.base_hypervectors[dna.DNA.C].numpy(), bins=30, density=True, alpha=0.7, color='green', edgecolor='black')\n",
    "axes[0, 1].set_title('Value of Base Hypervector C')\n",
    "axes[0, 1].set_xlabel('Value')\n",
    "axes[0, 1].set_ylabel('Density')\n",
    "\n",
    "axes[1, 0].hist(model.base_hypervectors[dna.DNA.G].numpy(), bins=30, density=True, alpha=0.7, color='orange', edgecolor='black')\n",
    "axes[1, 0].set_title('Value of Base Hypervector G')\n",
    "axes[1, 0].set_xlabel('Value')\n",
    "axes[1, 0].set_ylabel('Density')\n",
    "\n",
    "axes[1, 1].hist(model.base_hypervectors[dna.DNA.T].numpy(), bins=30, density=True, alpha=0.7, color='red', edgecolor='black')\n",
    "axes[1, 1].set_title('Value of Base Hypervector T')\n",
    "axes[1, 1].set_xlabel('Value')\n",
    "axes[1, 1].set_ylabel('Density')\n",
    "\n",
    "# Add a title for the entire figure\n",
    "plt.suptitle('Distribution of Full-Precision Base Hypervectors')\n",
    "\n",
    "# Adjust layout to prevent clipping of titles\n",
    "plt.tight_layout(rect=[0, 0, 1, 0.96])\n",
    "\n",
    "# Show the plot\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Experiment 2: Distribution of full-precision bound hypervector\n",
    "\n",
    "import dna\n",
    "import hdgim\n",
    "import torch\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "torch.set_printoptions(sci_mode=False)\n",
    "\n",
    "# [begin] hyperparameters\n",
    "HYPERVECTOR_DIMENSION = 1000  # denoted D in paper\n",
    "DNA_SEQUENCE_LENGTH = 1000\n",
    "DNA_SUBSEQUENCE_LENGTH = 20  # denoted n in paper\n",
    "BIT_PRECISION = 2  # denoted B in paper\n",
    "NOISE_PROBABILTY = 0.1  # denoted p in paper, probability: 0 ~ 1, simulate FeFET noise\n",
    "# [end] hyperparameters\n",
    "\n",
    "# create dataset\n",
    "dataset = dna.DNADataset(DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH)\n",
    "\n",
    "# create model\n",
    "model = hdgim.HDGIM(HYPERVECTOR_DIMENSION, DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH, BIT_PRECISION, NOISE_PROBABILTY)\n",
    "model.set_dataset(dataset)\n",
    "model.create_voltage_matrix()\n",
    "model.create_base_hypervectors()\n",
    "model.create_dna_sequence()\n",
    "model.create_dna_subsequences()\n",
    "model.bind()\n",
    "\n",
    "# draw plot\n",
    "plt.hist(model.encoded_hypervector.numpy(), bins=30, density=True, alpha=0.7, color='blue', edgecolor='black')\n",
    "\n",
    "plt.title('Distribution of Full-Precision Bound Hypervector')\n",
    "plt.xlabel('Value')\n",
    "plt.ylabel('Density')\n",
    "\n",
    "# Show the plot\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Experiment 3: Distribution of 2-bit precision bound hypervector, using min-max\n",
    "\n",
    "import dna\n",
    "import hdgim\n",
    "import torch\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "torch.set_printoptions(sci_mode=False)\n",
    "\n",
    "# [begin] hyperparameters\n",
    "HYPERVECTOR_DIMENSION = 1000  # denoted D in paper\n",
    "DNA_SEQUENCE_LENGTH = 1000\n",
    "DNA_SUBSEQUENCE_LENGTH = 20  # denoted n in paper\n",
    "BIT_PRECISION = 2  # denoted B in paper\n",
    "NOISE_PROBABILTY = 0.1  # denoted p in paper, probability: 0 ~ 1, simulate FeFET noise\n",
    "# [end] hyperparameters\n",
    "\n",
    "# create dataset\n",
    "dataset = dna.DNADataset(DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH)\n",
    "\n",
    "# create model\n",
    "model = hdgim.HDGIM(HYPERVECTOR_DIMENSION, DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH, BIT_PRECISION, NOISE_PROBABILTY)\n",
    "model.set_dataset(dataset)\n",
    "model.create_voltage_matrix()\n",
    "model.create_base_hypervectors()\n",
    "model.create_dna_sequence()\n",
    "model.create_dna_subsequences()\n",
    "model.bind()\n",
    "model.quantize_min_max()\n",
    "\n",
    "# draw plot\n",
    "plt.hist(model.quantized_hypervector.numpy(), bins=30, density=True, alpha=0.7, color='blue', edgecolor='black')\n",
    "\n",
    "plt.title('Distribution of 2-bit Precision Bound Hypervector')\n",
    "plt.xlabel('Value')\n",
    "plt.ylabel('Density')\n",
    "\n",
    "# Show the plot\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Experiment 4: Distribution of 2-bit precision bound hypervector, using cdf\n",
    "\n",
    "import dna\n",
    "import hdgim\n",
    "import torch\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "torch.set_printoptions(sci_mode=False)\n",
    "\n",
    "# [begin] hyperparameters\n",
    "HYPERVECTOR_DIMENSION = 1000  # denoted D in paper\n",
    "DNA_SEQUENCE_LENGTH = 1000\n",
    "DNA_SUBSEQUENCE_LENGTH = 20  # denoted n in paper\n",
    "BIT_PRECISION = 2  # denoted B in paper\n",
    "NOISE_PROBABILTY = 0.1  # denoted p in paper, probability: 0 ~ 1, simulate FeFET noise\n",
    "# [end] hyperparameters\n",
    "\n",
    "# create dataset\n",
    "dataset = dna.DNADataset(DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH)\n",
    "\n",
    "# create model\n",
    "model = hdgim.HDGIM(HYPERVECTOR_DIMENSION, DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH, BIT_PRECISION, NOISE_PROBABILTY)\n",
    "model.set_dataset(dataset)\n",
    "model.create_voltage_matrix()\n",
    "model.create_base_hypervectors()\n",
    "model.create_dna_sequence()\n",
    "model.create_dna_subsequences()\n",
    "model.bind()\n",
    "model.quantize_cdf()\n",
    "\n",
    "# draw plot\n",
    "plt.hist(model.quantized_hypervector.numpy(), bins=30, density=True, alpha=0.7, color='blue', edgecolor='black')\n",
    "\n",
    "plt.title('Distribution of 2-bit Precision Bound Hypervector')\n",
    "plt.xlabel('Value')\n",
    "plt.ylabel('Density')\n",
    "\n",
    "# Show the plot\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Experiment 5: Distribution of 2-bit precision bound hypervector\n",
    "\n",
    "import dna\n",
    "import hdgim\n",
    "import torch\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "torch.set_printoptions(sci_mode=False)\n",
    "\n",
    "# [begin] hyperparameters\n",
    "HYPERVECTOR_DIMENSION = 1000  # denoted D in paper\n",
    "DNA_SEQUENCE_LENGTH = 1000\n",
    "DNA_SUBSEQUENCE_LENGTH = 20  # denoted n in paper\n",
    "BIT_PRECISION = 2  # denoted B in paper\n",
    "NOISE_PROBABILTY = 0.1  # denoted p in paper, probability: 0 ~ 1, simulate FeFET noise\n",
    "# [end] hyperparameters\n",
    "\n",
    "# create dataset\n",
    "dataset = dna.DNADataset(DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH)\n",
    "\n",
    "# create model\n",
    "model = hdgim.HDGIM(HYPERVECTOR_DIMENSION, DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH, BIT_PRECISION, NOISE_PROBABILTY)\n",
    "model.set_dataset(dataset)\n",
    "model.create_voltage_matrix()\n",
    "model.create_base_hypervectors()\n",
    "model.create_dna_sequence()\n",
    "model.create_dna_subsequences()\n",
    "model.bind()\n",
    "model.quantize_min_max()\n",
    "model.noise()\n",
    "\n",
    "# draw plot\n",
    "plt.hist(model.quantized_hypervector.numpy(), bins=30, density=True, alpha=0.7, color='blue', edgecolor='black')\n",
    "\n",
    "plt.title('Distribution of 2-bit-Precision Bound Hypervector')\n",
    "plt.xlabel('Value')\n",
    "plt.ylabel('Density')\n",
    "\n",
    "# Show the plot\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train size: 1079\n",
      "Epoch 0: Accuracy 84.62%\n",
      "Average true similarity: -0.0009819166734814644, Average false similarity: -0.0069231330417096615\n",
      "True negative: 12, True positive: 901, False negative: 86, False positive: 80\n",
      "Epoch 1: Accuracy 86.28%\n",
      "Average true similarity: 0.004435630515217781, Average false similarity: -0.021236449480056763\n",
      "True negative: 19, True positive: 912, False negative: 79, False positive: 69\n",
      "Epoch 2: Accuracy 89.16%\n",
      "Average true similarity: 0.004473784938454628, Average false similarity: -0.03390280157327652\n",
      "True negative: 24, True positive: 938, False negative: 74, False positive: 43\n",
      "Epoch 3: Accuracy 91.2%\n",
      "Average true similarity: 0.005287759006023407, Average false similarity: -0.04816349595785141\n",
      "True negative: 37, True positive: 947, False negative: 61, False positive: 34\n",
      "Epoch 4: Accuracy 90.82%\n",
      "Average true similarity: 0.0034011502284556627, Average false similarity: -0.05120132490992546\n",
      "True negative: 33, True positive: 947, False negative: 65, False positive: 34\n",
      "Epoch 5: Accuracy 91.75%\n",
      "Average true similarity: 0.004838844295591116, Average false similarity: -0.0593191422522068\n",
      "True negative: 38, True positive: 952, False negative: 60, False positive: 29\n",
      "Epoch 6: Accuracy 91.29%\n",
      "Average true similarity: 0.004143522586673498, Average false similarity: -0.056434422731399536\n",
      "True negative: 37, True positive: 948, False negative: 61, False positive: 33\n",
      "Epoch 7: Accuracy 92.68%\n",
      "Average true similarity: 0.0024589623790234327, Average false similarity: -0.059918563812971115\n",
      "True negative: 46, True positive: 954, False negative: 52, False positive: 27\n",
      "Epoch 8: Accuracy 92.68%\n",
      "Average true similarity: 0.004798319656401873, Average false similarity: -0.05905062332749367\n",
      "True negative: 40, True positive: 960, False negative: 58, False positive: 21\n",
      "Epoch 9: Accuracy 95.0%\n",
      "Average true similarity: 0.0034724457655102015, Average false similarity: -0.0633263811469078\n",
      "True negative: 56, True positive: 969, False negative: 42, False positive: 12\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Experiment 6: Similarity Distribution of full-precision bound hypervector\n",
    "\n",
    "import dna\n",
    "import hdgim\n",
    "import torch\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "torch.set_printoptions(sci_mode=False)\n",
    "\n",
    "# [begin] hyperparameters\n",
    "HYPERVECTOR_DIMENSION = 1000  # denoted D in paper\n",
    "DNA_SEQUENCE_LENGTH = 1000\n",
    "DNA_SUBSEQUENCE_LENGTH = 20  # denoted n in paper\n",
    "BIT_PRECISION = 2  # denoted B in paper\n",
    "NOISE_PROBABILTY = 0.1  # denoted p in paper, probability: 0 ~ 1, simulate FeFET noise\n",
    "EPOCHES = 10\n",
    "FUNCTION = 'cosine'\n",
    "LEARNING_RATE = 10000\n",
    "THRESHOLD = -0.07\n",
    "# [end] hyperparameters\n",
    "\n",
    "# create dataset\n",
    "dataset = dna.DNADataset(DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH)\n",
    "\n",
    "# create model\n",
    "model = hdgim.HDGIM(HYPERVECTOR_DIMENSION, DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH, BIT_PRECISION, NOISE_PROBABILTY)\n",
    "model.set_dataset(dataset)\n",
    "model.create_voltage_matrix()\n",
    "model.create_base_hypervectors()\n",
    "model.create_dna_sequence()\n",
    "model.create_dna_subsequences()\n",
    "model.bind()\n",
    "model.quantize_min_max()\n",
    "model.noise()\n",
    "\n",
    "# train\n",
    "accuracies, true_similarities, false_similarities = model.train(EPOCHES, LEARNING_RATE, THRESHOLD, FUNCTION, True, True, True)\n",
    "\n",
    "# draw histogram which shows true similarity distribution and false similarity distribution\n",
    "plt.hist(true_similarities[EPOCHES-1], bins=30, density=True, alpha=0.5, color='blue', edgecolor='black', label='true')\n",
    "plt.hist(false_similarities[EPOCHES-1], bins=30, density=True, alpha=0.5, color='red', edgecolor='black', label='false')\n",
    "\n",
    "# add explanation of color\n",
    "plt.legend(prop={'size': 10})\n",
    "\n",
    "plt.title('Similarity Distribution Graph')\n",
    "plt.xlabel('Similarity')\n",
    "plt.ylabel('Density')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train size: 1079\n",
      "Epoch 0: Accuracy 9.08%\n",
      "Average true similarity: -127497.734375, Average false similarity: -127147.65625\n",
      "True negative: 98, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 1: Accuracy 9.08%\n",
      "Average true similarity: -127505.5625, Average false similarity: -127111.625\n",
      "True negative: 98, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 2: Accuracy 9.08%\n",
      "Average true similarity: -127494.265625, Average false similarity: -127136.0078125\n",
      "True negative: 98, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 3: Accuracy 9.08%\n",
      "Average true similarity: -127480.1953125, Average false similarity: -127071.578125\n",
      "True negative: 98, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 4: Accuracy 9.08%\n",
      "Average true similarity: -127473.921875, Average false similarity: -127056.078125\n",
      "True negative: 98, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 5: Accuracy 9.08%\n",
      "Average true similarity: -127476.7734375, Average false similarity: -127051.09375\n",
      "True negative: 98, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 6: Accuracy 9.08%\n",
      "Average true similarity: -127463.1015625, Average false similarity: -127038.921875\n",
      "True negative: 98, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 7: Accuracy 9.08%\n",
      "Average true similarity: -127451.734375, Average false similarity: -127065.03125\n",
      "True negative: 98, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 8: Accuracy 9.08%\n",
      "Average true similarity: -127460.8359375, Average false similarity: -127032.0625\n",
      "True negative: 98, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 9: Accuracy 9.08%\n",
      "Average true similarity: -127443.2109375, Average false similarity: -127003.5703125\n",
      "True negative: 98, True positive: 0, False negative: 0, False positive: 981\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Experiment 7: Similarity Distribution of 2-bit precision bound hypervector\n",
    "\n",
    "import dna\n",
    "import hdgim\n",
    "import torch\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "torch.set_printoptions(sci_mode=False)\n",
    "\n",
    "# [begin] hyperparameters\n",
    "HYPERVECTOR_DIMENSION = 1000  # denoted D in paper\n",
    "DNA_SEQUENCE_LENGTH = 1000\n",
    "DNA_SUBSEQUENCE_LENGTH = 20  # denoted n in paper\n",
    "BIT_PRECISION = 8  # denoted B in paper\n",
    "NOISE_PROBABILTY = 0.1  # denoted p in paper, probability: 0 ~ 1, simulate FeFET noise\n",
    "EPOCHES = 10\n",
    "FUNCTION = 'hamming'\n",
    "LEARNING_RATE = 10\n",
    "THRESHOLD = -131\n",
    "# [end] hyperparameters\n",
    "\n",
    "# create dataset\n",
    "dataset = dna.DNADataset(DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH)\n",
    "\n",
    "# create model\n",
    "model = hdgim.HDGIM(HYPERVECTOR_DIMENSION, DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH, BIT_PRECISION, 0)\n",
    "model.set_dataset(dataset)\n",
    "model.create_base_hypervectors()\n",
    "model.create_dna_sequence()\n",
    "model.create_dna_subsequences()\n",
    "model.bind()\n",
    "model.quantize_cdf()\n",
    "model.noise()\n",
    "\n",
    "# train\n",
    "accuracies, true_similarities, false_similarities = model.train(EPOCHES, LEARNING_RATE, THRESHOLD, FUNCTION, False, True, True)\n",
    "\n",
    "# draw histogram which shows true similarity distribution and false similarity distribution\n",
    "plt.hist(true_similarities[EPOCHES-1], bins=30, density=True, alpha=0.5, color='blue', edgecolor='black', label='true')\n",
    "plt.hist(false_similarities[EPOCHES-1], bins=30, density=True, alpha=0.5, color='red', edgecolor='black', label='false')\n",
    "\n",
    "# add explanation of color\n",
    "plt.legend(prop={'size': 10})\n",
    "\n",
    "plt.title('Similarity Distribution Graph')\n",
    "plt.xlabel('Similarity')\n",
    "plt.ylabel('Density')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Experiment 8: Sorting Encoded Hypervector, No Noise\n",
    "\n",
    "import dna\n",
    "import hdgim\n",
    "import torch\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "torch.set_printoptions(sci_mode=False)\n",
    "\n",
    "# [begin] hyperparameters\n",
    "HYPERVECTOR_DIMENSION = 1000  # denoted D in paper\n",
    "DNA_SEQUENCE_LENGTH = 1000\n",
    "DNA_SUBSEQUENCE_LENGTH = 20  # denoted n in paper\n",
    "BIT_PRECISION = 2  # denoted B in paper\n",
    "NOISE_PROBABILTY = 0.1  # denoted p in paper, probability: 0 ~ 1, simulate FeFET noise\n",
    "EPOCHES = 10\n",
    "FUNCTION = 'hamming'\n",
    "LEARNING_RATE = 10\n",
    "THRESHOLD = -131\n",
    "# [end] hyperparameters\n",
    "\n",
    "# create dataset\n",
    "dataset = dna.DNADataset(DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH)\n",
    "\n",
    "# create model\n",
    "model = hdgim.HDGIM(HYPERVECTOR_DIMENSION, DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH, BIT_PRECISION, 0)\n",
    "model.set_dataset(dataset)\n",
    "model.create_base_hypervectors()\n",
    "model.create_dna_sequence()\n",
    "model.create_dna_subsequences()\n",
    "model.bind()\n",
    "sorted_encoded_hypervector, _ = torch.sort(model.encoded_hypervector)\n",
    "# model.noise()\n",
    "\n",
    "# draw plot\n",
    "# Create a 2x2 grid of subplots\n",
    "fig, axes = plt.subplots(1, 2, figsize=(10, 8))\n",
    "\n",
    "# Plot the histograms in each subplot\n",
    "axes[0].plot(model.encoded_hypervector.numpy())\n",
    "axes[0].set_title('Value of Unsorted Encoded Hypervector')\n",
    "axes[0].set_xlabel('Index')\n",
    "axes[0].set_ylabel('Value')\n",
    "\n",
    "axes[1].plot(sorted_encoded_hypervector)\n",
    "axes[1].set_title('Value of Sorted Encoded Hypervector')\n",
    "axes[1].set_xlabel('Index')\n",
    "axes[1].set_ylabel('Value')\n",
    "\n",
    "# Add a title for the entire figure\n",
    "plt.suptitle('Value of Unsorted/Sorted Hypervectors')\n",
    "\n",
    "# Adjust layout to prevent clipping of titles\n",
    "plt.tight_layout(rect=[0, 0, 1, 0.96])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1500 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Experiment 9: CDF Test, No Noise\n",
    "\n",
    "import dna\n",
    "import hdgim\n",
    "import torch\n",
    "import matplotlib.pyplot as plt\n",
    "import scipy.stats as stats\n",
    "import numpy as np\n",
    "\n",
    "torch.set_printoptions(sci_mode=False)\n",
    "\n",
    "# [begin] hyperparameters\n",
    "HYPERVECTOR_DIMENSION = 1000  # denoted D in paper\n",
    "DNA_SEQUENCE_LENGTH = 1000\n",
    "DNA_SUBSEQUENCE_LENGTH = 20  # denoted n in paper\n",
    "BIT_PRECISION = 2  # denoted B in paper\n",
    "NOISE_PROBABILTY = 0.1  # denoted p in paper, probability: 0 ~ 1, simulate FeFET noise\n",
    "EPOCHES = 10\n",
    "FUNCTION = 'hamming'\n",
    "LEARNING_RATE = 10\n",
    "THRESHOLD = -131\n",
    "# [end] hyperparameters\n",
    "\n",
    "# create dataset\n",
    "dataset = dna.DNADataset(DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH)\n",
    "\n",
    "# create model\n",
    "model = hdgim.HDGIM(HYPERVECTOR_DIMENSION, DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH, BIT_PRECISION, 0)\n",
    "model.set_dataset(dataset)\n",
    "model.create_base_hypervectors()\n",
    "model.create_dna_sequence()\n",
    "model.create_dna_subsequences()\n",
    "model.bind()\n",
    "# model.noise()\n",
    "\n",
    "sorted_encoded_hypervector, indices = torch.sort(model.encoded_hypervector)\n",
    "np_sorted_encoded_hypervector = sorted_encoded_hypervector.numpy()\n",
    "np_normalized_sorted_encoded_hypervector = (np_sorted_encoded_hypervector - np_sorted_encoded_hypervector.mean()) / np_sorted_encoded_hypervector.std()\n",
    "binary_width = 1.0 / 2**BIT_PRECISION\n",
    "\n",
    "rv = stats.norm(0, 1) # assume standard normal distribution\n",
    "yl = rv.pdf(np_normalized_sorted_encoded_hypervector)\n",
    "cdf = rv.cdf(np_normalized_sorted_encoded_hypervector)\n",
    "quantized_cdf = np.floor(cdf / binary_width)\n",
    "quantized_hypervector = torch.from_numpy(quantized_cdf)[torch.argsort(indices)]\n",
    "\n",
    "# draw plot\n",
    "fig, axes = plt.subplots(5, 2, figsize=(10, 15))\n",
    "\n",
    "axes[0, 0].plot(np_sorted_encoded_hypervector)\n",
    "axes[0, 0].set_title(\"Result Value of Sorted Encoded Hypervector\")\n",
    "axes[0, 0].set_xlabel('Index')\n",
    "axes[0, 0].set_ylabel('Value')\n",
    "\n",
    "axes[0, 1].hist(np_sorted_encoded_hypervector, bins=30, density=True, color='blue', edgecolor='black', label='true')\n",
    "axes[0, 1].set_title(\"Result Distribution of Sorted Encoded Hypervector\")\n",
    "axes[0, 1].set_xlabel('Value')\n",
    "axes[0, 1].set_ylabel('Density')\n",
    "\n",
    "axes[1, 0].plot(np_normalized_sorted_encoded_hypervector, yl)\n",
    "axes[1, 0].set_title('Result Value of PDF')\n",
    "axes[1, 0].set_xlabel('Value (Normalized)')\n",
    "axes[1, 0].set_ylabel('Probablity')\n",
    "\n",
    "axes[1, 1].hist(yl, bins=30, density=True, color='blue', edgecolor='black', label='true')\n",
    "axes[1, 1].set_title('Result Distribution of PDF')\n",
    "axes[1, 1].set_xlabel('Probablity')\n",
    "axes[1, 1].set_ylabel('Density')\n",
    "\n",
    "axes[2, 0].plot(cdf)\n",
    "axes[2, 0].set_title('Result Value of CDF')\n",
    "axes[2, 0].set_xlabel('Index')\n",
    "axes[2, 0].set_ylabel('Probability')\n",
    "\n",
    "axes[2, 1].hist(cdf, bins=30, density=True, color='blue', edgecolor='black', label='true')\n",
    "axes[2, 1].set_title('Result Distribution of CDF')\n",
    "axes[2, 1].set_xlabel('Probablity')\n",
    "axes[2, 1].set_ylabel('Density')\n",
    "\n",
    "axes[3, 0].plot(quantized_cdf)\n",
    "axes[3, 0].set_title('Result Quantized Value of CDF')\n",
    "axes[3, 0].set_xlabel('Value')\n",
    "axes[3, 0].set_ylabel('Probability')\n",
    "\n",
    "axes[3, 1].hist(quantized_cdf, bins=30, density=True, color='blue', edgecolor='black', label='true')\n",
    "axes[3, 1].set_title('Result Distribution of Quantized CDF')\n",
    "axes[3, 1].set_xlabel('Value')\n",
    "axes[3, 1].set_ylabel('Density')\n",
    "\n",
    "axes[4, 0].plot(quantized_hypervector)\n",
    "axes[4, 0].set_title('Result Value of Quantized Hypervector')\n",
    "axes[4, 0].set_xlabel('Value')\n",
    "axes[4, 0].set_ylabel('Probability')\n",
    "\n",
    "axes[4, 1].hist(quantized_hypervector, bins=30, density=True, color='blue', edgecolor='black', label='true')\n",
    "axes[4, 1].set_title('Result Distribution of Quantized CDF')\n",
    "axes[4, 1].set_xlabel('Value')\n",
    "axes[4, 1].set_ylabel('Density')\n",
    "\n",
    "# Add a title for the entire figure\n",
    "plt.suptitle('Value of Unsorted/Sorted Hypervectors')\n",
    "\n",
    "# Adjust layout to prevent clipping of titles\n",
    "plt.tight_layout(rect=[0, 0, 1, 0.96])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0.98, 'Quantizing Test')"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Experiment 10: Quantizing Test, 10% Noise\n",
    "\n",
    "import dna\n",
    "import hdgim\n",
    "import torch\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "torch.set_printoptions(sci_mode=False)\n",
    "\n",
    "# [begin] hyperparameters\n",
    "HYPERVECTOR_DIMENSION = 1000  # denoted D in paper\n",
    "DNA_SEQUENCE_LENGTH = 1000\n",
    "DNA_SUBSEQUENCE_LENGTH = 20  # denoted n in paper\n",
    "BIT_PRECISION = 2  # denoted B in paper\n",
    "NOISE_PROBABILTY = 0.1  # denoted p in paper, probability: 0 ~ 1, simulate FeFET noise\n",
    "EPOCHES = 10\n",
    "FUNCTION = 'hamming'\n",
    "LEARNING_RATE = 10\n",
    "THRESHOLD = -131\n",
    "# [end] hyperparameters\n",
    "\n",
    "# create dataset\n",
    "dataset = dna.DNADataset(DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH)\n",
    "\n",
    "# create model\n",
    "model = hdgim.HDGIM(HYPERVECTOR_DIMENSION, DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH, BIT_PRECISION, NOISE_PROBABILTY)\n",
    "model.set_dataset(dataset)\n",
    "model.create_base_hypervectors()\n",
    "model.create_dna_sequence()\n",
    "model.create_dna_subsequences()\n",
    "model.bind()\n",
    "model.quantize_cdf()\n",
    "model.noise()\n",
    "\n",
    "fig, axes = plt.subplots(1, 2, figsize=(10, 5))\n",
    "\n",
    "# Plot the histograms in each subplot\n",
    "axes[0].plot(model.quantized_hypervector.numpy())\n",
    "axes[0].set_title('Value of Quantized Hypervector')\n",
    "axes[0].set_xlabel('Index')\n",
    "axes[0].set_ylabel('Value')\n",
    "\n",
    "axes[1].hist(model.quantized_hypervector.numpy(), bins=30, density=True, color='blue', edgecolor='black', label='true')\n",
    "axes[1].set_title('Distribution of Quantized Hypervector')\n",
    "axes[1].set_xlabel('Value')\n",
    "axes[1].set_ylabel('Density')\n",
    "\n",
    "plt.suptitle('Quantizing Test')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train size: 1962\n",
      "Epoch 0: Accuracy 50.0%\n",
      "Average true similarity: -4.166407747196737, Average false similarity: -4.167495412844033\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 1: Accuracy 50.0%\n",
      "Average true similarity: -4.168817533129455, Average false similarity: -4.170624872579005\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 2: Accuracy 50.0%\n",
      "Average true similarity: -4.172084607543321, Average false similarity: -4.173128440366979\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 3: Accuracy 50.0%\n",
      "Average true similarity: -4.174560652395512, Average false similarity: -4.1763710499490285\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 4: Accuracy 50.0%\n",
      "Average true similarity: -4.1746116207951065, Average false similarity: -4.177295616717633\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 5: Accuracy 50.0%\n",
      "Average true similarity: -4.168554536187562, Average false similarity: -4.171725790010195\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 6: Accuracy 50.0%\n",
      "Average true similarity: -4.1675290519877635, Average false similarity: -4.170611620795109\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 7: Accuracy 50.0%\n",
      "Average true similarity: -4.161121304791026, Average false similarity: -4.163838939857286\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 8: Accuracy 50.0%\n",
      "Average true similarity: -4.1573017329255855, Average false similarity: -4.161489296636081\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 9: Accuracy 50.0%\n",
      "Average true similarity: -4.147639143730888, Average false similarity: -4.151387359836899\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 10: Accuracy 50.0%\n",
      "Average true similarity: -4.140969418960246, Average false similarity: -4.145339449541283\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 11: Accuracy 50.0%\n",
      "Average true similarity: -4.129718654434253, Average false similarity: -4.133299694189606\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 12: Accuracy 50.0%\n",
      "Average true similarity: -4.111939857288481, Average false similarity: -4.116286442405707\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 13: Accuracy 50.0%\n",
      "Average true similarity: -4.095735983690107, Average false similarity: -4.1009255861365945\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 14: Accuracy 50.0%\n",
      "Average true similarity: -4.086756371049949, Average false similarity: -4.090990825688075\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 15: Accuracy 50.0%\n",
      "Average true similarity: -4.070239551478082, Average false similarity: -4.074759429153927\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 16: Accuracy 50.0%\n",
      "Average true similarity: -4.049945973496428, Average false similarity: -4.0543119266055045\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 17: Accuracy 50.0%\n",
      "Average true similarity: -4.03159633027523, Average false similarity: -4.03737003058104\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 18: Accuracy 50.0%\n",
      "Average true similarity: -4.019510703363914, Average false similarity: -4.024977573904178\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 19: Accuracy 50.0%\n",
      "Average true similarity: -3.998524974515796, Average false similarity: -4.003004077471965\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 20: Accuracy 50.0%\n",
      "Average true similarity: -3.9724362895005143, Average false similarity: -3.9781467889908195\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 21: Accuracy 50.0%\n",
      "Average true similarity: -3.9522752293577956, Average false similarity: -3.9576901121304773\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 22: Accuracy 50.0%\n",
      "Average true similarity: -3.9323761467889935, Average false similarity: -3.9384138634046897\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 23: Accuracy 50.0%\n",
      "Average true similarity: -3.9102222222222194, Average false similarity: -3.916464831804275\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 24: Accuracy 50.0%\n",
      "Average true similarity: -3.88905708460754, Average false similarity: -3.8954954128440322\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 25: Accuracy 50.0%\n",
      "Average true similarity: -3.8651753312945925, Average false similarity: -3.870350662589192\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 26: Accuracy 50.0%\n",
      "Average true similarity: -3.836825688073394, Average false similarity: -3.8432262996941917\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 27: Accuracy 50.0%\n",
      "Average true similarity: -3.8148328236493443, Average false similarity: -3.821204892966357\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 28: Accuracy 50.0%\n",
      "Average true similarity: -3.7966931702344566, Average false similarity: -3.8042589194699254\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 29: Accuracy 50.0%\n",
      "Average true similarity: -3.775841997961267, Average false similarity: -3.7824189602446494\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 30: Accuracy 50.0%\n",
      "Average true similarity: -3.7657104994903214, Average false similarity: -3.773754332313968\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 31: Accuracy 50.0%\n",
      "Average true similarity: -3.744781855249744, Average false similarity: -3.751248725790015\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 32: Accuracy 50.0%\n",
      "Average true similarity: -3.713159021406731, Average false similarity: -3.7203547400611594\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 33: Accuracy 50.0%\n",
      "Average true similarity: -3.6871100917431208, Average false similarity: -3.6940305810397533\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 34: Accuracy 50.0%\n",
      "Average true similarity: -3.662961264016311, Average false similarity: -3.6697859327217137\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 35: Accuracy 50.0%\n",
      "Average true similarity: -3.6449602446483165, Average false similarity: -3.652034658511727\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 36: Accuracy 50.0%\n",
      "Average true similarity: -3.625624872578995, Average false similarity: -3.6327859327217142\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 37: Accuracy 50.0%\n",
      "Average true similarity: -3.598475025484202, Average false similarity: -3.605338430173293\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 38: Accuracy 50.0%\n",
      "Average true similarity: -3.5760937818552447, Average false similarity: -3.5833333333333353\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 39: Accuracy 50.0%\n",
      "Average true similarity: -3.5505575942915404, Average false similarity: -3.558729867482159\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 40: Accuracy 50.0%\n",
      "Average true similarity: -3.5320142711518803, Average false similarity: -3.540002038735984\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 41: Accuracy 50.0%\n",
      "Average true similarity: -3.5172650356778803, Average false similarity: -3.5256452599388433\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 42: Accuracy 50.0%\n",
      "Average true similarity: -3.4960754332314004, Average false similarity: -3.5039449541284404\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 43: Accuracy 50.0%\n",
      "Average true similarity: -3.4740193679918496, Average false similarity: -3.481695208970439\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 44: Accuracy 50.0%\n",
      "Average true similarity: -3.4549622833842997, Average false similarity: -3.4639429153924604\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 45: Accuracy 50.0%\n",
      "Average true similarity: -3.432677879714579, Average false similarity: -3.440571865443423\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 46: Accuracy 50.0%\n",
      "Average true similarity: -3.4174617737003055, Average false similarity: -3.4252986748216157\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 47: Accuracy 50.0%\n",
      "Average true similarity: -3.403115188583078, Average false similarity: -3.410593272171253\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 48: Accuracy 50.0%\n",
      "Average true similarity: -3.3835769622833842, Average false similarity: -3.392558613659527\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 49: Accuracy 50.0%\n",
      "Average true similarity: -3.366755351681955, Average false similarity: -3.374550458715596\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 50: Accuracy 50.0%\n",
      "Average true similarity: -3.3433761467889855, Average false similarity: -3.351426095820593\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 51: Accuracy 50.0%\n",
      "Average true similarity: -3.3231957186544294, Average false similarity: -3.33160244648318\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 52: Accuracy 50.0%\n",
      "Average true similarity: -3.3051080530071393, Average false similarity: -3.3138246687054003\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 53: Accuracy 50.0%\n",
      "Average true similarity: -3.2867003058104, Average false similarity: -3.2957033639143742\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 54: Accuracy 50.0%\n",
      "Average true similarity: -3.262447502548425, Average false similarity: -3.2710244648317994\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 55: Accuracy 50.0%\n",
      "Average true similarity: -3.23984607543323, Average false similarity: -3.2485300713557588\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 56: Accuracy 50.0%\n",
      "Average true similarity: -3.2184057084607547, Average false similarity: -3.2268307849133544\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 57: Accuracy 50.0%\n",
      "Average true similarity: -3.20452089704383, Average false similarity: -3.2142925586136575\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 58: Accuracy 50.0%\n",
      "Average true similarity: -3.1865983690112105, Average false similarity: -3.196271151885823\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 59: Accuracy 50.0%\n",
      "Average true similarity: -3.1684383282364936, Average false similarity: -3.177749235474\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 60: Accuracy 50.0%\n",
      "Average true similarity: -3.157145769622835, Average false similarity: -3.1671192660550465\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 61: Accuracy 50.0%\n",
      "Average true similarity: -3.1384159021406703, Average false similarity: -3.147033639143734\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 62: Accuracy 50.0%\n",
      "Average true similarity: -3.12814780835882, Average false similarity: -3.138522935779813\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 63: Accuracy 50.0%\n",
      "Average true similarity: -3.122074413863407, Average false similarity: -3.1325575942915407\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 64: Accuracy 50.0%\n",
      "Average true similarity: -3.1117828746177385, Average false similarity: -3.121437308868502\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 65: Accuracy 50.0%\n",
      "Average true similarity: -3.101643221202855, Average false similarity: -3.112018348623854\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 66: Accuracy 50.0%\n",
      "Average true similarity: -3.087389398572884, Average false similarity: -3.0969867482161075\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 67: Accuracy 50.0%\n",
      "Average true similarity: -3.0660356778797135, Average false similarity: -3.075699286442404\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 68: Accuracy 50.0%\n",
      "Average true similarity: -3.0486401630988778, Average false similarity: -3.0588593272171245\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 69: Accuracy 50.0%\n",
      "Average true similarity: -3.040143730886852, Average false similarity: -3.050287461773698\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 70: Accuracy 50.0%\n",
      "Average true similarity: -3.0231845056065243, Average false similarity: -3.0324311926605536\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 71: Accuracy 50.0%\n",
      "Average true similarity: -3.0142262996941884, Average false similarity: -3.0243751274209965\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 72: Accuracy 50.0%\n",
      "Average true similarity: -2.99814882772681, Average false similarity: -3.008180428134557\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 73: Accuracy 50.0%\n",
      "Average true similarity: -2.980623853211009, Average false similarity: -2.990403669724771\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 74: Accuracy 50.0%\n",
      "Average true similarity: -2.9690540265035703, Average false similarity: -2.9794373088685018\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 75: Accuracy 50.0%\n",
      "Average true similarity: -2.9589622833843006, Average false similarity: -2.9692599388379235\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 76: Accuracy 50.0%\n",
      "Average true similarity: -2.945092762487257, Average false similarity: -2.9546218144750274\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 77: Accuracy 50.0%\n",
      "Average true similarity: -2.9321845056065263, Average false similarity: -2.9430346585117215\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 78: Accuracy 50.0%\n",
      "Average true similarity: -2.921212028542305, Average false similarity: -2.932084607543322\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 79: Accuracy 50.0%\n",
      "Average true similarity: -2.9109602446483183, Average false similarity: -2.922478083588171\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 80: Accuracy 50.0%\n",
      "Average true similarity: -2.899382262996943, Average false similarity: -2.9094903160040744\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 81: Accuracy 50.0%\n",
      "Average true similarity: -2.888858307849137, Average false similarity: -2.899965341488277\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 82: Accuracy 50.0%\n",
      "Average true similarity: -2.8788114169215135, Average false similarity: -2.8899011213047916\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 83: Accuracy 50.0%\n",
      "Average true similarity: -2.870799184505611, Average false similarity: -2.8812660550458715\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 84: Accuracy 50.0%\n",
      "Average true similarity: -2.860385321100919, Average false similarity: -2.8714179408766585\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 85: Accuracy 50.0%\n",
      "Average true similarity: -2.851392456676857, Average false similarity: -2.8620509683995934\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 86: Accuracy 50.0%\n",
      "Average true similarity: -2.843583078491337, Average false similarity: -2.8546024464831827\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 87: Accuracy 50.0%\n",
      "Average true similarity: -2.8364475025484213, Average false similarity: -2.8484118246687045\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 88: Accuracy 50.0%\n",
      "Average true similarity: -2.824606523955145, Average false similarity: -2.8357288481141714\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 89: Accuracy 50.0%\n",
      "Average true similarity: -2.8103353720693205, Average false similarity: -2.8217023445463854\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 90: Accuracy 50.0%\n",
      "Average true similarity: -2.8006666666666673, Average false similarity: -2.8118053007135555\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 91: Accuracy 50.0%\n",
      "Average true similarity: -2.793678899082566, Average false similarity: -2.805309887869519\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 92: Accuracy 50.0%\n",
      "Average true similarity: -2.785284403669724, Average false similarity: -2.79767380224261\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 93: Accuracy 50.0%\n",
      "Average true similarity: -2.7780071355759395, Average false similarity: -2.789720693170236\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 94: Accuracy 50.0%\n",
      "Average true similarity: -2.7614097859327247, Average false similarity: -2.7731569826707423\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 95: Accuracy 50.0%\n",
      "Average true similarity: -2.748213047910295, Average false similarity: -2.7608685015290484\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 96: Accuracy 50.0%\n",
      "Average true similarity: -2.7397573904179406, Average false similarity: -2.751360856269113\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 97: Accuracy 50.0%\n",
      "Average true similarity: -2.7314260958205936, Average false similarity: -2.7436360856269095\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 98: Accuracy 50.0%\n",
      "Average true similarity: -2.7251620795107017, Average false similarity: -2.7369633027522915\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 99: Accuracy 50.0%\n",
      "Average true similarity: -2.7164322120285447, Average false similarity: -2.7284118246687052\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 100: Accuracy 50.0%\n",
      "Average true similarity: -2.7075983690112113, Average false similarity: -2.719354740061161\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 101: Accuracy 50.0%\n",
      "Average true similarity: -2.701285423037713, Average false similarity: -2.713861365953105\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 102: Accuracy 50.0%\n",
      "Average true similarity: -2.6973160040774737, Average false similarity: -2.708444444444443\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 103: Accuracy 50.0%\n",
      "Average true similarity: -2.6901926605504594, Average false similarity: -2.702969418960243\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 104: Accuracy 50.0%\n",
      "Average true similarity: -2.679321100917429, Average false similarity: -2.6910774719673842\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 105: Accuracy 50.0%\n",
      "Average true similarity: -2.674169215086644, Average false similarity: -2.68623241590214\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 106: Accuracy 50.0%\n",
      "Average true similarity: -2.6712324159021374, Average false similarity: -2.6831855249745136\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 107: Accuracy 50.0%\n",
      "Average true similarity: -2.667254841997964, Average false similarity: -2.67982262996942\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 108: Accuracy 50.0%\n",
      "Average true similarity: -2.659753312945967, Average false similarity: -2.6716493374108037\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 109: Accuracy 50.0%\n",
      "Average true similarity: -2.6528440366972506, Average false similarity: -2.665070336391439\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 110: Accuracy 50.0%\n",
      "Average true similarity: -2.646824668705399, Average false similarity: -2.659282364933741\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 111: Accuracy 50.0%\n",
      "Average true similarity: -2.6378399592252793, Average false similarity: -2.6495545361875616\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 112: Accuracy 50.0%\n",
      "Average true similarity: -2.629072375127423, Average false similarity: -2.6418705402650327\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 113: Accuracy 50.0%\n",
      "Average true similarity: -2.6232558613659527, Average false similarity: -2.6353231396534174\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 114: Accuracy 50.0%\n",
      "Average true similarity: -2.6228022426095823, Average false similarity: -2.634976554536189\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 115: Accuracy 50.0%\n",
      "Average true similarity: -2.615901121304792, Average false similarity: -2.6279898063200777\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 116: Accuracy 50.0%\n",
      "Average true similarity: -2.609027522935778, Average false similarity: -2.620491335372069\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 117: Accuracy 50.0%\n",
      "Average true similarity: -2.6033323139653435, Average false similarity: -2.6147145769622813\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 118: Accuracy 50.0%\n",
      "Average true similarity: -2.597860346585118, Average false similarity: -2.609598369011212\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 119: Accuracy 50.0%\n",
      "Average true similarity: -2.592703363914372, Average false similarity: -2.6044607543323117\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 120: Accuracy 50.0%\n",
      "Average true similarity: -2.5855698267074407, Average false similarity: -2.5973047910295617\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 121: Accuracy 50.0%\n",
      "Average true similarity: -2.5765463812436287, Average false similarity: -2.588898063200813\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 122: Accuracy 50.0%\n",
      "Average true similarity: -2.5660540265035676, Average false similarity: -2.5788797145769675\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 123: Accuracy 50.0%\n",
      "Average true similarity: -2.5597278287461758, Average false similarity: -2.572033639143732\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 124: Accuracy 50.0%\n",
      "Average true similarity: -2.553404689092763, Average false similarity: -2.5658909276248716\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 125: Accuracy 50.0%\n",
      "Average true similarity: -2.545536187563709, Average false similarity: -2.558168195718653\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 126: Accuracy 50.0%\n",
      "Average true similarity: -2.542768603465853, Average false similarity: -2.5546523955147844\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 127: Accuracy 50.0%\n",
      "Average true similarity: -2.542537206931701, Average false similarity: -2.554766564729862\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 128: Accuracy 50.0%\n",
      "Average true similarity: -2.53693374108053, Average false similarity: -2.550198776758411\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 129: Accuracy 50.0%\n",
      "Average true similarity: -2.531988786952089, Average false similarity: -2.5442670744138653\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 130: Accuracy 50.0%\n",
      "Average true similarity: -2.5288797145769655, Average false similarity: -2.5412150866462815\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 131: Accuracy 50.0%\n",
      "Average true similarity: -2.5232446483180455, Average false similarity: -2.535344546381247\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 132: Accuracy 50.0%\n",
      "Average true similarity: -2.519687054026504, Average false similarity: -2.5318389398572854\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 133: Accuracy 50.0%\n",
      "Average true similarity: -2.516848114169215, Average false similarity: -2.5286666666666657\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 134: Accuracy 50.0%\n",
      "Average true similarity: -2.5133944954128427, Average false similarity: -2.5256156982670763\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 135: Accuracy 50.0%\n",
      "Average true similarity: -2.5104209989806323, Average false similarity: -2.5232354740061163\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 136: Accuracy 50.0%\n",
      "Average true similarity: -2.506225280326194, Average false similarity: -2.5186167176350662\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 137: Accuracy 50.0%\n",
      "Average true similarity: -2.503752293577981, Average false similarity: -2.515771661569828\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 138: Accuracy 50.0%\n",
      "Average true similarity: -2.502231396534149, Average false similarity: -2.514806320081551\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 139: Accuracy 50.0%\n",
      "Average true similarity: -2.4986605504587116, Average false similarity: -2.511713557594295\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 140: Accuracy 50.0%\n",
      "Average true similarity: -2.494023445463809, Average false similarity: -2.5073649337410826\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 141: Accuracy 50.0%\n",
      "Average true similarity: -2.4901253822629954, Average false similarity: -2.5033374108053033\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 142: Accuracy 50.0%\n",
      "Average true similarity: -2.4814362895005107, Average false similarity: -2.4951518858307846\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 143: Accuracy 50.0%\n",
      "Average true similarity: -2.474567787971459, Average false similarity: -2.4880101936799184\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 144: Accuracy 50.0%\n",
      "Average true similarity: -2.4726004077471986, Average false similarity: -2.4854872579001013\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 145: Accuracy 50.0%\n",
      "Average true similarity: -2.4665626911314993, Average false similarity: -2.4790020387359832\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 146: Accuracy 50.0%\n",
      "Average true similarity: -2.461826707441387, Average false similarity: -2.4743211009174306\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 147: Accuracy 50.0%\n",
      "Average true similarity: -2.4589643221202846, Average false similarity: -2.472195718654434\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 148: Accuracy 50.0%\n",
      "Average true similarity: -2.457795107033634, Average false similarity: -2.4701131498470934\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 149: Accuracy 50.0%\n",
      "Average true similarity: -2.453150866462794, Average false similarity: -2.4660030581039756\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 150: Accuracy 50.0%\n",
      "Average true similarity: -2.4499806320081525, Average false similarity: -2.462754332313968\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 151: Accuracy 50.0%\n",
      "Average true similarity: -2.448768603465848, Average false similarity: -2.4615810397553513\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 152: Accuracy 50.0%\n",
      "Average true similarity: -2.4446921508664627, Average false similarity: -2.458532110091746\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 153: Accuracy 50.0%\n",
      "Average true similarity: -2.442556574923549, Average false similarity: -2.4557064220183515\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 154: Accuracy 50.0%\n",
      "Average true similarity: -2.4390061162079517, Average false similarity: -2.452423037716616\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 155: Accuracy 50.0%\n",
      "Average true similarity: -2.4349021406727815, Average false similarity: -2.4480428134556615\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 156: Accuracy 50.0%\n",
      "Average true similarity: -2.4293068297655447, Average false similarity: -2.442561671763505\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 157: Accuracy 50.0%\n",
      "Average true similarity: -2.424416921508668, Average false similarity: -2.437459734964324\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 158: Accuracy 50.0%\n",
      "Average true similarity: -2.420672782874616, Average false similarity: -2.4332691131498447\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 159: Accuracy 50.0%\n",
      "Average true similarity: -2.4175830784913344, Average false similarity: -2.4305637104994884\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 160: Accuracy 50.0%\n",
      "Average true similarity: -2.4134617737003072, Average false similarity: -2.426672782874617\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 161: Accuracy 50.0%\n",
      "Average true similarity: -2.41030275229358, Average false similarity: -2.4237033639143717\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 162: Accuracy 50.0%\n",
      "Average true similarity: -2.409001019367993, Average false similarity: -2.4216422018348607\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 163: Accuracy 50.0%\n",
      "Average true similarity: -2.408479102956165, Average false similarity: -2.4211885830784885\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 164: Accuracy 50.0%\n",
      "Average true similarity: -2.4042844036697244, Average false similarity: -2.4178562691131487\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 165: Accuracy 50.0%\n",
      "Average true similarity: -2.401836901121302, Average false similarity: -2.4149745158002025\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 166: Accuracy 50.0%\n",
      "Average true similarity: -2.3992691131498467, Average false similarity: -2.411847094801225\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 167: Accuracy 50.0%\n",
      "Average true similarity: -2.3966636085626893, Average false similarity: -2.4085963302752313\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 168: Accuracy 50.0%\n",
      "Average true similarity: -2.394713557594298, Average false similarity: -2.4082140672782884\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 169: Accuracy 50.0%\n",
      "Average true similarity: -2.390475025484199, Average false similarity: -2.403870540265036\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 170: Accuracy 50.0%\n",
      "Average true similarity: -2.3885779816513755, Average false similarity: -2.4015606523955144\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 171: Accuracy 50.0%\n",
      "Average true similarity: -2.3887329255861367, Average false similarity: -2.4016615698267065\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 172: Accuracy 50.0%\n",
      "Average true similarity: -2.385327217125385, Average false similarity: -2.39852599388379\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 173: Accuracy 50.0%\n",
      "Average true similarity: -2.383437308868501, Average false similarity: -2.3966065239551493\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 174: Accuracy 50.0%\n",
      "Average true similarity: -2.379801223241589, Average false similarity: -2.39282364933741\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 175: Accuracy 50.0%\n",
      "Average true similarity: -2.375012232415904, Average false similarity: -2.388229357798163\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 176: Accuracy 50.0%\n",
      "Average true similarity: -2.3700214067278296, Average false similarity: -2.3828644240570864\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 177: Accuracy 50.0%\n",
      "Average true similarity: -2.366016309887869, Average false similarity: -2.378627930682977\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 178: Accuracy 50.0%\n",
      "Average true similarity: -2.3641590214067247, Average false similarity: -2.377541284403671\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 179: Accuracy 50.0%\n",
      "Average true similarity: -2.3604668705402627, Average false similarity: -2.3737645259938818\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 180: Accuracy 50.0%\n",
      "Average true similarity: -2.3565912334352714, Average false similarity: -2.3700275229357812\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 181: Accuracy 50.0%\n",
      "Average true similarity: -2.355255861365951, Average false similarity: -2.367770642201835\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 182: Accuracy 50.0%\n",
      "Average true similarity: -2.3546544342507647, Average false similarity: -2.3681753312945957\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 183: Accuracy 50.0%\n",
      "Average true similarity: -2.3529704383282333, Average false similarity: -2.366439347604484\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 184: Accuracy 50.0%\n",
      "Average true similarity: -2.350618756371049, Average false similarity: -2.3642833843017352\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 185: Accuracy 50.0%\n",
      "Average true similarity: -2.3481743119266074, Average false similarity: -2.3612324159021405\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 186: Accuracy 50.0%\n",
      "Average true similarity: -2.3460632008154914, Average false similarity: -2.35911009174312\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 187: Accuracy 50.0%\n",
      "Average true similarity: -2.342195718654437, Average false similarity: -2.355908256880731\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 188: Accuracy 50.0%\n",
      "Average true similarity: -2.3405361875637087, Average false similarity: -2.353958205912336\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 189: Accuracy 50.0%\n",
      "Average true similarity: -2.3365840978593315, Average false similarity: -2.34962691131498\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 190: Accuracy 50.0%\n",
      "Average true similarity: -2.333101936799184, Average false similarity: -2.346139653414881\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 191: Accuracy 50.0%\n",
      "Average true similarity: -2.327768603465856, Average false similarity: -2.3404607543323146\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 192: Accuracy 50.0%\n",
      "Average true similarity: -2.324266055045872, Average false similarity: -2.33669622833843\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 193: Accuracy 50.0%\n",
      "Average true similarity: -2.322785932721713, Average false similarity: -2.33628236493374\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 194: Accuracy 50.0%\n",
      "Average true similarity: -2.319586136595309, Average false similarity: -2.33238124362895\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 195: Accuracy 50.0%\n",
      "Average true similarity: -2.3146625891946995, Average false similarity: -2.327851172273189\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 196: Accuracy 50.0%\n",
      "Average true similarity: -2.312748216106014, Average false similarity: -2.3263761467889936\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 197: Accuracy 50.0%\n",
      "Average true similarity: -2.3113761467889873, Average false similarity: -2.325233435270132\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 198: Accuracy 50.0%\n",
      "Average true similarity: -2.3097339449541283, Average false similarity: -2.3230224260958208\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 199: Accuracy 50.0%\n",
      "Average true similarity: -2.30917329255861, Average false similarity: -2.3229378185524956\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 200: Accuracy 50.0%\n",
      "Average true similarity: -2.3050428134556573, Average false similarity: -2.318900101936801\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 201: Accuracy 50.0%\n",
      "Average true similarity: -2.3033139653414887, Average false similarity: -2.3169938837920503\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 202: Accuracy 50.0%\n",
      "Average true similarity: -2.303508664627932, Average false similarity: -2.316998980632007\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 203: Accuracy 50.0%\n",
      "Average true similarity: -2.303252803261976, Average false similarity: -2.3161253822629964\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 204: Accuracy 50.0%\n",
      "Average true similarity: -2.303705402650355, Average false similarity: -2.317379204892968\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 205: Accuracy 50.0%\n",
      "Average true similarity: -2.301688073394496, Average false similarity: -2.315533129459735\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 206: Accuracy 50.0%\n",
      "Average true similarity: -2.301428134556577, Average false similarity: -2.315393476044855\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 207: Accuracy 50.0%\n",
      "Average true similarity: -2.3016707441386304, Average false similarity: -2.3153160040774714\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 208: Accuracy 50.0%\n",
      "Average true similarity: -2.302062181447505, Average false similarity: -2.31602242609582\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 209: Accuracy 50.0%\n",
      "Average true similarity: -2.302470948012232, Average false similarity: -2.3163312945973518\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 210: Accuracy 50.0%\n",
      "Average true similarity: -2.302958205912335, Average false similarity: -2.3164903160040775\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 211: Accuracy 50.0%\n",
      "Average true similarity: -2.3022813455657487, Average false similarity: -2.3165412844036704\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 212: Accuracy 50.0%\n",
      "Average true similarity: -2.301627930682975, Average false similarity: -2.314825688073394\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 213: Accuracy 50.0%\n",
      "Average true similarity: -2.2990265035677875, Average false similarity: -2.312826707441387\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 214: Accuracy 50.0%\n",
      "Average true similarity: -2.2957920489296644, Average false similarity: -2.3094678899082592\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 215: Accuracy 50.0%\n",
      "Average true similarity: -2.2941304791029538, Average false similarity: -2.3076034658511735\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 216: Accuracy 50.0%\n",
      "Average true similarity: -2.292174311926603, Average false similarity: -2.3061325178389414\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 217: Accuracy 50.0%\n",
      "Average true similarity: -2.290747196738026, Average false similarity: -2.303923547400607\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 218: Accuracy 50.0%\n",
      "Average true similarity: -2.290109072375127, Average false similarity: -2.3040479102956204\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 219: Accuracy 50.0%\n",
      "Average true similarity: -2.287300713557592, Average false similarity: -2.301531090723751\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 220: Accuracy 50.0%\n",
      "Average true similarity: -2.2847074413863426, Average false similarity: -2.2988868501529045\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 221: Accuracy 50.0%\n",
      "Average true similarity: -2.282565749235476, Average false similarity: -2.295615698267078\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 222: Accuracy 50.0%\n",
      "Average true similarity: -2.2817319062181447, Average false similarity: -2.2945728848114175\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 223: Accuracy 50.0%\n",
      "Average true similarity: -2.279632008154944, Average false similarity: -2.2929959225280276\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 224: Accuracy 50.0%\n",
      "Average true similarity: -2.2763873598368987, Average false similarity: -2.290337410805297\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 225: Accuracy 50.0%\n",
      "Average true similarity: -2.275756371049948, Average false similarity: -2.2899276248725764\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 226: Accuracy 50.0%\n",
      "Average true similarity: -2.274826707441386, Average false similarity: -2.288293577981654\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 227: Accuracy 50.0%\n",
      "Average true similarity: -2.271883792048934, Average false similarity: -2.2861814475025484\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 228: Accuracy 50.0%\n",
      "Average true similarity: -2.2690723751274215, Average false similarity: -2.282824668705405\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 229: Accuracy 50.0%\n",
      "Average true similarity: -2.2655096839959223, Average false similarity: -2.279930682976554\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 230: Accuracy 50.0%\n",
      "Average true similarity: -2.2648358817533096, Average false similarity: -2.278482161060139\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 231: Accuracy 50.0%\n",
      "Average true similarity: -2.2645096839959207, Average false similarity: -2.2785626911314973\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 232: Accuracy 50.0%\n",
      "Average true similarity: -2.2641845056065253, Average false similarity: -2.2771253822629958\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 233: Accuracy 50.0%\n",
      "Average true similarity: -2.2613160040774716, Average false similarity: -2.2755178389398574\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 234: Accuracy 50.0%\n",
      "Average true similarity: -2.260424057084607, Average false similarity: -2.2744036697247716\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 235: Accuracy 50.0%\n",
      "Average true similarity: -2.2603486238532104, Average false similarity: -2.274261977573903\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 236: Accuracy 50.0%\n",
      "Average true similarity: -2.258560652395513, Average false similarity: -2.272798165137617\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 237: Accuracy 50.0%\n",
      "Average true similarity: -2.2587838939857297, Average false similarity: -2.272970438328238\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 238: Accuracy 50.0%\n",
      "Average true similarity: -2.2557043832823647, Average false similarity: -2.2692466870540264\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 239: Accuracy 50.0%\n",
      "Average true similarity: -2.255317023445466, Average false similarity: -2.268995922528035\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 240: Accuracy 50.0%\n",
      "Average true similarity: -2.2549429153924554, Average false similarity: -2.2686207951070307\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 241: Accuracy 50.0%\n",
      "Average true similarity: -2.253574923547403, Average false similarity: -2.2672262996941877\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 242: Accuracy 50.0%\n",
      "Average true similarity: -2.2524658511722744, Average false similarity: -2.265929663608561\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 243: Accuracy 50.0%\n",
      "Average true similarity: -2.250531090723754, Average false similarity: -2.265474006116204\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 244: Accuracy 50.0%\n",
      "Average true similarity: -2.2476523955147796, Average false similarity: -2.2608583078491304\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 245: Accuracy 50.0%\n",
      "Average true similarity: -2.2453435270132522, Average false similarity: -2.259266055045872\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 246: Accuracy 50.0%\n",
      "Average true similarity: -2.244084607543323, Average false similarity: -2.257967380224263\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 247: Accuracy 50.0%\n",
      "Average true similarity: -2.2421518858307836, Average false similarity: -2.2561580020387355\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 248: Accuracy 50.0%\n",
      "Average true similarity: -2.2396228338430175, Average false similarity: -2.2539204892966382\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 249: Accuracy 50.0%\n",
      "Average true similarity: -2.2353465851172274, Average false similarity: -2.2494515800203883\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 250: Accuracy 50.0%\n",
      "Average true similarity: -2.2337441386340484, Average false similarity: -2.2469867482161066\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 251: Accuracy 50.0%\n",
      "Average true similarity: -2.2336697247706403, Average false similarity: -2.2467084607543315\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 252: Accuracy 50.0%\n",
      "Average true similarity: -2.232771661569828, Average false similarity: -2.247658511722736\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 253: Accuracy 50.0%\n",
      "Average true similarity: -2.2338134556574913, Average false similarity: -2.2478226299694195\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 254: Accuracy 50.0%\n",
      "Average true similarity: -2.2319102956167165, Average false similarity: -2.246167176350662\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 255: Accuracy 50.0%\n",
      "Average true similarity: -2.230862385321099, Average false similarity: -2.2449867482161046\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 256: Accuracy 50.0%\n",
      "Average true similarity: -2.2301569826707404, Average false similarity: -2.243683995922527\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 257: Accuracy 50.0%\n",
      "Average true similarity: -2.229755351681958, Average false similarity: -2.2437257900101946\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 258: Accuracy 50.0%\n",
      "Average true similarity: -2.228815494393472, Average false similarity: -2.242962283384303\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 259: Accuracy 50.0%\n",
      "Average true similarity: -2.228351681957185, Average false similarity: -2.2426809378185557\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 260: Accuracy 50.0%\n",
      "Average true similarity: -2.227227319062182, Average false similarity: -2.242165137614678\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 261: Accuracy 50.0%\n",
      "Average true similarity: -2.229302752293577, Average false similarity: -2.2429898063200815\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 262: Accuracy 50.0%\n",
      "Average true similarity: -2.227336391437312, Average false similarity: -2.2415310907237505\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 263: Accuracy 50.0%\n",
      "Average true similarity: -2.2251753312945985, Average false similarity: -2.239374108053007\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 264: Accuracy 50.0%\n",
      "Average true similarity: -2.2231019367991833, Average false similarity: -2.237563710499491\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 265: Accuracy 50.05%\n",
      "Average true similarity: -2.2213730886850147, Average false similarity: -2.2362089704383292\n",
      "True negative: 981, True positive: 1, False negative: 0, False positive: 980\n",
      "Epoch 266: Accuracy 50.0%\n",
      "Average true similarity: -2.2191039755351714, Average false similarity: -2.23369724770642\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 267: Accuracy 50.0%\n",
      "Average true similarity: -2.21979510703364, Average false similarity: -2.2338786952089698\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 268: Accuracy 50.0%\n",
      "Average true similarity: -2.2209918450560644, Average false similarity: -2.2343353720693178\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 269: Accuracy 50.0%\n",
      "Average true similarity: -2.2206799184505592, Average false similarity: -2.2344322120285423\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 270: Accuracy 50.0%\n",
      "Average true similarity: -2.2192008154943963, Average false similarity: -2.2333017329255833\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 271: Accuracy 50.0%\n",
      "Average true similarity: -2.2189296636085607, Average false similarity: -2.232980632008155\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 272: Accuracy 50.0%\n",
      "Average true similarity: -2.2182558613659555, Average false similarity: -2.232494393476044\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 273: Accuracy 50.0%\n",
      "Average true similarity: -2.2184821610601393, Average false similarity: -2.2322191641182454\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 274: Accuracy 50.0%\n",
      "Average true similarity: -2.217759429153924, Average false similarity: -2.2312436289500517\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 275: Accuracy 50.0%\n",
      "Average true similarity: -2.217067278287462, Average false similarity: -2.230706422018348\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 276: Accuracy 50.0%\n",
      "Average true similarity: -2.2175688073394513, Average false similarity: -2.231295616717636\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 277: Accuracy 50.0%\n",
      "Average true similarity: -2.217605504587153, Average false similarity: -2.230720693170232\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 278: Accuracy 50.0%\n",
      "Average true similarity: -2.2162181447502567, Average false similarity: -2.230253822629967\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 279: Accuracy 50.0%\n",
      "Average true similarity: -2.2144984709480138, Average false similarity: -2.2297451580020384\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 280: Accuracy 50.0%\n",
      "Average true similarity: -2.214705402650358, Average false similarity: -2.2288644240570834\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 281: Accuracy 50.0%\n",
      "Average true similarity: -2.214194699286441, Average false similarity: -2.2282293577981624\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 282: Accuracy 50.0%\n",
      "Average true similarity: -2.214875637104995, Average false similarity: -2.2289215086646257\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 283: Accuracy 50.0%\n",
      "Average true similarity: -2.2145657492354753, Average false similarity: -2.2290387359836887\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 284: Accuracy 50.0%\n",
      "Average true similarity: -2.2135280326197777, Average false similarity: -2.228381243628949\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 285: Accuracy 50.05%\n",
      "Average true similarity: -2.2111049949031623, Average false similarity: -2.224891946992863\n",
      "True negative: 981, True positive: 1, False negative: 0, False positive: 980\n",
      "Epoch 286: Accuracy 50.05%\n",
      "Average true similarity: -2.210274209989807, Average false similarity: -2.224758409785936\n",
      "True negative: 981, True positive: 1, False negative: 0, False positive: 980\n",
      "Epoch 287: Accuracy 50.1%\n",
      "Average true similarity: -2.2091936799184513, Average false similarity: -2.223300713557595\n",
      "True negative: 981, True positive: 2, False negative: 0, False positive: 979\n",
      "Epoch 288: Accuracy 50.1%\n",
      "Average true similarity: -2.206680937818551, Average false similarity: -2.2207726809378174\n",
      "True negative: 981, True positive: 2, False negative: 0, False positive: 979\n",
      "Epoch 289: Accuracy 50.05%\n",
      "Average true similarity: -2.2062680937818544, Average false similarity: -2.220716615698268\n",
      "True negative: 981, True positive: 1, False negative: 0, False positive: 980\n",
      "Epoch 290: Accuracy 50.1%\n",
      "Average true similarity: -2.2056116207951084, Average false similarity: -2.2196432212028556\n",
      "True negative: 981, True positive: 2, False negative: 0, False positive: 979\n",
      "Epoch 291: Accuracy 50.05%\n",
      "Average true similarity: -2.2044271151885857, Average false similarity: -2.2192813455657516\n",
      "True negative: 981, True positive: 1, False negative: 0, False positive: 980\n",
      "Epoch 292: Accuracy 50.05%\n",
      "Average true similarity: -2.2049816513761487, Average false similarity: -2.219488277268095\n",
      "True negative: 981, True positive: 1, False negative: 0, False positive: 980\n",
      "Epoch 293: Accuracy 50.2%\n",
      "Average true similarity: -2.2050254841998, Average false similarity: -2.219424057084611\n",
      "True negative: 981, True positive: 4, False negative: 0, False positive: 977\n",
      "Epoch 294: Accuracy 50.15%\n",
      "Average true similarity: -2.204567787971457, Average false similarity: -2.2186992864424075\n",
      "True negative: 981, True positive: 3, False negative: 0, False positive: 978\n",
      "Epoch 295: Accuracy 50.05%\n",
      "Average true similarity: -2.2049021406727807, Average false similarity: -2.218821610601424\n",
      "True negative: 981, True positive: 1, False negative: 0, False positive: 980\n",
      "Epoch 296: Accuracy 50.0%\n",
      "Average true similarity: -2.2048603465851158, Average false similarity: -2.2191865443425067\n",
      "True negative: 981, True positive: 0, False negative: 0, False positive: 981\n",
      "Epoch 297: Accuracy 50.05%\n",
      "Average true similarity: -2.2052395514780847, Average false similarity: -2.219476044852194\n",
      "True negative: 980, True positive: 2, False negative: 1, False positive: 979\n",
      "Epoch 298: Accuracy 50.2%\n",
      "Average true similarity: -2.206160040774721, Average false similarity: -2.219909276248725\n",
      "True negative: 981, True positive: 4, False negative: 0, False positive: 977\n",
      "Epoch 299: Accuracy 50.15%\n",
      "Average true similarity: -2.2048868501529064, Average false similarity: -2.219200815494393\n",
      "True negative: 981, True positive: 3, False negative: 0, False positive: 978\n",
      "Epoch 300: Accuracy 50.1%\n",
      "Average true similarity: -2.2047604485219168, Average false similarity: -2.2193944954128457\n",
      "True negative: 980, True positive: 3, False negative: 1, False positive: 978\n",
      "Epoch 301: Accuracy 50.1%\n",
      "Average true similarity: -2.2036024464831776, Average false similarity: -2.218258919469927\n",
      "True negative: 981, True positive: 2, False negative: 0, False positive: 979\n",
      "Epoch 302: Accuracy 50.36%\n",
      "Average true similarity: -2.202007135575942, Average false similarity: -2.215986748216106\n",
      "True negative: 981, True positive: 7, False negative: 0, False positive: 974\n",
      "Epoch 303: Accuracy 50.15%\n",
      "Average true similarity: -2.201707441386338, Average false similarity: -2.2160479102956154\n",
      "True negative: 981, True positive: 3, False negative: 0, False positive: 978\n",
      "Epoch 304: Accuracy 50.15%\n",
      "Average true similarity: -2.201946992864426, Average false similarity: -2.2156095820591215\n",
      "True negative: 981, True positive: 3, False negative: 0, False positive: 978\n",
      "Epoch 305: Accuracy 50.2%\n",
      "Average true similarity: -2.2016432212028536, Average false similarity: -2.216322120285423\n",
      "True negative: 981, True positive: 4, False negative: 0, False positive: 977\n",
      "Epoch 306: Accuracy 50.15%\n",
      "Average true similarity: -2.2017186544342513, Average false similarity: -2.2163710499490326\n",
      "True negative: 981, True positive: 3, False negative: 0, False positive: 978\n",
      "Epoch 307: Accuracy 50.15%\n",
      "Average true similarity: -2.201395514780838, Average false similarity: -2.215752293577983\n",
      "True negative: 980, True positive: 4, False negative: 1, False positive: 977\n",
      "Epoch 308: Accuracy 50.2%\n",
      "Average true similarity: -2.200558613659534, Average false similarity: -2.2157675840978617\n",
      "True negative: 981, True positive: 4, False negative: 0, False positive: 977\n",
      "Epoch 309: Accuracy 50.15%\n",
      "Average true similarity: -2.201750254841997, Average false similarity: -2.2169887869520903\n",
      "True negative: 981, True positive: 3, False negative: 0, False positive: 978\n",
      "Epoch 310: Accuracy 50.15%\n",
      "Average true similarity: -2.200543323139648, Average false similarity: -2.214695208970441\n",
      "True negative: 981, True positive: 3, False negative: 0, False positive: 978\n",
      "Epoch 311: Accuracy 50.2%\n",
      "Average true similarity: -2.197903160040773, Average false similarity: -2.2130764525993927\n",
      "True negative: 981, True positive: 4, False negative: 0, False positive: 977\n",
      "Epoch 312: Accuracy 50.25%\n",
      "Average true similarity: -2.196050968399592, Average false similarity: -2.210452599388378\n",
      "True negative: 980, True positive: 6, False negative: 1, False positive: 975\n",
      "Epoch 313: Accuracy 50.36%\n",
      "Average true similarity: -2.1949561671763527, Average false similarity: -2.2093965341488255\n",
      "True negative: 980, True positive: 8, False negative: 1, False positive: 973\n",
      "Epoch 314: Accuracy 50.36%\n",
      "Average true similarity: -2.193328236493373, Average false similarity: -2.2078613659531126\n",
      "True negative: 980, True positive: 8, False negative: 1, False positive: 973\n",
      "Epoch 315: Accuracy 50.46%\n",
      "Average true similarity: -2.192625891946995, Average false similarity: -2.2071447502548427\n",
      "True negative: 980, True positive: 10, False negative: 1, False positive: 971\n",
      "Epoch 316: Accuracy 50.36%\n",
      "Average true similarity: -2.191968399592249, Average false similarity: -2.2064322120285436\n",
      "True negative: 980, True positive: 8, False negative: 1, False positive: 973\n",
      "Epoch 317: Accuracy 50.36%\n",
      "Average true similarity: -2.1905677879714593, Average false similarity: -2.2050968399592272\n",
      "True negative: 980, True positive: 8, False negative: 1, False positive: 973\n",
      "Epoch 318: Accuracy 50.36%\n",
      "Average true similarity: -2.188433231396535, Average false similarity: -2.202348623853211\n",
      "True negative: 980, True positive: 8, False negative: 1, False positive: 973\n",
      "Epoch 319: Accuracy 50.61%\n",
      "Average true similarity: -2.1875463812436324, Average false similarity: -2.2022813455657504\n",
      "True negative: 980, True positive: 13, False negative: 1, False positive: 968\n",
      "Epoch 320: Accuracy 50.51%\n",
      "Average true similarity: -2.1881049949031586, Average false similarity: -2.2023465851172275\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 321: Accuracy 50.56%\n",
      "Average true similarity: -2.1878970438328267, Average false similarity: -2.202533129459734\n",
      "True negative: 980, True positive: 12, False negative: 1, False positive: 969\n",
      "Epoch 322: Accuracy 50.51%\n",
      "Average true similarity: -2.188319062181447, Average false similarity: -2.2036279306829782\n",
      "True negative: 980, True positive: 11, False negative: 1, False positive: 970\n",
      "Epoch 323: Accuracy 50.51%\n",
      "Average true similarity: -2.1876106014271164, Average false similarity: -2.2014281345565747\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 324: Accuracy 50.36%\n",
      "Average true similarity: -2.186817533129459, Average false similarity: -2.200838939857288\n",
      "True negative: 979, True positive: 9, False negative: 2, False positive: 972\n",
      "Epoch 325: Accuracy 50.56%\n",
      "Average true similarity: -2.186549439347603, Average false similarity: -2.2017064220183493\n",
      "True negative: 979, True positive: 13, False negative: 2, False positive: 968\n",
      "Epoch 326: Accuracy 50.31%\n",
      "Average true similarity: -2.1861875637104986, Average false similarity: -2.2001151885830765\n",
      "True negative: 980, True positive: 7, False negative: 1, False positive: 974\n",
      "Epoch 327: Accuracy 50.41%\n",
      "Average true similarity: -2.1861325178389412, Average false similarity: -2.20095514780836\n",
      "True negative: 979, True positive: 10, False negative: 2, False positive: 971\n",
      "Epoch 328: Accuracy 50.61%\n",
      "Average true similarity: -2.183445463812435, Average false similarity: -2.198768603465852\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 329: Accuracy 50.66%\n",
      "Average true similarity: -2.1828807339449527, Average false similarity: -2.1980377166157012\n",
      "True negative: 981, True positive: 13, False negative: 0, False positive: 968\n",
      "Epoch 330: Accuracy 50.61%\n",
      "Average true similarity: -2.182840978593271, Average false similarity: -2.196959225280328\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 331: Accuracy 50.71%\n",
      "Average true similarity: -2.1830693170234463, Average false similarity: -2.197402650356777\n",
      "True negative: 978, True positive: 17, False negative: 3, False positive: 964\n",
      "Epoch 332: Accuracy 50.82%\n",
      "Average true similarity: -2.18221712538226, Average false similarity: -2.196033639143732\n",
      "True negative: 980, True positive: 17, False negative: 1, False positive: 964\n",
      "Epoch 333: Accuracy 50.66%\n",
      "Average true similarity: -2.1822059123343576, Average false similarity: -2.1965718654434236\n",
      "True negative: 976, True positive: 18, False negative: 5, False positive: 963\n",
      "Epoch 334: Accuracy 50.71%\n",
      "Average true similarity: -2.1825025484199787, Average false similarity: -2.1972864424057104\n",
      "True negative: 979, True positive: 16, False negative: 2, False positive: 965\n",
      "Epoch 335: Accuracy 50.76%\n",
      "Average true similarity: -2.1830020387359843, Average false similarity: -2.197426095820591\n",
      "True negative: 980, True positive: 16, False negative: 1, False positive: 965\n",
      "Epoch 336: Accuracy 50.71%\n",
      "Average true similarity: -2.182124362895002, Average false similarity: -2.196531090723753\n",
      "True negative: 979, True positive: 16, False negative: 2, False positive: 965\n",
      "Epoch 337: Accuracy 50.61%\n",
      "Average true similarity: -2.181551478083586, Average false similarity: -2.1960713557594267\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 338: Accuracy 50.71%\n",
      "Average true similarity: -2.1808756371049913, Average false similarity: -2.1954658511722744\n",
      "True negative: 978, True positive: 17, False negative: 3, False positive: 964\n",
      "Epoch 339: Accuracy 50.66%\n",
      "Average true similarity: -2.1802884811416896, Average false similarity: -2.19446483180428\n",
      "True negative: 978, True positive: 16, False negative: 3, False positive: 965\n",
      "Epoch 340: Accuracy 50.71%\n",
      "Average true similarity: -2.180941896024464, Average false similarity: -2.195745158002038\n",
      "True negative: 980, True positive: 15, False negative: 1, False positive: 966\n",
      "Epoch 341: Accuracy 50.56%\n",
      "Average true similarity: -2.182237512742102, Average false similarity: -2.196270132517838\n",
      "True negative: 979, True positive: 13, False negative: 2, False positive: 968\n",
      "Epoch 342: Accuracy 50.66%\n",
      "Average true similarity: -2.184672782874616, Average false similarity: -2.1991325178389425\n",
      "True negative: 980, True positive: 14, False negative: 1, False positive: 967\n",
      "Epoch 343: Accuracy 50.51%\n",
      "Average true similarity: -2.184538226299694, Average false similarity: -2.199200815494394\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 344: Accuracy 50.36%\n",
      "Average true similarity: -2.185416921508667, Average false similarity: -2.200356778797147\n",
      "True negative: 980, True positive: 8, False negative: 1, False positive: 973\n",
      "Epoch 345: Accuracy 50.41%\n",
      "Average true similarity: -2.1841620795107026, Average false similarity: -2.198528032619775\n",
      "True negative: 978, True positive: 11, False negative: 3, False positive: 970\n",
      "Epoch 346: Accuracy 50.51%\n",
      "Average true similarity: -2.1817003058103976, Average false similarity: -2.1963363914373093\n",
      "True negative: 978, True positive: 13, False negative: 3, False positive: 968\n",
      "Epoch 347: Accuracy 50.76%\n",
      "Average true similarity: -2.181510703363914, Average false similarity: -2.196311926605506\n",
      "True negative: 980, True positive: 16, False negative: 1, False positive: 965\n",
      "Epoch 348: Accuracy 50.61%\n",
      "Average true similarity: -2.180100917431192, Average false similarity: -2.1947492354740037\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 349: Accuracy 50.61%\n",
      "Average true similarity: -2.1794352701325197, Average false similarity: -2.1940040774719667\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 350: Accuracy 50.71%\n",
      "Average true similarity: -2.1790703363914394, Average false similarity: -2.1936676860346584\n",
      "True negative: 978, True positive: 17, False negative: 3, False positive: 964\n",
      "Epoch 351: Accuracy 50.66%\n",
      "Average true similarity: -2.1784994903160046, Average false similarity: -2.1928501529051965\n",
      "True negative: 978, True positive: 16, False negative: 3, False positive: 965\n",
      "Epoch 352: Accuracy 50.71%\n",
      "Average true similarity: -2.179720693170236, Average false similarity: -2.1935698267074417\n",
      "True negative: 978, True positive: 17, False negative: 3, False positive: 964\n",
      "Epoch 353: Accuracy 50.56%\n",
      "Average true similarity: -2.180585117227317, Average false similarity: -2.1955504587155956\n",
      "True negative: 980, True positive: 12, False negative: 1, False positive: 969\n",
      "Epoch 354: Accuracy 50.46%\n",
      "Average true similarity: -2.1820998980632007, Average false similarity: -2.1968195718654444\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 355: Accuracy 50.46%\n",
      "Average true similarity: -2.1830601427115193, Average false similarity: -2.197254841997961\n",
      "True negative: 981, True positive: 9, False negative: 0, False positive: 972\n",
      "Epoch 356: Accuracy 50.51%\n",
      "Average true similarity: -2.183378185524973, Average false similarity: -2.19719367991845\n",
      "True negative: 980, True positive: 11, False negative: 1, False positive: 970\n",
      "Epoch 357: Accuracy 50.36%\n",
      "Average true similarity: -2.183881753312942, Average false similarity: -2.1987828746177374\n",
      "True negative: 980, True positive: 8, False negative: 1, False positive: 973\n",
      "Epoch 358: Accuracy 50.51%\n",
      "Average true similarity: -2.184471967380219, Average false similarity: -2.1992079510703393\n",
      "True negative: 980, True positive: 11, False negative: 1, False positive: 970\n",
      "Epoch 359: Accuracy 50.41%\n",
      "Average true similarity: -2.184822629969416, Average false similarity: -2.199591233435269\n",
      "True negative: 980, True positive: 9, False negative: 1, False positive: 972\n",
      "Epoch 360: Accuracy 50.41%\n",
      "Average true similarity: -2.184410805300712, Average false similarity: -2.1985739041794097\n",
      "True negative: 980, True positive: 9, False negative: 1, False positive: 972\n",
      "Epoch 361: Accuracy 50.56%\n",
      "Average true similarity: -2.183828746177368, Average false similarity: -2.1988837920489295\n",
      "True negative: 981, True positive: 11, False negative: 0, False positive: 970\n",
      "Epoch 362: Accuracy 50.51%\n",
      "Average true similarity: -2.184668705402652, Average false similarity: -2.199310907237515\n",
      "True negative: 980, True positive: 11, False negative: 1, False positive: 970\n",
      "Epoch 363: Accuracy 50.61%\n",
      "Average true similarity: -2.1841651376146793, Average false similarity: -2.199548419979614\n",
      "True negative: 980, True positive: 13, False negative: 1, False positive: 968\n",
      "Epoch 364: Accuracy 50.46%\n",
      "Average true similarity: -2.1841019367991823, Average false similarity: -2.1986605504587162\n",
      "True negative: 980, True positive: 10, False negative: 1, False positive: 971\n",
      "Epoch 365: Accuracy 50.25%\n",
      "Average true similarity: -2.1850326197757384, Average false similarity: -2.199179408766564\n",
      "True negative: 979, True positive: 7, False negative: 2, False positive: 974\n",
      "Epoch 366: Accuracy 50.31%\n",
      "Average true similarity: -2.1839653414882765, Average false similarity: -2.198242609582056\n",
      "True negative: 979, True positive: 8, False negative: 2, False positive: 973\n",
      "Epoch 367: Accuracy 50.2%\n",
      "Average true similarity: -2.184869520897044, Average false similarity: -2.199195718654434\n",
      "True negative: 979, True positive: 6, False negative: 2, False positive: 975\n",
      "Epoch 368: Accuracy 50.36%\n",
      "Average true similarity: -2.1841325178389384, Average false similarity: -2.1992426095820607\n",
      "True negative: 981, True positive: 7, False negative: 0, False positive: 974\n",
      "Epoch 369: Accuracy 50.41%\n",
      "Average true similarity: -2.1839337410805273, Average false similarity: -2.1984933741080552\n",
      "True negative: 979, True positive: 10, False negative: 2, False positive: 971\n",
      "Epoch 370: Accuracy 50.41%\n",
      "Average true similarity: -2.182571865443423, Average false similarity: -2.1969164118246676\n",
      "True negative: 979, True positive: 10, False negative: 2, False positive: 971\n",
      "Epoch 371: Accuracy 50.61%\n",
      "Average true similarity: -2.1830305810397537, Average false similarity: -2.1972833843017363\n",
      "True negative: 980, True positive: 13, False negative: 1, False positive: 968\n",
      "Epoch 372: Accuracy 50.41%\n",
      "Average true similarity: -2.1834189602446474, Average false similarity: -2.197776758409787\n",
      "True negative: 980, True positive: 9, False negative: 1, False positive: 972\n",
      "Epoch 373: Accuracy 50.41%\n",
      "Average true similarity: -2.1827553516819576, Average false similarity: -2.197259938837921\n",
      "True negative: 978, True positive: 11, False negative: 3, False positive: 970\n",
      "Epoch 374: Accuracy 50.41%\n",
      "Average true similarity: -2.183280326197755, Average false similarity: -2.198437308868503\n",
      "True negative: 979, True positive: 10, False negative: 2, False positive: 971\n",
      "Epoch 375: Accuracy 50.46%\n",
      "Average true similarity: -2.1829561671763487, Average false similarity: -2.1977482161060133\n",
      "True negative: 980, True positive: 10, False negative: 1, False positive: 971\n",
      "Epoch 376: Accuracy 50.46%\n",
      "Average true similarity: -2.184478083588177, Average false similarity: -2.198898063200817\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 377: Accuracy 50.31%\n",
      "Average true similarity: -2.1831977573904187, Average false similarity: -2.197328236493378\n",
      "True negative: 980, True positive: 7, False negative: 1, False positive: 974\n",
      "Epoch 378: Accuracy 50.61%\n",
      "Average true similarity: -2.181804281345565, Average false similarity: -2.196804281345566\n",
      "True negative: 980, True positive: 13, False negative: 1, False positive: 968\n",
      "Epoch 379: Accuracy 50.46%\n",
      "Average true similarity: -2.1809041794087665, Average false similarity: -2.195800203873598\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 380: Accuracy 50.51%\n",
      "Average true similarity: -2.1809785932721697, Average false similarity: -2.1959500509684013\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 381: Accuracy 50.56%\n",
      "Average true similarity: -2.181989806320078, Average false similarity: -2.196300713557594\n",
      "True negative: 979, True positive: 13, False negative: 2, False positive: 968\n",
      "Epoch 382: Accuracy 50.36%\n",
      "Average true similarity: -2.1808205912334375, Average false similarity: -2.195115188583081\n",
      "True negative: 980, True positive: 8, False negative: 1, False positive: 973\n",
      "Epoch 383: Accuracy 50.41%\n",
      "Average true similarity: -2.1800784913353732, Average false similarity: -2.194946992864421\n",
      "True negative: 978, True positive: 11, False negative: 3, False positive: 970\n",
      "Epoch 384: Accuracy 50.51%\n",
      "Average true similarity: -2.178276248725789, Average false similarity: -2.192924566768602\n",
      "True negative: 978, True positive: 13, False negative: 3, False positive: 968\n",
      "Epoch 385: Accuracy 50.46%\n",
      "Average true similarity: -2.1783017329255876, Average false similarity: -2.192881753312946\n",
      "True negative: 978, True positive: 12, False negative: 3, False positive: 969\n",
      "Epoch 386: Accuracy 50.66%\n",
      "Average true similarity: -2.178444444444446, Average false similarity: -2.192668705402651\n",
      "True negative: 977, True positive: 17, False negative: 4, False positive: 964\n",
      "Epoch 387: Accuracy 50.51%\n",
      "Average true similarity: -2.177698267074416, Average false similarity: -2.1928562691131503\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 388: Accuracy 50.31%\n",
      "Average true similarity: -2.1783893985728815, Average false similarity: -2.1931957186544357\n",
      "True negative: 978, True positive: 9, False negative: 3, False positive: 972\n",
      "Epoch 389: Accuracy 50.46%\n",
      "Average true similarity: -2.1785168195718665, Average false similarity: -2.1929408766564715\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 390: Accuracy 50.71%\n",
      "Average true similarity: -2.1780519877675864, Average false similarity: -2.1928338430173295\n",
      "True negative: 980, True positive: 15, False negative: 1, False positive: 966\n",
      "Epoch 391: Accuracy 50.56%\n",
      "Average true similarity: -2.178304791029564, Average false similarity: -2.193145769622836\n",
      "True negative: 979, True positive: 13, False negative: 2, False positive: 968\n",
      "Epoch 392: Accuracy 50.56%\n",
      "Average true similarity: -2.178847094801226, Average false similarity: -2.193045871559637\n",
      "True negative: 979, True positive: 13, False negative: 2, False positive: 968\n",
      "Epoch 393: Accuracy 50.66%\n",
      "Average true similarity: -2.179112130479104, Average false similarity: -2.1934729867482146\n",
      "True negative: 981, True positive: 13, False negative: 0, False positive: 968\n",
      "Epoch 394: Accuracy 50.46%\n",
      "Average true similarity: -2.177872579001018, Average false similarity: -2.1934128440367\n",
      "True negative: 976, True positive: 14, False negative: 5, False positive: 967\n",
      "Epoch 395: Accuracy 50.41%\n",
      "Average true similarity: -2.1779113149847094, Average false similarity: -2.1921834862385317\n",
      "True negative: 977, True positive: 12, False negative: 4, False positive: 969\n",
      "Epoch 396: Accuracy 50.61%\n",
      "Average true similarity: -2.1771590214067293, Average false similarity: -2.191851172273192\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 397: Accuracy 50.66%\n",
      "Average true similarity: -2.1771304791029547, Average false similarity: -2.191958205912333\n",
      "True negative: 980, True positive: 14, False negative: 1, False positive: 967\n",
      "Epoch 398: Accuracy 50.61%\n",
      "Average true similarity: -2.1761304791029534, Average false similarity: -2.1907400611620815\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 399: Accuracy 50.71%\n",
      "Average true similarity: -2.175634046890927, Average false similarity: -2.1904729867482144\n",
      "True negative: 978, True positive: 17, False negative: 3, False positive: 964\n",
      "Epoch 400: Accuracy 50.51%\n",
      "Average true similarity: -2.1767288481141716, Average false similarity: -2.191997961264014\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 401: Accuracy 50.56%\n",
      "Average true similarity: -2.1765565749235436, Average false similarity: -2.192285423037715\n",
      "True negative: 979, True positive: 13, False negative: 2, False positive: 968\n",
      "Epoch 402: Accuracy 50.61%\n",
      "Average true similarity: -2.178134556574922, Average false similarity: -2.1925657492354715\n",
      "True negative: 980, True positive: 13, False negative: 1, False positive: 968\n",
      "Epoch 403: Accuracy 50.71%\n",
      "Average true similarity: -2.177125382262995, Average false similarity: -2.191812436289502\n",
      "True negative: 979, True positive: 16, False negative: 2, False positive: 965\n",
      "Epoch 404: Accuracy 50.61%\n",
      "Average true similarity: -2.1760356778797143, Average false similarity: -2.191039755351684\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 405: Accuracy 50.61%\n",
      "Average true similarity: -2.176211009174312, Average false similarity: -2.190698267074416\n",
      "True negative: 980, True positive: 13, False negative: 1, False positive: 968\n",
      "Epoch 406: Accuracy 50.56%\n",
      "Average true similarity: -2.1750621814475033, Average false similarity: -2.1902222222222174\n",
      "True negative: 978, True positive: 14, False negative: 3, False positive: 967\n",
      "Epoch 407: Accuracy 50.82%\n",
      "Average true similarity: -2.174462793068298, Average false similarity: -2.1889816513761438\n",
      "True negative: 979, True positive: 18, False negative: 2, False positive: 963\n",
      "Epoch 408: Accuracy 50.51%\n",
      "Average true similarity: -2.174009174311924, Average false similarity: -2.188046890927625\n",
      "True negative: 976, True positive: 15, False negative: 5, False positive: 966\n",
      "Epoch 409: Accuracy 50.46%\n",
      "Average true similarity: -2.1739184505606515, Average false similarity: -2.1890397553516845\n",
      "True negative: 974, True positive: 16, False negative: 7, False positive: 965\n",
      "Epoch 410: Accuracy 50.76%\n",
      "Average true similarity: -2.173652395514783, Average false similarity: -2.1885983690112134\n",
      "True negative: 978, True positive: 18, False negative: 3, False positive: 963\n",
      "Epoch 411: Accuracy 50.87%\n",
      "Average true similarity: -2.171708460754332, Average false similarity: -2.186674821610603\n",
      "True negative: 978, True positive: 20, False negative: 3, False positive: 961\n",
      "Epoch 412: Accuracy 50.61%\n",
      "Average true similarity: -2.1729653414882786, Average false similarity: -2.1879510703363914\n",
      "True negative: 976, True positive: 17, False negative: 5, False positive: 964\n",
      "Epoch 413: Accuracy 50.92%\n",
      "Average true similarity: -2.1727329255861374, Average false similarity: -2.187540265035677\n",
      "True negative: 976, True positive: 23, False negative: 5, False positive: 958\n",
      "Epoch 414: Accuracy 50.92%\n",
      "Average true similarity: -2.172025484199794, Average false similarity: -2.186847094801225\n",
      "True negative: 977, True positive: 22, False negative: 4, False positive: 959\n",
      "Epoch 415: Accuracy 51.02%\n",
      "Average true similarity: -2.171385321100918, Average false similarity: -2.1867716615698254\n",
      "True negative: 978, True positive: 23, False negative: 3, False positive: 958\n",
      "Epoch 416: Accuracy 50.66%\n",
      "Average true similarity: -2.1725810397553547, Average false similarity: -2.187592252803261\n",
      "True negative: 978, True positive: 16, False negative: 3, False positive: 965\n",
      "Epoch 417: Accuracy 50.82%\n",
      "Average true similarity: -2.172713557594293, Average false similarity: -2.187223241590211\n",
      "True negative: 978, True positive: 19, False negative: 3, False positive: 962\n",
      "Epoch 418: Accuracy 50.76%\n",
      "Average true similarity: -2.1728705402650377, Average false similarity: -2.1876636085626893\n",
      "True negative: 977, True positive: 19, False negative: 4, False positive: 962\n",
      "Epoch 419: Accuracy 50.71%\n",
      "Average true similarity: -2.1739225280326195, Average false similarity: -2.189338430173294\n",
      "True negative: 979, True positive: 16, False negative: 2, False positive: 965\n",
      "Epoch 420: Accuracy 50.66%\n",
      "Average true similarity: -2.174608562691131, Average false similarity: -2.1890336391437315\n",
      "True negative: 979, True positive: 15, False negative: 2, False positive: 966\n",
      "Epoch 421: Accuracy 50.71%\n",
      "Average true similarity: -2.174861365953109, Average false similarity: -2.1894209989806312\n",
      "True negative: 978, True positive: 17, False negative: 3, False positive: 964\n",
      "Epoch 422: Accuracy 50.66%\n",
      "Average true similarity: -2.175639143730886, Average false similarity: -2.190963302752296\n",
      "True negative: 980, True positive: 14, False negative: 1, False positive: 967\n",
      "Epoch 423: Accuracy 50.51%\n",
      "Average true similarity: -2.175581039755348, Average false similarity: -2.1905249745158004\n",
      "True negative: 976, True positive: 15, False negative: 5, False positive: 966\n",
      "Epoch 424: Accuracy 50.51%\n",
      "Average true similarity: -2.1758532110091755, Average false similarity: -2.1907135575942913\n",
      "True negative: 976, True positive: 15, False negative: 5, False positive: 966\n",
      "Epoch 425: Accuracy 50.51%\n",
      "Average true similarity: -2.176013251783895, Average false similarity: -2.189513761467889\n",
      "True negative: 978, True positive: 13, False negative: 3, False positive: 968\n",
      "Epoch 426: Accuracy 50.76%\n",
      "Average true similarity: -2.176245667686037, Average false similarity: -2.190964322120283\n",
      "True negative: 979, True positive: 17, False negative: 2, False positive: 964\n",
      "Epoch 427: Accuracy 50.66%\n",
      "Average true similarity: -2.1775096839959227, Average false similarity: -2.192469928644241\n",
      "True negative: 980, True positive: 14, False negative: 1, False positive: 967\n",
      "Epoch 428: Accuracy 50.76%\n",
      "Average true similarity: -2.1770295616717656, Average false similarity: -2.191812436289502\n",
      "True negative: 980, True positive: 16, False negative: 1, False positive: 965\n",
      "Epoch 429: Accuracy 50.66%\n",
      "Average true similarity: -2.1780101936799148, Average false similarity: -2.1929429153924582\n",
      "True negative: 979, True positive: 15, False negative: 2, False positive: 966\n",
      "Epoch 430: Accuracy 50.46%\n",
      "Average true similarity: -2.1793822629969406, Average false similarity: -2.195106014271153\n",
      "True negative: 980, True positive: 10, False negative: 1, False positive: 971\n",
      "Epoch 431: Accuracy 50.56%\n",
      "Average true similarity: -2.179145769622835, Average false similarity: -2.1937115188583074\n",
      "True negative: 978, True positive: 14, False negative: 3, False positive: 967\n",
      "Epoch 432: Accuracy 50.56%\n",
      "Average true similarity: -2.178753312945972, Average false similarity: -2.1940866462793083\n",
      "True negative: 981, True positive: 11, False negative: 0, False positive: 970\n",
      "Epoch 433: Accuracy 50.41%\n",
      "Average true similarity: -2.1796605504587165, Average false similarity: -2.194688073394495\n",
      "True negative: 979, True positive: 10, False negative: 2, False positive: 971\n",
      "Epoch 434: Accuracy 50.46%\n",
      "Average true similarity: -2.1801794087665645, Average false similarity: -2.194963302752291\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 435: Accuracy 50.56%\n",
      "Average true similarity: -2.1802976554536166, Average false similarity: -2.195557594291539\n",
      "True negative: 980, True positive: 12, False negative: 1, False positive: 969\n",
      "Epoch 436: Accuracy 50.36%\n",
      "Average true similarity: -2.1800723751274202, Average false similarity: -2.1948165137614657\n",
      "True negative: 979, True positive: 9, False negative: 2, False positive: 972\n",
      "Epoch 437: Accuracy 50.41%\n",
      "Average true similarity: -2.1808083588175324, Average false similarity: -2.1954464831804295\n",
      "True negative: 979, True positive: 10, False negative: 2, False positive: 971\n",
      "Epoch 438: Accuracy 50.31%\n",
      "Average true similarity: -2.180386340468909, Average false similarity: -2.195148827726809\n",
      "True negative: 978, True positive: 9, False negative: 3, False positive: 972\n",
      "Epoch 439: Accuracy 50.51%\n",
      "Average true similarity: -2.1809357798165117, Average false similarity: -2.1955208970438322\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 440: Accuracy 50.46%\n",
      "Average true similarity: -2.1812446483180428, Average false similarity: -2.196598369011213\n",
      "True negative: 980, True positive: 10, False negative: 1, False positive: 971\n",
      "Epoch 441: Accuracy 50.46%\n",
      "Average true similarity: -2.180858307849133, Average false similarity: -2.195086646279302\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 442: Accuracy 50.51%\n",
      "Average true similarity: -2.1796472986748214, Average false similarity: -2.194826707441389\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 443: Accuracy 50.46%\n",
      "Average true similarity: -2.179754332313963, Average false similarity: -2.1945545361875634\n",
      "True negative: 977, True positive: 13, False negative: 4, False positive: 968\n",
      "Epoch 444: Accuracy 50.36%\n",
      "Average true similarity: -2.178906218144748, Average false similarity: -2.193941896024463\n",
      "True negative: 980, True positive: 8, False negative: 1, False positive: 973\n",
      "Epoch 445: Accuracy 50.56%\n",
      "Average true similarity: -2.1772528032619785, Average false similarity: -2.1927686034658516\n",
      "True negative: 978, True positive: 14, False negative: 3, False positive: 967\n",
      "Epoch 446: Accuracy 50.66%\n",
      "Average true similarity: -2.176384301732924, Average false similarity: -2.191023445463812\n",
      "True negative: 980, True positive: 14, False negative: 1, False positive: 967\n",
      "Epoch 447: Accuracy 50.61%\n",
      "Average true similarity: -2.176869520897042, Average false similarity: -2.1914260958205913\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 448: Accuracy 50.61%\n",
      "Average true similarity: -2.1763445463812423, Average false similarity: -2.191036697247709\n",
      "True negative: 977, True positive: 16, False negative: 4, False positive: 965\n",
      "Epoch 449: Accuracy 50.82%\n",
      "Average true similarity: -2.1759296636085628, Average false similarity: -2.190662589194697\n",
      "True negative: 979, True positive: 18, False negative: 2, False positive: 963\n",
      "Epoch 450: Accuracy 50.66%\n",
      "Average true similarity: -2.1767370030581032, Average false similarity: -2.1912548419979574\n",
      "True negative: 979, True positive: 15, False negative: 2, False positive: 966\n",
      "Epoch 451: Accuracy 50.76%\n",
      "Average true similarity: -2.1776829765545376, Average false similarity: -2.1926095820591236\n",
      "True negative: 979, True positive: 17, False negative: 2, False positive: 964\n",
      "Epoch 452: Accuracy 50.61%\n",
      "Average true similarity: -2.1775270132517797, Average false similarity: -2.1926197757390393\n",
      "True negative: 980, True positive: 13, False negative: 1, False positive: 968\n",
      "Epoch 453: Accuracy 50.46%\n",
      "Average true similarity: -2.1780214067278316, Average false similarity: -2.1925423037716634\n",
      "True negative: 977, True positive: 13, False negative: 4, False positive: 968\n",
      "Epoch 454: Accuracy 50.41%\n",
      "Average true similarity: -2.1789235474006126, Average false similarity: -2.1941743119266093\n",
      "True negative: 980, True positive: 9, False negative: 1, False positive: 972\n",
      "Epoch 455: Accuracy 50.51%\n",
      "Average true similarity: -2.1787400611620793, Average false similarity: -2.19385626911315\n",
      "True negative: 980, True positive: 11, False negative: 1, False positive: 970\n",
      "Epoch 456: Accuracy 50.76%\n",
      "Average true similarity: -2.1780377166156986, Average false similarity: -2.1927533129459755\n",
      "True negative: 980, True positive: 16, False negative: 1, False positive: 965\n",
      "Epoch 457: Accuracy 50.46%\n",
      "Average true similarity: -2.1774546381243636, Average false similarity: -2.192371049949032\n",
      "True negative: 980, True positive: 10, False negative: 1, False positive: 971\n",
      "Epoch 458: Accuracy 50.36%\n",
      "Average true similarity: -2.1776748216105974, Average false similarity: -2.1922395514780844\n",
      "True negative: 976, True positive: 12, False negative: 5, False positive: 969\n",
      "Epoch 459: Accuracy 50.66%\n",
      "Average true similarity: -2.177106014271152, Average false similarity: -2.193068297655453\n",
      "True negative: 979, True positive: 15, False negative: 2, False positive: 966\n",
      "Epoch 460: Accuracy 50.2%\n",
      "Average true similarity: -2.1770530071355783, Average false similarity: -2.1920682976554544\n",
      "True negative: 977, True positive: 8, False negative: 4, False positive: 973\n",
      "Epoch 461: Accuracy 50.56%\n",
      "Average true similarity: -2.176800203873598, Average false similarity: -2.191878695208973\n",
      "True negative: 979, True positive: 13, False negative: 2, False positive: 968\n",
      "Epoch 462: Accuracy 50.61%\n",
      "Average true similarity: -2.177084607543322, Average false similarity: -2.1913292558613677\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 463: Accuracy 50.31%\n",
      "Average true similarity: -2.1764077471967394, Average false similarity: -2.191585117227317\n",
      "True negative: 978, True positive: 9, False negative: 3, False positive: 972\n",
      "Epoch 464: Accuracy 50.61%\n",
      "Average true similarity: -2.1760917431192657, Average false similarity: -2.1907849133537214\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 465: Accuracy 50.46%\n",
      "Average true similarity: -2.176282364933743, Average false similarity: -2.1912844036697243\n",
      "True negative: 978, True positive: 12, False negative: 3, False positive: 969\n",
      "Epoch 466: Accuracy 50.56%\n",
      "Average true similarity: -2.1747115188583104, Average false similarity: -2.189709480122325\n",
      "True negative: 978, True positive: 14, False negative: 3, False positive: 967\n",
      "Epoch 467: Accuracy 50.51%\n",
      "Average true similarity: -2.1735239551478096, Average false similarity: -2.1882823649337393\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 468: Accuracy 50.82%\n",
      "Average true similarity: -2.173202854230378, Average false similarity: -2.1878450560652407\n",
      "True negative: 977, True positive: 20, False negative: 4, False positive: 961\n",
      "Epoch 469: Accuracy 50.66%\n",
      "Average true similarity: -2.172954128440367, Average false similarity: -2.187532110091744\n",
      "True negative: 978, True positive: 16, False negative: 3, False positive: 965\n",
      "Epoch 470: Accuracy 50.61%\n",
      "Average true similarity: -2.1735616717635047, Average false similarity: -2.188892966360857\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 471: Accuracy 50.56%\n",
      "Average true similarity: -2.174367991845055, Average false similarity: -2.1887420998980645\n",
      "True negative: 980, True positive: 12, False negative: 1, False positive: 969\n",
      "Epoch 472: Accuracy 50.61%\n",
      "Average true similarity: -2.175060142711518, Average false similarity: -2.1900122324159037\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 473: Accuracy 50.46%\n",
      "Average true similarity: -2.1757991845056055, Average false similarity: -2.1903506625891946\n",
      "True negative: 981, True positive: 9, False negative: 0, False positive: 972\n",
      "Epoch 474: Accuracy 50.36%\n",
      "Average true similarity: -2.1762731906218153, Average false similarity: -2.191238532110091\n",
      "True negative: 977, True positive: 11, False negative: 4, False positive: 970\n",
      "Epoch 475: Accuracy 50.51%\n",
      "Average true similarity: -2.1757359836901164, Average false similarity: -2.190635066258916\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 476: Accuracy 50.46%\n",
      "Average true similarity: -2.1761997961264012, Average false similarity: -2.1907573904179394\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 477: Accuracy 50.82%\n",
      "Average true similarity: -2.1759082568807337, Average false similarity: -2.191307849133537\n",
      "True negative: 980, True positive: 17, False negative: 1, False positive: 964\n",
      "Epoch 478: Accuracy 50.56%\n",
      "Average true similarity: -2.1758277268093793, Average false similarity: -2.190691131498471\n",
      "True negative: 978, True positive: 14, False negative: 3, False positive: 967\n",
      "Epoch 479: Accuracy 50.51%\n",
      "Average true similarity: -2.1754291539245627, Average false similarity: -2.190101936799187\n",
      "True negative: 981, True positive: 10, False negative: 0, False positive: 971\n",
      "Epoch 480: Accuracy 50.51%\n",
      "Average true similarity: -2.175385321100914, Average false similarity: -2.1899215086646273\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 481: Accuracy 50.56%\n",
      "Average true similarity: -2.1752782874617718, Average false similarity: -2.1900061162079494\n",
      "True negative: 978, True positive: 14, False negative: 3, False positive: 967\n",
      "Epoch 482: Accuracy 50.66%\n",
      "Average true similarity: -2.1745575942915423, Average false similarity: -2.1889194699286447\n",
      "True negative: 977, True positive: 17, False negative: 4, False positive: 964\n",
      "Epoch 483: Accuracy 50.46%\n",
      "Average true similarity: -2.175195718654434, Average false similarity: -2.1911580020387373\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 484: Accuracy 50.51%\n",
      "Average true similarity: -2.1752752293578013, Average false similarity: -2.189409785932723\n",
      "True negative: 978, True positive: 13, False negative: 3, False positive: 968\n",
      "Epoch 485: Accuracy 50.41%\n",
      "Average true similarity: -2.1752140672782887, Average false similarity: -2.1907237512742084\n",
      "True negative: 981, True positive: 8, False negative: 0, False positive: 973\n",
      "Epoch 486: Accuracy 50.76%\n",
      "Average true similarity: -2.174771661569824, Average false similarity: -2.19005912334353\n",
      "True negative: 980, True positive: 16, False negative: 1, False positive: 965\n",
      "Epoch 487: Accuracy 50.46%\n",
      "Average true similarity: -2.175375127420996, Average false similarity: -2.1909714576962274\n",
      "True negative: 980, True positive: 10, False negative: 1, False positive: 971\n",
      "Epoch 488: Accuracy 50.66%\n",
      "Average true similarity: -2.1752334352701306, Average false similarity: -2.1903720693170214\n",
      "True negative: 979, True positive: 15, False negative: 2, False positive: 966\n",
      "Epoch 489: Accuracy 50.46%\n",
      "Average true similarity: -2.175499490316003, Average false similarity: -2.1899836901121326\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 490: Accuracy 50.51%\n",
      "Average true similarity: -2.1742599388379205, Average false similarity: -2.1894046890927608\n",
      "True negative: 980, True positive: 11, False negative: 1, False positive: 970\n",
      "Epoch 491: Accuracy 50.66%\n",
      "Average true similarity: -2.1750214067278306, Average false similarity: -2.1899541284403687\n",
      "True negative: 978, True positive: 16, False negative: 3, False positive: 965\n",
      "Epoch 492: Accuracy 50.61%\n",
      "Average true similarity: -2.1756799184505606, Average false similarity: -2.189767584097861\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 493: Accuracy 50.56%\n",
      "Average true similarity: -2.176483180428132, Average false similarity: -2.1914291539245654\n",
      "True negative: 980, True positive: 12, False negative: 1, False positive: 969\n",
      "Epoch 494: Accuracy 50.66%\n",
      "Average true similarity: -2.1753170234454626, Average false similarity: -2.1903323139653392\n",
      "True negative: 979, True positive: 15, False negative: 2, False positive: 966\n",
      "Epoch 495: Accuracy 50.61%\n",
      "Average true similarity: -2.1755973496432213, Average false similarity: -2.190702344546384\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 496: Accuracy 50.51%\n",
      "Average true similarity: -2.175182466870539, Average false similarity: -2.1899847094801217\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 497: Accuracy 50.76%\n",
      "Average true similarity: -2.1754301732925607, Average false similarity: -2.191034658511723\n",
      "True negative: 980, True positive: 16, False negative: 1, False positive: 965\n",
      "Epoch 498: Accuracy 50.76%\n",
      "Average true similarity: -2.1749480122324165, Average false similarity: -2.1895820591233437\n",
      "True negative: 977, True positive: 19, False negative: 4, False positive: 962\n",
      "Epoch 499: Accuracy 50.61%\n",
      "Average true similarity: -2.174430173292562, Average false similarity: -2.1898746177370016\n",
      "True negative: 977, True positive: 16, False negative: 4, False positive: 965\n",
      "Epoch 500: Accuracy 50.46%\n",
      "Average true similarity: -2.1747614678899083, Average false similarity: -2.1895871559633053\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 501: Accuracy 50.76%\n",
      "Average true similarity: -2.175125382263, Average false similarity: -2.1899001019368014\n",
      "True negative: 978, True positive: 18, False negative: 3, False positive: 963\n",
      "Epoch 502: Accuracy 50.76%\n",
      "Average true similarity: -2.1744974515800215, Average false similarity: -2.189458715596331\n",
      "True negative: 980, True positive: 16, False negative: 1, False positive: 965\n",
      "Epoch 503: Accuracy 50.56%\n",
      "Average true similarity: -2.174510703363913, Average false similarity: -2.1898002038735984\n",
      "True negative: 980, True positive: 12, False negative: 1, False positive: 969\n",
      "Epoch 504: Accuracy 50.61%\n",
      "Average true similarity: -2.1747176350662594, Average false similarity: -2.1902701325178358\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 505: Accuracy 50.56%\n",
      "Average true similarity: -2.174794087665647, Average false similarity: -2.1895647298674836\n",
      "True negative: 978, True positive: 14, False negative: 3, False positive: 967\n",
      "Epoch 506: Accuracy 50.51%\n",
      "Average true similarity: -2.1743567787971427, Average false similarity: -2.1891304791029533\n",
      "True negative: 978, True positive: 13, False negative: 3, False positive: 968\n",
      "Epoch 507: Accuracy 50.66%\n",
      "Average true similarity: -2.1725820591233407, Average false similarity: -2.1883853211009185\n",
      "True negative: 979, True positive: 15, False negative: 2, False positive: 966\n",
      "Epoch 508: Accuracy 50.71%\n",
      "Average true similarity: -2.1725086646279332, Average false similarity: -2.1879459734964324\n",
      "True negative: 978, True positive: 17, False negative: 3, False positive: 964\n",
      "Epoch 509: Accuracy 50.56%\n",
      "Average true similarity: -2.1706860346585133, Average false similarity: -2.185495412844035\n",
      "True negative: 979, True positive: 13, False negative: 2, False positive: 968\n",
      "Epoch 510: Accuracy 50.71%\n",
      "Average true similarity: -2.1703751274210004, Average false similarity: -2.185570846075433\n",
      "True negative: 974, True positive: 21, False negative: 7, False positive: 960\n",
      "Epoch 511: Accuracy 50.71%\n",
      "Average true similarity: -2.170146788990826, Average false similarity: -2.185090723751275\n",
      "True negative: 976, True positive: 19, False negative: 5, False positive: 962\n",
      "Epoch 512: Accuracy 50.97%\n",
      "Average true similarity: -2.169453618756373, Average false similarity: -2.1848399592252834\n",
      "True negative: 978, True positive: 22, False negative: 3, False positive: 959\n",
      "Epoch 513: Accuracy 50.92%\n",
      "Average true similarity: -2.169540265035677, Average false similarity: -2.185172273190623\n",
      "True negative: 978, True positive: 21, False negative: 3, False positive: 960\n",
      "Epoch 514: Accuracy 50.61%\n",
      "Average true similarity: -2.170187563710496, Average false similarity: -2.1848970438328212\n",
      "True negative: 976, True positive: 17, False negative: 5, False positive: 964\n",
      "Epoch 515: Accuracy 50.66%\n",
      "Average true similarity: -2.1696401630988786, Average false similarity: -2.184642201834863\n",
      "True negative: 978, True positive: 16, False negative: 3, False positive: 965\n",
      "Epoch 516: Accuracy 50.66%\n",
      "Average true similarity: -2.170425076452599, Average false similarity: -2.184804281345562\n",
      "True negative: 978, True positive: 16, False negative: 3, False positive: 965\n",
      "Epoch 517: Accuracy 50.51%\n",
      "Average true similarity: -2.1710978593272188, Average false similarity: -2.185204892966363\n",
      "True negative: 977, True positive: 14, False negative: 4, False positive: 967\n",
      "Epoch 518: Accuracy 50.82%\n",
      "Average true similarity: -2.1719072375127437, Average false similarity: -2.186497451580019\n",
      "True negative: 979, True positive: 18, False negative: 2, False positive: 963\n",
      "Epoch 519: Accuracy 50.76%\n",
      "Average true similarity: -2.172090723751272, Average false similarity: -2.187170234454637\n",
      "True negative: 980, True positive: 16, False negative: 1, False positive: 965\n",
      "Epoch 520: Accuracy 50.61%\n",
      "Average true similarity: -2.170682976554536, Average false similarity: -2.185187563710499\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 521: Accuracy 50.76%\n",
      "Average true similarity: -2.168886850152907, Average false similarity: -2.1832752293578004\n",
      "True negative: 976, True positive: 20, False negative: 5, False positive: 961\n",
      "Epoch 522: Accuracy 50.82%\n",
      "Average true similarity: -2.1686136595310934, Average false similarity: -2.184013251783894\n",
      "True negative: 977, True positive: 20, False negative: 4, False positive: 961\n",
      "Epoch 523: Accuracy 51.02%\n",
      "Average true similarity: -2.168633027522933, Average false similarity: -2.18374617737003\n",
      "True negative: 978, True positive: 23, False negative: 3, False positive: 958\n",
      "Epoch 524: Accuracy 50.97%\n",
      "Average true similarity: -2.1685983690112116, Average false similarity: -2.183358817533129\n",
      "True negative: 978, True positive: 22, False negative: 3, False positive: 959\n",
      "Epoch 525: Accuracy 50.71%\n",
      "Average true similarity: -2.169542303771663, Average false similarity: -2.1838012232415895\n",
      "True negative: 977, True positive: 18, False negative: 4, False positive: 963\n",
      "Epoch 526: Accuracy 50.66%\n",
      "Average true similarity: -2.1688032619775743, Average false similarity: -2.1837777777777796\n",
      "True negative: 976, True positive: 18, False negative: 5, False positive: 963\n",
      "Epoch 527: Accuracy 50.97%\n",
      "Average true similarity: -2.168166156982671, Average false similarity: -2.1829531090723737\n",
      "True negative: 977, True positive: 23, False negative: 4, False positive: 958\n",
      "Epoch 528: Accuracy 50.82%\n",
      "Average true similarity: -2.1668185524974497, Average false similarity: -2.181763506625892\n",
      "True negative: 975, True positive: 22, False negative: 6, False positive: 959\n",
      "Epoch 529: Accuracy 50.87%\n",
      "Average true similarity: -2.166792048929662, Average false similarity: -2.181746177370031\n",
      "True negative: 975, True positive: 23, False negative: 6, False positive: 958\n",
      "Epoch 530: Accuracy 50.76%\n",
      "Average true similarity: -2.1669245667686012, Average false similarity: -2.1820275229357806\n",
      "True negative: 974, True positive: 22, False negative: 7, False positive: 959\n",
      "Epoch 531: Accuracy 50.82%\n",
      "Average true similarity: -2.166196738022427, Average false similarity: -2.181316004077474\n",
      "True negative: 978, True positive: 19, False negative: 3, False positive: 962\n",
      "Epoch 532: Accuracy 50.87%\n",
      "Average true similarity: -2.166689092762487, Average false similarity: -2.1811641182466905\n",
      "True negative: 977, True positive: 21, False negative: 4, False positive: 960\n",
      "Epoch 533: Accuracy 50.92%\n",
      "Average true similarity: -2.1676829765545347, Average false similarity: -2.183200815494392\n",
      "True negative: 977, True positive: 22, False negative: 4, False positive: 959\n",
      "Epoch 534: Accuracy 50.76%\n",
      "Average true similarity: -2.1687054026503563, Average false similarity: -2.183638124362897\n",
      "True negative: 975, True positive: 21, False negative: 6, False positive: 960\n",
      "Epoch 535: Accuracy 50.41%\n",
      "Average true similarity: -2.169389398572884, Average false similarity: -2.184404689092762\n",
      "True negative: 976, True positive: 13, False negative: 5, False positive: 968\n",
      "Epoch 536: Accuracy 50.66%\n",
      "Average true similarity: -2.1697808358817516, Average false similarity: -2.1842059123343516\n",
      "True negative: 978, True positive: 16, False negative: 3, False positive: 965\n",
      "Epoch 537: Accuracy 50.82%\n",
      "Average true similarity: -2.1685963302752294, Average false similarity: -2.1837359836901107\n",
      "True negative: 979, True positive: 18, False negative: 2, False positive: 963\n",
      "Epoch 538: Accuracy 50.56%\n",
      "Average true similarity: -2.169572884811417, Average false similarity: -2.184673802242611\n",
      "True negative: 978, True positive: 14, False negative: 3, False positive: 967\n",
      "Epoch 539: Accuracy 50.56%\n",
      "Average true similarity: -2.169030581039755, Average false similarity: -2.1845249745158006\n",
      "True negative: 978, True positive: 14, False negative: 3, False positive: 967\n",
      "Epoch 540: Accuracy 50.76%\n",
      "Average true similarity: -2.1691508664627905, Average false similarity: -2.1844036697247677\n",
      "True negative: 979, True positive: 17, False negative: 2, False positive: 964\n",
      "Epoch 541: Accuracy 50.56%\n",
      "Average true similarity: -2.169660550458716, Average false similarity: -2.1845310907237527\n",
      "True negative: 976, True positive: 16, False negative: 5, False positive: 965\n",
      "Epoch 542: Accuracy 50.71%\n",
      "Average true similarity: -2.1695168195718666, Average false similarity: -2.1846890927624867\n",
      "True negative: 980, True positive: 15, False negative: 1, False positive: 966\n",
      "Epoch 543: Accuracy 50.61%\n",
      "Average true similarity: -2.169029561671761, Average false similarity: -2.1836442405708434\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 544: Accuracy 51.12%\n",
      "Average true similarity: -2.1688175331294595, Average false similarity: -2.1837604485219138\n",
      "True negative: 979, True positive: 24, False negative: 2, False positive: 957\n",
      "Epoch 545: Accuracy 50.71%\n",
      "Average true similarity: -2.168279306829764, Average false similarity: -2.182699286442407\n",
      "True negative: 978, True positive: 17, False negative: 3, False positive: 964\n",
      "Epoch 546: Accuracy 50.76%\n",
      "Average true similarity: -2.1681396534148827, Average false similarity: -2.1837471967380244\n",
      "True negative: 977, True positive: 19, False negative: 4, False positive: 962\n",
      "Epoch 547: Accuracy 50.61%\n",
      "Average true similarity: -2.168582059123344, Average false similarity: -2.183113149847092\n",
      "True negative: 977, True positive: 16, False negative: 4, False positive: 965\n",
      "Epoch 548: Accuracy 50.82%\n",
      "Average true similarity: -2.1689734964322134, Average false similarity: -2.184445463812437\n",
      "True negative: 979, True positive: 18, False negative: 2, False positive: 963\n",
      "Epoch 549: Accuracy 50.66%\n",
      "Average true similarity: -2.169276248725787, Average false similarity: -2.1841580020387363\n",
      "True negative: 980, True positive: 14, False negative: 1, False positive: 967\n",
      "Epoch 550: Accuracy 50.71%\n",
      "Average true similarity: -2.169694189602449, Average false similarity: -2.1855586136595306\n",
      "True negative: 979, True positive: 16, False negative: 2, False positive: 965\n",
      "Epoch 551: Accuracy 50.76%\n",
      "Average true similarity: -2.1705616717635063, Average false similarity: -2.18555657492355\n",
      "True negative: 979, True positive: 17, False negative: 2, False positive: 964\n",
      "Epoch 552: Accuracy 50.61%\n",
      "Average true similarity: -2.169360856269111, Average false similarity: -2.1843669724770667\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 553: Accuracy 50.56%\n",
      "Average true similarity: -2.1691314984709473, Average false similarity: -2.1842680937818564\n",
      "True negative: 977, True positive: 15, False negative: 4, False positive: 966\n",
      "Epoch 554: Accuracy 50.82%\n",
      "Average true similarity: -2.1680458715596336, Average false similarity: -2.183298674821611\n",
      "True negative: 979, True positive: 18, False negative: 2, False positive: 963\n",
      "Epoch 555: Accuracy 50.97%\n",
      "Average true similarity: -2.168333333333334, Average false similarity: -2.1834169215086656\n",
      "True negative: 980, True positive: 20, False negative: 1, False positive: 961\n",
      "Epoch 556: Accuracy 50.87%\n",
      "Average true similarity: -2.169046890927624, Average false similarity: -2.1839755351681958\n",
      "True negative: 979, True positive: 19, False negative: 2, False positive: 962\n",
      "Epoch 557: Accuracy 51.02%\n",
      "Average true similarity: -2.1695321100917435, Average false similarity: -2.184661569826706\n",
      "True negative: 979, True positive: 22, False negative: 2, False positive: 959\n",
      "Epoch 558: Accuracy 50.71%\n",
      "Average true similarity: -2.1690428134556594, Average false similarity: -2.183848114169215\n",
      "True negative: 979, True positive: 16, False negative: 2, False positive: 965\n",
      "Epoch 559: Accuracy 50.71%\n",
      "Average true similarity: -2.169169215086648, Average false similarity: -2.1838786952089673\n",
      "True negative: 978, True positive: 17, False negative: 3, False positive: 964\n",
      "Epoch 560: Accuracy 51.02%\n",
      "Average true similarity: -2.168793068297653, Average false similarity: -2.1836126401630986\n",
      "True negative: 979, True positive: 22, False negative: 2, False positive: 959\n",
      "Epoch 561: Accuracy 50.61%\n",
      "Average true similarity: -2.1690581039755337, Average false similarity: -2.184080530071355\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 562: Accuracy 50.82%\n",
      "Average true similarity: -2.1698277268093817, Average false similarity: -2.1845820591233416\n",
      "True negative: 979, True positive: 18, False negative: 2, False positive: 963\n",
      "Epoch 563: Accuracy 50.66%\n",
      "Average true similarity: -2.1690591233435286, Average false similarity: -2.1836717635066263\n",
      "True negative: 977, True positive: 17, False negative: 4, False positive: 964\n",
      "Epoch 564: Accuracy 50.71%\n",
      "Average true similarity: -2.1694556574923536, Average false similarity: -2.1847675840978606\n",
      "True negative: 979, True positive: 16, False negative: 2, False positive: 965\n",
      "Epoch 565: Accuracy 50.87%\n",
      "Average true similarity: -2.1705076452599408, Average false similarity: -2.186290519877674\n",
      "True negative: 980, True positive: 18, False negative: 1, False positive: 963\n",
      "Epoch 566: Accuracy 50.82%\n",
      "Average true similarity: -2.1706483180428147, Average false similarity: -2.185705402650355\n",
      "True negative: 980, True positive: 17, False negative: 1, False positive: 964\n",
      "Epoch 567: Accuracy 50.66%\n",
      "Average true similarity: -2.1708827726809403, Average false similarity: -2.1860326197757387\n",
      "True negative: 979, True positive: 15, False negative: 2, False positive: 966\n",
      "Epoch 568: Accuracy 50.61%\n",
      "Average true similarity: -2.17210499490316, Average false similarity: -2.1865851172273136\n",
      "True negative: 981, True positive: 12, False negative: 0, False positive: 969\n",
      "Epoch 569: Accuracy 50.61%\n",
      "Average true similarity: -2.171089704383283, Average false similarity: -2.185680937818551\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 570: Accuracy 50.82%\n",
      "Average true similarity: -2.170654434250766, Average false similarity: -2.1863272171253842\n",
      "True negative: 978, True positive: 19, False negative: 3, False positive: 962\n",
      "Epoch 571: Accuracy 50.87%\n",
      "Average true similarity: -2.170539245667686, Average false similarity: -2.1854913353720717\n",
      "True negative: 981, True positive: 17, False negative: 0, False positive: 964\n",
      "Epoch 572: Accuracy 50.76%\n",
      "Average true similarity: -2.1708715596330275, Average false similarity: -2.1858267074413837\n",
      "True negative: 978, True positive: 18, False negative: 3, False positive: 963\n",
      "Epoch 573: Accuracy 50.56%\n",
      "Average true similarity: -2.1718144750254877, Average false similarity: -2.186780835881753\n",
      "True negative: 980, True positive: 12, False negative: 1, False positive: 969\n",
      "Epoch 574: Accuracy 50.51%\n",
      "Average true similarity: -2.1716982670744103, Average false similarity: -2.1866992864424035\n",
      "True negative: 977, True positive: 14, False negative: 4, False positive: 967\n",
      "Epoch 575: Accuracy 50.61%\n",
      "Average true similarity: -2.172394495412844, Average false similarity: -2.1874923547400607\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 576: Accuracy 50.46%\n",
      "Average true similarity: -2.1727084607543325, Average false similarity: -2.1876177370030567\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 577: Accuracy 50.46%\n",
      "Average true similarity: -2.173030581039754, Average false similarity: -2.1878042813455627\n",
      "True negative: 980, True positive: 10, False negative: 1, False positive: 971\n",
      "Epoch 578: Accuracy 50.71%\n",
      "Average true similarity: -2.172334352701325, Average false similarity: -2.187913353720694\n",
      "True negative: 979, True positive: 16, False negative: 2, False positive: 965\n",
      "Epoch 579: Accuracy 50.66%\n",
      "Average true similarity: -2.172585117227318, Average false similarity: -2.187396534148829\n",
      "True negative: 979, True positive: 15, False negative: 2, False positive: 966\n",
      "Epoch 580: Accuracy 50.46%\n",
      "Average true similarity: -2.17107849133537, Average false similarity: -2.1859989806320064\n",
      "True negative: 977, True positive: 13, False negative: 4, False positive: 968\n",
      "Epoch 581: Accuracy 50.66%\n",
      "Average true similarity: -2.170965341488279, Average false similarity: -2.1862487257900103\n",
      "True negative: 979, True positive: 15, False negative: 2, False positive: 966\n",
      "Epoch 582: Accuracy 50.76%\n",
      "Average true similarity: -2.171626911314983, Average false similarity: -2.1865677879714593\n",
      "True negative: 979, True positive: 17, False negative: 2, False positive: 964\n",
      "Epoch 583: Accuracy 50.66%\n",
      "Average true similarity: -2.1703924566768595, Average false similarity: -2.185657492354741\n",
      "True negative: 978, True positive: 16, False negative: 3, False positive: 965\n",
      "Epoch 584: Accuracy 50.61%\n",
      "Average true similarity: -2.1720540265035684, Average false similarity: -2.187081549439346\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 585: Accuracy 50.61%\n",
      "Average true similarity: -2.1725290519877687, Average false similarity: -2.1876228338430193\n",
      "True negative: 981, True positive: 12, False negative: 0, False positive: 969\n",
      "Epoch 586: Accuracy 50.61%\n",
      "Average true similarity: -2.1713527013251785, Average false similarity: -2.186721712538226\n",
      "True negative: 981, True positive: 12, False negative: 0, False positive: 969\n",
      "Epoch 587: Accuracy 50.31%\n",
      "Average true similarity: -2.1718919469928633, Average false similarity: -2.187097859327218\n",
      "True negative: 978, True positive: 9, False negative: 3, False positive: 972\n",
      "Epoch 588: Accuracy 50.51%\n",
      "Average true similarity: -2.1719653414882747, Average false similarity: -2.1873843017329264\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 589: Accuracy 50.51%\n",
      "Average true similarity: -2.1723710499490303, Average false similarity: -2.1877054026503595\n",
      "True negative: 980, True positive: 11, False negative: 1, False positive: 970\n",
      "Epoch 590: Accuracy 50.46%\n",
      "Average true similarity: -2.1722099898063205, Average false similarity: -2.1881508664627964\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 591: Accuracy 50.36%\n",
      "Average true similarity: -2.1724077471967367, Average false similarity: -2.1872375127421027\n",
      "True negative: 977, True positive: 11, False negative: 4, False positive: 970\n",
      "Epoch 592: Accuracy 50.51%\n",
      "Average true similarity: -2.172973496432212, Average false similarity: -2.188254841997961\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 593: Accuracy 50.31%\n",
      "Average true similarity: -2.173680937818555, Average false similarity: -2.1891243628950057\n",
      "True negative: 979, True positive: 8, False negative: 2, False positive: 973\n",
      "Epoch 594: Accuracy 50.66%\n",
      "Average true similarity: -2.1734862385321096, Average false similarity: -2.1892538226299676\n",
      "True negative: 980, True positive: 14, False negative: 1, False positive: 967\n",
      "Epoch 595: Accuracy 50.66%\n",
      "Average true similarity: -2.1734220183486235, Average false similarity: -2.1884831804281357\n",
      "True negative: 980, True positive: 14, False negative: 1, False positive: 967\n",
      "Epoch 596: Accuracy 50.66%\n",
      "Average true similarity: -2.172253822629969, Average false similarity: -2.1875769622833867\n",
      "True negative: 980, True positive: 14, False negative: 1, False positive: 967\n",
      "Epoch 597: Accuracy 50.46%\n",
      "Average true similarity: -2.172980632008156, Average false similarity: -2.188496432212031\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 598: Accuracy 50.51%\n",
      "Average true similarity: -2.1726666666666663, Average false similarity: -2.187584097859326\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 599: Accuracy 50.51%\n",
      "Average true similarity: -2.173317023445465, Average false similarity: -2.1881753312945973\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 600: Accuracy 50.76%\n",
      "Average true similarity: -2.172517838939856, Average false similarity: -2.187916411824671\n",
      "True negative: 978, True positive: 18, False negative: 3, False positive: 963\n",
      "Epoch 601: Accuracy 50.71%\n",
      "Average true similarity: -2.172321100917428, Average false similarity: -2.187327217125382\n",
      "True negative: 979, True positive: 16, False negative: 2, False positive: 965\n",
      "Epoch 602: Accuracy 50.36%\n",
      "Average true similarity: -2.1734311926605505, Average false similarity: -2.188927624872577\n",
      "True negative: 978, True positive: 10, False negative: 3, False positive: 971\n",
      "Epoch 603: Accuracy 50.46%\n",
      "Average true similarity: -2.173244648318043, Average false similarity: -2.188555555555555\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 604: Accuracy 50.51%\n",
      "Average true similarity: -2.172672782874616, Average false similarity: -2.1883353720693197\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 605: Accuracy 50.66%\n",
      "Average true similarity: -2.172709480122323, Average false similarity: -2.187941896024467\n",
      "True negative: 979, True positive: 15, False negative: 2, False positive: 966\n",
      "Epoch 606: Accuracy 50.71%\n",
      "Average true similarity: -2.1733017329255855, Average false similarity: -2.187851172273189\n",
      "True negative: 980, True positive: 15, False negative: 1, False positive: 966\n",
      "Epoch 607: Accuracy 50.46%\n",
      "Average true similarity: -2.172538226299696, Average false similarity: -2.1876738022426094\n",
      "True negative: 978, True positive: 12, False negative: 3, False positive: 969\n",
      "Epoch 608: Accuracy 50.76%\n",
      "Average true similarity: -2.1716146788990818, Average false similarity: -2.1871702344546384\n",
      "True negative: 978, True positive: 18, False negative: 3, False positive: 963\n",
      "Epoch 609: Accuracy 50.41%\n",
      "Average true similarity: -2.172614678899083, Average false similarity: -2.187681957186545\n",
      "True negative: 979, True positive: 10, False negative: 2, False positive: 971\n",
      "Epoch 610: Accuracy 50.71%\n",
      "Average true similarity: -2.172387359836901, Average false similarity: -2.188001019367987\n",
      "True negative: 980, True positive: 15, False negative: 1, False positive: 966\n",
      "Epoch 611: Accuracy 50.46%\n",
      "Average true similarity: -2.171624872579002, Average false similarity: -2.1870214067278293\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 612: Accuracy 50.56%\n",
      "Average true similarity: -2.1718664627930675, Average false similarity: -2.186924566768604\n",
      "True negative: 979, True positive: 13, False negative: 2, False positive: 968\n",
      "Epoch 613: Accuracy 50.51%\n",
      "Average true similarity: -2.1721192660550503, Average false similarity: -2.187870540265032\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 614: Accuracy 50.51%\n",
      "Average true similarity: -2.172218144750255, Average false similarity: -2.1866646279306847\n",
      "True negative: 978, True positive: 13, False negative: 3, False positive: 968\n",
      "Epoch 615: Accuracy 50.41%\n",
      "Average true similarity: -2.1722405708460757, Average false similarity: -2.186608562691128\n",
      "True negative: 978, True positive: 11, False negative: 3, False positive: 970\n",
      "Epoch 616: Accuracy 50.51%\n",
      "Average true similarity: -2.172044852191643, Average false similarity: -2.187664627930685\n",
      "True negative: 978, True positive: 13, False negative: 3, False positive: 968\n",
      "Epoch 617: Accuracy 50.71%\n",
      "Average true similarity: -2.172809378185525, Average false similarity: -2.1875891946992856\n",
      "True negative: 980, True positive: 15, False negative: 1, False positive: 966\n",
      "Epoch 618: Accuracy 50.46%\n",
      "Average true similarity: -2.1725443425076443, Average false similarity: -2.188699286442407\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 619: Accuracy 50.71%\n",
      "Average true similarity: -2.1728950050968376, Average false similarity: -2.1876748216106012\n",
      "True negative: 978, True positive: 17, False negative: 3, False positive: 964\n",
      "Epoch 620: Accuracy 50.66%\n",
      "Average true similarity: -2.1727166156982674, Average false similarity: -2.188241590214066\n",
      "True negative: 979, True positive: 15, False negative: 2, False positive: 966\n",
      "Epoch 621: Accuracy 50.61%\n",
      "Average true similarity: -2.1728267074413856, Average false similarity: -2.187849133537207\n",
      "True negative: 980, True positive: 13, False negative: 1, False positive: 968\n",
      "Epoch 622: Accuracy 50.46%\n",
      "Average true similarity: -2.173050968399591, Average false similarity: -2.188508664627933\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 623: Accuracy 50.66%\n",
      "Average true similarity: -2.1729765545361848, Average false similarity: -2.187875637104993\n",
      "True negative: 980, True positive: 14, False negative: 1, False positive: 967\n",
      "Epoch 624: Accuracy 50.46%\n",
      "Average true similarity: -2.1732385321100938, Average false similarity: -2.187808358817533\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 625: Accuracy 50.25%\n",
      "Average true similarity: -2.1726676860346585, Average false similarity: -2.1879490316004073\n",
      "True negative: 978, True positive: 8, False negative: 3, False positive: 973\n",
      "Epoch 626: Accuracy 50.41%\n",
      "Average true similarity: -2.1736941896024486, Average false similarity: -2.1885820591233403\n",
      "True negative: 979, True positive: 10, False negative: 2, False positive: 971\n",
      "Epoch 627: Accuracy 50.25%\n",
      "Average true similarity: -2.173420998980633, Average false similarity: -2.187709480122324\n",
      "True negative: 977, True positive: 9, False negative: 4, False positive: 972\n",
      "Epoch 628: Accuracy 50.51%\n",
      "Average true similarity: -2.1743302752293587, Average false similarity: -2.189422018348626\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 629: Accuracy 50.46%\n",
      "Average true similarity: -2.173446483180428, Average false similarity: -2.1878114169215097\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 630: Accuracy 50.2%\n",
      "Average true similarity: -2.174170234454637, Average false similarity: -2.188871559633028\n",
      "True negative: 977, True positive: 8, False negative: 4, False positive: 973\n",
      "Epoch 631: Accuracy 50.51%\n",
      "Average true similarity: -2.1736850152905185, Average false similarity: -2.1883822629969463\n",
      "True negative: 978, True positive: 13, False negative: 3, False positive: 968\n",
      "Epoch 632: Accuracy 50.56%\n",
      "Average true similarity: -2.1734108053007124, Average false similarity: -2.1889296636085622\n",
      "True negative: 978, True positive: 14, False negative: 3, False positive: 967\n",
      "Epoch 633: Accuracy 50.36%\n",
      "Average true similarity: -2.173951070336392, Average false similarity: -2.1885351681957177\n",
      "True negative: 979, True positive: 9, False negative: 2, False positive: 972\n",
      "Epoch 634: Accuracy 50.51%\n",
      "Average true similarity: -2.172994903160038, Average false similarity: -2.187998980632009\n",
      "True negative: 980, True positive: 11, False negative: 1, False positive: 970\n",
      "Epoch 635: Accuracy 50.82%\n",
      "Average true similarity: -2.1743007135575936, Average false similarity: -2.1887584097859314\n",
      "True negative: 981, True positive: 16, False negative: 0, False positive: 965\n",
      "Epoch 636: Accuracy 50.36%\n",
      "Average true similarity: -2.174603465851172, Average false similarity: -2.189305810397554\n",
      "True negative: 977, True positive: 11, False negative: 4, False positive: 970\n",
      "Epoch 637: Accuracy 50.51%\n",
      "Average true similarity: -2.17508154943935, Average false similarity: -2.190500509683995\n",
      "True negative: 980, True positive: 11, False negative: 1, False positive: 970\n",
      "Epoch 638: Accuracy 50.2%\n",
      "Average true similarity: -2.1754322120285456, Average false similarity: -2.1911590214067274\n",
      "True negative: 980, True positive: 5, False negative: 1, False positive: 976\n",
      "Epoch 639: Accuracy 50.25%\n",
      "Average true similarity: -2.175571865443424, Average false similarity: -2.1900448521916425\n",
      "True negative: 979, True positive: 7, False negative: 2, False positive: 974\n",
      "Epoch 640: Accuracy 50.56%\n",
      "Average true similarity: -2.174051987767581, Average false similarity: -2.189464831804282\n",
      "True negative: 980, True positive: 12, False negative: 1, False positive: 969\n",
      "Epoch 641: Accuracy 50.51%\n",
      "Average true similarity: -2.1738501529051995, Average false similarity: -2.1886676860346594\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 642: Accuracy 50.36%\n",
      "Average true similarity: -2.173777777777779, Average false similarity: -2.188830784913354\n",
      "True negative: 979, True positive: 9, False negative: 2, False positive: 972\n",
      "Epoch 643: Accuracy 50.25%\n",
      "Average true similarity: -2.1736564729867456, Average false similarity: -2.188216106014274\n",
      "True negative: 978, True positive: 8, False negative: 3, False positive: 973\n",
      "Epoch 644: Accuracy 50.66%\n",
      "Average true similarity: -2.1735871559633027, Average false similarity: -2.1886788990825683\n",
      "True negative: 981, True positive: 13, False negative: 0, False positive: 968\n",
      "Epoch 645: Accuracy 50.36%\n",
      "Average true similarity: -2.1740713557594282, Average false similarity: -2.1889816513761446\n",
      "True negative: 979, True positive: 9, False negative: 2, False positive: 972\n",
      "Epoch 646: Accuracy 50.51%\n",
      "Average true similarity: -2.1746411824668725, Average false similarity: -2.189702344546382\n",
      "True negative: 980, True positive: 11, False negative: 1, False positive: 970\n",
      "Epoch 647: Accuracy 50.56%\n",
      "Average true similarity: -2.175774719673803, Average false similarity: -2.1919317023445455\n",
      "True negative: 980, True positive: 12, False negative: 1, False positive: 969\n",
      "Epoch 648: Accuracy 50.46%\n",
      "Average true similarity: -2.1757206931702333, Average false similarity: -2.1914617737003064\n",
      "True negative: 980, True positive: 10, False negative: 1, False positive: 971\n",
      "Epoch 649: Accuracy 50.36%\n",
      "Average true similarity: -2.1763465851172263, Average false similarity: -2.190344546381245\n",
      "True negative: 979, True positive: 9, False negative: 2, False positive: 972\n",
      "Epoch 650: Accuracy 50.36%\n",
      "Average true similarity: -2.176234454638121, Average false similarity: -2.190794087665648\n",
      "True negative: 979, True positive: 9, False negative: 2, False positive: 972\n",
      "Epoch 651: Accuracy 50.41%\n",
      "Average true similarity: -2.176434250764528, Average false similarity: -2.1918297655453616\n",
      "True negative: 981, True positive: 8, False negative: 0, False positive: 973\n",
      "Epoch 652: Accuracy 50.36%\n",
      "Average true similarity: -2.176539245667686, Average false similarity: -2.1910428134556588\n",
      "True negative: 981, True positive: 7, False negative: 0, False positive: 974\n",
      "Epoch 653: Accuracy 50.46%\n",
      "Average true similarity: -2.175472986748216, Average false similarity: -2.1906116207951065\n",
      "True negative: 978, True positive: 12, False negative: 3, False positive: 969\n",
      "Epoch 654: Accuracy 50.46%\n",
      "Average true similarity: -2.1758950050968395, Average false similarity: -2.191085626911314\n",
      "True negative: 980, True positive: 10, False negative: 1, False positive: 971\n",
      "Epoch 655: Accuracy 50.51%\n",
      "Average true similarity: -2.1760346585117247, Average false similarity: -2.1914260958205913\n",
      "True negative: 980, True positive: 11, False negative: 1, False positive: 970\n",
      "Epoch 656: Accuracy 50.25%\n",
      "Average true similarity: -2.177160040774718, Average false similarity: -2.191996941896024\n",
      "True negative: 979, True positive: 7, False negative: 2, False positive: 974\n",
      "Epoch 657: Accuracy 50.31%\n",
      "Average true similarity: -2.17666870540265, Average false similarity: -2.192357798165137\n",
      "True negative: 981, True positive: 6, False negative: 0, False positive: 975\n",
      "Epoch 658: Accuracy 50.46%\n",
      "Average true similarity: -2.1767696228338447, Average false similarity: -2.19184505606524\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 659: Accuracy 50.41%\n",
      "Average true similarity: -2.176745158002041, Average false similarity: -2.1922171253822618\n",
      "True negative: 979, True positive: 10, False negative: 2, False positive: 971\n",
      "Epoch 660: Accuracy 50.51%\n",
      "Average true similarity: -2.176771661569825, Average false similarity: -2.1927390417940864\n",
      "True negative: 980, True positive: 11, False negative: 1, False positive: 970\n",
      "Epoch 661: Accuracy 50.31%\n",
      "Average true similarity: -2.1766860346585126, Average false similarity: -2.1921987767584104\n",
      "True negative: 980, True positive: 7, False negative: 1, False positive: 974\n",
      "Epoch 662: Accuracy 50.15%\n",
      "Average true similarity: -2.176959225280329, Average false similarity: -2.1916982670744147\n",
      "True negative: 979, True positive: 5, False negative: 2, False positive: 976\n",
      "Epoch 663: Accuracy 50.46%\n",
      "Average true similarity: -2.176701325178391, Average false similarity: -2.191761467889909\n",
      "True negative: 980, True positive: 10, False negative: 1, False positive: 971\n",
      "Epoch 664: Accuracy 50.56%\n",
      "Average true similarity: -2.176868501529056, Average false similarity: -2.191773700305813\n",
      "True negative: 980, True positive: 12, False negative: 1, False positive: 969\n",
      "Epoch 665: Accuracy 50.46%\n",
      "Average true similarity: -2.1763058103975528, Average false similarity: -2.1909204892966376\n",
      "True negative: 981, True positive: 9, False negative: 0, False positive: 972\n",
      "Epoch 666: Accuracy 50.51%\n",
      "Average true similarity: -2.175876656472988, Average false similarity: -2.1905891946992857\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 667: Accuracy 50.56%\n",
      "Average true similarity: -2.1750917431192636, Average false similarity: -2.190588175331294\n",
      "True negative: 980, True positive: 12, False negative: 1, False positive: 969\n",
      "Epoch 668: Accuracy 50.51%\n",
      "Average true similarity: -2.174949031600408, Average false similarity: -2.1899694189602457\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 669: Accuracy 50.61%\n",
      "Average true similarity: -2.1740693170234446, Average false similarity: -2.1888603465851153\n",
      "True negative: 980, True positive: 13, False negative: 1, False positive: 968\n",
      "Epoch 670: Accuracy 50.71%\n",
      "Average true similarity: -2.1740214067278267, Average false similarity: -2.189509683995923\n",
      "True negative: 980, True positive: 15, False negative: 1, False positive: 966\n",
      "Epoch 671: Accuracy 50.25%\n",
      "Average true similarity: -2.174289500509683, Average false similarity: -2.1893955147808364\n",
      "True negative: 978, True positive: 8, False negative: 3, False positive: 973\n",
      "Epoch 672: Accuracy 50.71%\n",
      "Average true similarity: -2.175218144750256, Average false similarity: -2.1904046890927646\n",
      "True negative: 980, True positive: 15, False negative: 1, False positive: 966\n",
      "Epoch 673: Accuracy 50.56%\n",
      "Average true similarity: -2.1745208970438363, Average false similarity: -2.189240570846078\n",
      "True negative: 979, True positive: 13, False negative: 2, False positive: 968\n",
      "Epoch 674: Accuracy 50.46%\n",
      "Average true similarity: -2.1731855249745147, Average false similarity: -2.187792048929661\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 675: Accuracy 50.61%\n",
      "Average true similarity: -2.1730611620795104, Average false similarity: -2.1879255861365934\n",
      "True negative: 980, True positive: 13, False negative: 1, False positive: 968\n",
      "Epoch 676: Accuracy 50.51%\n",
      "Average true similarity: -2.1728073394495415, Average false similarity: -2.187934760448522\n",
      "True negative: 978, True positive: 13, False negative: 3, False positive: 968\n",
      "Epoch 677: Accuracy 50.51%\n",
      "Average true similarity: -2.1727257900101913, Average false similarity: -2.188047910295619\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 678: Accuracy 50.51%\n",
      "Average true similarity: -2.1732222222222206, Average false similarity: -2.187879714576963\n",
      "True negative: 978, True positive: 13, False negative: 3, False positive: 968\n",
      "Epoch 679: Accuracy 50.36%\n",
      "Average true similarity: -2.1733017329255873, Average false similarity: -2.1876401630988798\n",
      "True negative: 979, True positive: 9, False negative: 2, False positive: 972\n",
      "Epoch 680: Accuracy 50.51%\n",
      "Average true similarity: -2.1724597349643227, Average false similarity: -2.187457696228342\n",
      "True negative: 980, True positive: 11, False negative: 1, False positive: 970\n",
      "Epoch 681: Accuracy 50.56%\n",
      "Average true similarity: -2.1724159021406733, Average false similarity: -2.1873598369011202\n",
      "True negative: 979, True positive: 13, False negative: 2, False positive: 968\n",
      "Epoch 682: Accuracy 50.66%\n",
      "Average true similarity: -2.172397553516821, Average false similarity: -2.1870438328236483\n",
      "True negative: 979, True positive: 15, False negative: 2, False positive: 966\n",
      "Epoch 683: Accuracy 50.66%\n",
      "Average true similarity: -2.172281345565748, Average false similarity: -2.1869694189602473\n",
      "True negative: 980, True positive: 14, False negative: 1, False positive: 967\n",
      "Epoch 684: Accuracy 50.66%\n",
      "Average true similarity: -2.17199592252803, Average false similarity: -2.187215086646281\n",
      "True negative: 980, True positive: 14, False negative: 1, False positive: 967\n",
      "Epoch 685: Accuracy 50.46%\n",
      "Average true similarity: -2.1714373088685033, Average false similarity: -2.1867482161060128\n",
      "True negative: 977, True positive: 13, False negative: 4, False positive: 968\n",
      "Epoch 686: Accuracy 50.76%\n",
      "Average true similarity: -2.1716839959225296, Average false similarity: -2.1860407747196735\n",
      "True negative: 979, True positive: 17, False negative: 2, False positive: 964\n",
      "Epoch 687: Accuracy 50.76%\n",
      "Average true similarity: -2.171082568807337, Average false similarity: -2.1851804281345544\n",
      "True negative: 979, True positive: 17, False negative: 2, False positive: 964\n",
      "Epoch 688: Accuracy 50.82%\n",
      "Average true similarity: -2.171522935779813, Average false similarity: -2.186385321100917\n",
      "True negative: 981, True positive: 16, False negative: 0, False positive: 965\n",
      "Epoch 689: Accuracy 50.71%\n",
      "Average true similarity: -2.171222222222224, Average false similarity: -2.186396534148832\n",
      "True negative: 981, True positive: 14, False negative: 0, False positive: 967\n",
      "Epoch 690: Accuracy 50.66%\n",
      "Average true similarity: -2.170987767584095, Average false similarity: -2.1861641182466873\n",
      "True negative: 978, True positive: 16, False negative: 3, False positive: 965\n",
      "Epoch 691: Accuracy 50.76%\n",
      "Average true similarity: -2.1713700305810386, Average false similarity: -2.186120285423039\n",
      "True negative: 980, True positive: 16, False negative: 1, False positive: 965\n",
      "Epoch 692: Accuracy 50.61%\n",
      "Average true similarity: -2.1714169215086616, Average false similarity: -2.186309887869522\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 693: Accuracy 50.76%\n",
      "Average true similarity: -2.171659531090724, Average false similarity: -2.1861457696228315\n",
      "True negative: 979, True positive: 17, False negative: 2, False positive: 964\n",
      "Epoch 694: Accuracy 50.82%\n",
      "Average true similarity: -2.171912334352701, Average false similarity: -2.1868226299694187\n",
      "True negative: 980, True positive: 17, False negative: 1, False positive: 964\n",
      "Epoch 695: Accuracy 50.51%\n",
      "Average true similarity: -2.1713078491335396, Average false similarity: -2.1860122324158984\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 696: Accuracy 50.61%\n",
      "Average true similarity: -2.1710642201834904, Average false similarity: -2.186509683995921\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 697: Accuracy 50.56%\n",
      "Average true similarity: -2.171953109072376, Average false similarity: -2.187715596330276\n",
      "True negative: 980, True positive: 12, False negative: 1, False positive: 969\n",
      "Epoch 698: Accuracy 50.61%\n",
      "Average true similarity: -2.172126401630987, Average false similarity: -2.187764525993885\n",
      "True negative: 980, True positive: 13, False negative: 1, False positive: 968\n",
      "Epoch 699: Accuracy 50.66%\n",
      "Average true similarity: -2.1721243628950053, Average false similarity: -2.187815494393476\n",
      "True negative: 979, True positive: 15, False negative: 2, False positive: 966\n",
      "Epoch 700: Accuracy 50.71%\n",
      "Average true similarity: -2.1719092762487238, Average false similarity: -2.187598369011212\n",
      "True negative: 978, True positive: 17, False negative: 3, False positive: 964\n",
      "Epoch 701: Accuracy 50.61%\n",
      "Average true similarity: -2.172323139653413, Average false similarity: -2.187839959225281\n",
      "True negative: 980, True positive: 13, False negative: 1, False positive: 968\n",
      "Epoch 702: Accuracy 50.46%\n",
      "Average true similarity: -2.172668705402648, Average false similarity: -2.188085626911318\n",
      "True negative: 978, True positive: 12, False negative: 3, False positive: 969\n",
      "Epoch 703: Accuracy 50.51%\n",
      "Average true similarity: -2.1713914373088716, Average false similarity: -2.1865168195718665\n",
      "True negative: 978, True positive: 13, False negative: 3, False positive: 968\n",
      "Epoch 704: Accuracy 50.87%\n",
      "Average true similarity: -2.1713567787971466, Average false similarity: -2.186015290519874\n",
      "True negative: 980, True positive: 18, False negative: 1, False positive: 963\n",
      "Epoch 705: Accuracy 50.76%\n",
      "Average true similarity: -2.1712833843017307, Average false similarity: -2.1864138634046912\n",
      "True negative: 978, True positive: 18, False negative: 3, False positive: 963\n",
      "Epoch 706: Accuracy 50.61%\n",
      "Average true similarity: -2.1706320081549437, Average false similarity: -2.184987767584099\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 707: Accuracy 50.61%\n",
      "Average true similarity: -2.1707237512742106, Average false similarity: -2.1863017329255867\n",
      "True negative: 977, True positive: 16, False negative: 4, False positive: 965\n",
      "Epoch 708: Accuracy 50.56%\n",
      "Average true similarity: -2.1712670744138647, Average false similarity: -2.186679918450561\n",
      "True negative: 978, True positive: 14, False negative: 3, False positive: 967\n",
      "Epoch 709: Accuracy 50.61%\n",
      "Average true similarity: -2.170546381243628, Average false similarity: -2.1863792048929676\n",
      "True negative: 977, True positive: 16, False negative: 4, False positive: 965\n",
      "Epoch 710: Accuracy 50.61%\n",
      "Average true similarity: -2.1703832823649334, Average false similarity: -2.185355759429153\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 711: Accuracy 50.97%\n",
      "Average true similarity: -2.169176350662588, Average false similarity: -2.1851580020387344\n",
      "True negative: 978, True positive: 22, False negative: 3, False positive: 959\n",
      "Epoch 712: Accuracy 50.61%\n",
      "Average true similarity: -2.1694393476044858, Average false similarity: -2.184668705402651\n",
      "True negative: 976, True positive: 17, False negative: 5, False positive: 964\n",
      "Epoch 713: Accuracy 50.51%\n",
      "Average true similarity: -2.170043832823649, Average false similarity: -2.1850468909276284\n",
      "True negative: 978, True positive: 13, False negative: 3, False positive: 968\n",
      "Epoch 714: Accuracy 50.82%\n",
      "Average true similarity: -2.168913353720692, Average false similarity: -2.1835912334352683\n",
      "True negative: 978, True positive: 19, False negative: 3, False positive: 962\n",
      "Epoch 715: Accuracy 50.82%\n",
      "Average true similarity: -2.168680937818553, Average false similarity: -2.1833761467889903\n",
      "True negative: 977, True positive: 20, False negative: 4, False positive: 961\n",
      "Epoch 716: Accuracy 50.46%\n",
      "Average true similarity: -2.1689755351681943, Average false similarity: -2.18416004077472\n",
      "True negative: 975, True positive: 15, False negative: 6, False positive: 966\n",
      "Epoch 717: Accuracy 50.76%\n",
      "Average true similarity: -2.1693272171253803, Average false similarity: -2.183309887869518\n",
      "True negative: 978, True positive: 18, False negative: 3, False positive: 963\n",
      "Epoch 718: Accuracy 50.82%\n",
      "Average true similarity: -2.1690061162079535, Average false similarity: -2.1837981651376164\n",
      "True negative: 980, True positive: 17, False negative: 1, False positive: 964\n",
      "Epoch 719: Accuracy 50.76%\n",
      "Average true similarity: -2.1689072375127414, Average false similarity: -2.1838827726809384\n",
      "True negative: 979, True positive: 17, False negative: 2, False positive: 964\n",
      "Epoch 720: Accuracy 50.71%\n",
      "Average true similarity: -2.1694199796126403, Average false similarity: -2.184796126401631\n",
      "True negative: 978, True positive: 17, False negative: 3, False positive: 964\n",
      "Epoch 721: Accuracy 50.56%\n",
      "Average true similarity: -2.1688277268093774, Average false similarity: -2.1837828746177346\n",
      "True negative: 977, True positive: 15, False negative: 4, False positive: 966\n",
      "Epoch 722: Accuracy 50.66%\n",
      "Average true similarity: -2.1688797145769585, Average false similarity: -2.1834403669724782\n",
      "True negative: 978, True positive: 16, False negative: 3, False positive: 965\n",
      "Epoch 723: Accuracy 50.92%\n",
      "Average true similarity: -2.1695045871559637, Average false similarity: -2.1840948012232424\n",
      "True negative: 977, True positive: 22, False negative: 4, False positive: 959\n",
      "Epoch 724: Accuracy 50.71%\n",
      "Average true similarity: -2.170622833843018, Average false similarity: -2.185005096839958\n",
      "True negative: 978, True positive: 17, False negative: 3, False positive: 964\n",
      "Epoch 725: Accuracy 50.76%\n",
      "Average true similarity: -2.170036697247704, Average false similarity: -2.1849439347604505\n",
      "True negative: 978, True positive: 18, False negative: 3, False positive: 963\n",
      "Epoch 726: Accuracy 50.97%\n",
      "Average true similarity: -2.169895005096837, Average false similarity: -2.1853374108053\n",
      "True negative: 981, True positive: 19, False negative: 0, False positive: 962\n",
      "Epoch 727: Accuracy 50.76%\n",
      "Average true similarity: -2.1693537206931706, Average false similarity: -2.1848562691131477\n",
      "True negative: 979, True positive: 17, False negative: 2, False positive: 964\n",
      "Epoch 728: Accuracy 50.97%\n",
      "Average true similarity: -2.1692110091743158, Average false similarity: -2.1848267074413883\n",
      "True negative: 981, True positive: 19, False negative: 0, False positive: 962\n",
      "Epoch 729: Accuracy 50.61%\n",
      "Average true similarity: -2.1691702344546413, Average false similarity: -2.1849317023445454\n",
      "True negative: 976, True positive: 17, False negative: 5, False positive: 964\n",
      "Epoch 730: Accuracy 50.92%\n",
      "Average true similarity: -2.16944342507645, Average false similarity: -2.1844505606523943\n",
      "True negative: 979, True positive: 20, False negative: 2, False positive: 961\n",
      "Epoch 731: Accuracy 50.66%\n",
      "Average true similarity: -2.169371049949032, Average false similarity: -2.1848552497451577\n",
      "True negative: 978, True positive: 16, False negative: 3, False positive: 965\n",
      "Epoch 732: Accuracy 50.56%\n",
      "Average true similarity: -2.168952089704383, Average false similarity: -2.184006116207952\n",
      "True negative: 977, True positive: 15, False negative: 4, False positive: 966\n",
      "Epoch 733: Accuracy 50.82%\n",
      "Average true similarity: -2.169913353720694, Average false similarity: -2.184501529051986\n",
      "True negative: 978, True positive: 19, False negative: 3, False positive: 962\n",
      "Epoch 734: Accuracy 50.71%\n",
      "Average true similarity: -2.1696024464831805, Average false similarity: -2.1849989806320087\n",
      "True negative: 979, True positive: 16, False negative: 2, False positive: 965\n",
      "Epoch 735: Accuracy 50.76%\n",
      "Average true similarity: -2.1691080530071334, Average false similarity: -2.184188583078493\n",
      "True negative: 977, True positive: 19, False negative: 4, False positive: 962\n",
      "Epoch 736: Accuracy 50.66%\n",
      "Average true similarity: -2.1693588175331313, Average false similarity: -2.184339449541284\n",
      "True negative: 977, True positive: 17, False negative: 4, False positive: 964\n",
      "Epoch 737: Accuracy 50.66%\n",
      "Average true similarity: -2.1698572884811393, Average false similarity: -2.184950050968396\n",
      "True negative: 977, True positive: 17, False negative: 4, False positive: 964\n",
      "Epoch 738: Accuracy 50.56%\n",
      "Average true similarity: -2.169696228338429, Average false similarity: -2.1843353720693184\n",
      "True negative: 978, True positive: 14, False negative: 3, False positive: 967\n",
      "Epoch 739: Accuracy 50.61%\n",
      "Average true similarity: -2.1701060142711546, Average false similarity: -2.1853934760448546\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 740: Accuracy 50.61%\n",
      "Average true similarity: -2.169621814475024, Average false similarity: -2.185051987767586\n",
      "True negative: 980, True positive: 13, False negative: 1, False positive: 968\n",
      "Epoch 741: Accuracy 50.76%\n",
      "Average true similarity: -2.1695290519877672, Average false similarity: -2.1846075433231364\n",
      "True negative: 980, True positive: 16, False negative: 1, False positive: 965\n",
      "Epoch 742: Accuracy 50.76%\n",
      "Average true similarity: -2.1695178389398553, Average false similarity: -2.1841855249745183\n",
      "True negative: 977, True positive: 19, False negative: 4, False positive: 962\n",
      "Epoch 743: Accuracy 50.61%\n",
      "Average true similarity: -2.1693710499490337, Average false similarity: -2.184951070336391\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 744: Accuracy 50.71%\n",
      "Average true similarity: -2.1690621814475004, Average false similarity: -2.184382262996942\n",
      "True negative: 978, True positive: 17, False negative: 3, False positive: 964\n",
      "Epoch 745: Accuracy 50.61%\n",
      "Average true similarity: -2.1699001019367965, Average false similarity: -2.184545361875637\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 746: Accuracy 50.76%\n",
      "Average true similarity: -2.169630988786951, Average false similarity: -2.1846116207951067\n",
      "True negative: 977, True positive: 19, False negative: 4, False positive: 962\n",
      "Epoch 747: Accuracy 50.87%\n",
      "Average true similarity: -2.169299694189603, Average false similarity: -2.1839011213047907\n",
      "True negative: 977, True positive: 21, False negative: 4, False positive: 960\n",
      "Epoch 748: Accuracy 50.97%\n",
      "Average true similarity: -2.1690784913353722, Average false similarity: -2.184876656472986\n",
      "True negative: 979, True positive: 21, False negative: 2, False positive: 960\n",
      "Epoch 749: Accuracy 50.76%\n",
      "Average true similarity: -2.170044852191641, Average false similarity: -2.183594291539246\n",
      "True negative: 978, True positive: 18, False negative: 3, False positive: 963\n",
      "Epoch 750: Accuracy 50.97%\n",
      "Average true similarity: -2.169528032619775, Average false similarity: -2.1838154943934787\n",
      "True negative: 977, True positive: 23, False negative: 4, False positive: 958\n",
      "Epoch 751: Accuracy 50.97%\n",
      "Average true similarity: -2.1692354740061166, Average false similarity: -2.184528032619774\n",
      "True negative: 979, True positive: 21, False negative: 2, False positive: 960\n",
      "Epoch 752: Accuracy 50.87%\n",
      "Average true similarity: -2.1699062181447513, Average false similarity: -2.184499490316003\n",
      "True negative: 978, True positive: 20, False negative: 3, False positive: 961\n",
      "Epoch 753: Accuracy 50.87%\n",
      "Average true similarity: -2.17006625891947, Average false similarity: -2.185085626911316\n",
      "True negative: 979, True positive: 19, False negative: 2, False positive: 962\n",
      "Epoch 754: Accuracy 50.97%\n",
      "Average true similarity: -2.169597349643222, Average false similarity: -2.184138634046889\n",
      "True negative: 978, True positive: 22, False negative: 3, False positive: 959\n",
      "Epoch 755: Accuracy 50.82%\n",
      "Average true similarity: -2.169438328236493, Average false similarity: -2.183864424057086\n",
      "True negative: 978, True positive: 19, False negative: 3, False positive: 962\n",
      "Epoch 756: Accuracy 50.31%\n",
      "Average true similarity: -2.1698134556574913, Average false similarity: -2.1843873598369\n",
      "True negative: 976, True positive: 11, False negative: 5, False positive: 970\n",
      "Epoch 757: Accuracy 50.97%\n",
      "Average true similarity: -2.169624872579002, Average false similarity: -2.18464525993884\n",
      "True negative: 978, True positive: 22, False negative: 3, False positive: 959\n",
      "Epoch 758: Accuracy 50.82%\n",
      "Average true similarity: -2.1706850152905215, Average false similarity: -2.1841997961264012\n",
      "True negative: 979, True positive: 18, False negative: 2, False positive: 963\n",
      "Epoch 759: Accuracy 50.71%\n",
      "Average true similarity: -2.1710764525993906, Average false similarity: -2.1857767584097854\n",
      "True negative: 977, True positive: 18, False negative: 4, False positive: 963\n",
      "Epoch 760: Accuracy 50.66%\n",
      "Average true similarity: -2.1715535168195736, Average false similarity: -2.186883792048932\n",
      "True negative: 980, True positive: 14, False negative: 1, False positive: 967\n",
      "Epoch 761: Accuracy 50.71%\n",
      "Average true similarity: -2.1714077471967355, Average false similarity: -2.186261977573905\n",
      "True negative: 978, True positive: 17, False negative: 3, False positive: 964\n",
      "Epoch 762: Accuracy 50.61%\n",
      "Average true similarity: -2.1717461773700304, Average false similarity: -2.1864556574923535\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 763: Accuracy 50.76%\n",
      "Average true similarity: -2.1719429153924557, Average false similarity: -2.1867278287461795\n",
      "True negative: 979, True positive: 17, False negative: 2, False positive: 964\n",
      "Epoch 764: Accuracy 50.56%\n",
      "Average true similarity: -2.171309887869521, Average false similarity: -2.1860795107033635\n",
      "True negative: 979, True positive: 13, False negative: 2, False positive: 968\n",
      "Epoch 765: Accuracy 50.56%\n",
      "Average true similarity: -2.172030581039758, Average false similarity: -2.1865300713557607\n",
      "True negative: 977, True positive: 15, False negative: 4, False positive: 966\n",
      "Epoch 766: Accuracy 50.76%\n",
      "Average true similarity: -2.1724923547400623, Average false similarity: -2.1872181447502537\n",
      "True negative: 979, True positive: 17, False negative: 2, False positive: 964\n",
      "Epoch 767: Accuracy 50.87%\n",
      "Average true similarity: -2.1724128440366974, Average false similarity: -2.186888888888888\n",
      "True negative: 980, True positive: 18, False negative: 1, False positive: 963\n",
      "Epoch 768: Accuracy 50.51%\n",
      "Average true similarity: -2.172621814475026, Average false similarity: -2.1877920489296643\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 769: Accuracy 50.46%\n",
      "Average true similarity: -2.1713190621814498, Average false similarity: -2.1857787971457685\n",
      "True negative: 977, True positive: 13, False negative: 4, False positive: 968\n",
      "Epoch 770: Accuracy 50.61%\n",
      "Average true similarity: -2.1706095820591207, Average false similarity: -2.1867217125382257\n",
      "True negative: 976, True positive: 17, False negative: 5, False positive: 964\n",
      "Epoch 771: Accuracy 50.56%\n",
      "Average true similarity: -2.170914373088685, Average false similarity: -2.185590214067278\n",
      "True negative: 979, True positive: 13, False negative: 2, False positive: 968\n",
      "Epoch 772: Accuracy 50.92%\n",
      "Average true similarity: -2.169624872579003, Average false similarity: -2.184099898063198\n",
      "True negative: 978, True positive: 21, False negative: 3, False positive: 960\n",
      "Epoch 773: Accuracy 50.41%\n",
      "Average true similarity: -2.170062181447505, Average false similarity: -2.184276248725792\n",
      "True negative: 978, True positive: 11, False negative: 3, False positive: 970\n",
      "Epoch 774: Accuracy 50.76%\n",
      "Average true similarity: -2.1695392456676847, Average false similarity: -2.184468909276248\n",
      "True negative: 979, True positive: 17, False negative: 2, False positive: 964\n",
      "Epoch 775: Accuracy 50.97%\n",
      "Average true similarity: -2.168444444444446, Average false similarity: -2.183256880733945\n",
      "True negative: 978, True positive: 22, False negative: 3, False positive: 959\n",
      "Epoch 776: Accuracy 51.12%\n",
      "Average true similarity: -2.168015290519874, Average false similarity: -2.1831804281345577\n",
      "True negative: 979, True positive: 24, False negative: 2, False positive: 957\n",
      "Epoch 777: Accuracy 50.66%\n",
      "Average true similarity: -2.168300713557594, Average false similarity: -2.1829847094801225\n",
      "True negative: 979, True positive: 15, False negative: 2, False positive: 966\n",
      "Epoch 778: Accuracy 50.71%\n",
      "Average true similarity: -2.1683374108053024, Average false similarity: -2.183049949031599\n",
      "True negative: 978, True positive: 17, False negative: 3, False positive: 964\n",
      "Epoch 779: Accuracy 50.82%\n",
      "Average true similarity: -2.1683058103975554, Average false similarity: -2.18307033639144\n",
      "True negative: 978, True positive: 19, False negative: 3, False positive: 962\n",
      "Epoch 780: Accuracy 50.61%\n",
      "Average true similarity: -2.1689113149847095, Average false similarity: -2.1843985728848097\n",
      "True negative: 976, True positive: 17, False negative: 5, False positive: 964\n",
      "Epoch 781: Accuracy 50.76%\n",
      "Average true similarity: -2.169796126401633, Average false similarity: -2.184818552497453\n",
      "True negative: 976, True positive: 20, False negative: 5, False positive: 961\n",
      "Epoch 782: Accuracy 50.76%\n",
      "Average true similarity: -2.169219164118247, Average false similarity: -2.1834169215086634\n",
      "True negative: 977, True positive: 19, False negative: 4, False positive: 962\n",
      "Epoch 783: Accuracy 50.97%\n",
      "Average true similarity: -2.169672782874618, Average false similarity: -2.184299694189601\n",
      "True negative: 980, True positive: 20, False negative: 1, False positive: 961\n",
      "Epoch 784: Accuracy 50.87%\n",
      "Average true similarity: -2.168467889908256, Average false similarity: -2.1838552497451587\n",
      "True negative: 976, True positive: 22, False negative: 5, False positive: 959\n",
      "Epoch 785: Accuracy 50.76%\n",
      "Average true similarity: -2.168687054026503, Average false similarity: -2.1828878695208984\n",
      "True negative: 977, True positive: 19, False negative: 4, False positive: 962\n",
      "Epoch 786: Accuracy 50.82%\n",
      "Average true similarity: -2.1684658511722743, Average false similarity: -2.183536187563711\n",
      "True negative: 977, True positive: 20, False negative: 4, False positive: 961\n",
      "Epoch 787: Accuracy 50.82%\n",
      "Average true similarity: -2.1670621814475, Average false similarity: -2.182948012232413\n",
      "True negative: 976, True positive: 21, False negative: 5, False positive: 960\n",
      "Epoch 788: Accuracy 51.07%\n",
      "Average true similarity: -2.16750254841998, Average false similarity: -2.18261977573904\n",
      "True negative: 976, True positive: 26, False negative: 5, False positive: 955\n",
      "Epoch 789: Accuracy 50.92%\n",
      "Average true similarity: -2.1672191641182477, Average false similarity: -2.182571865443427\n",
      "True negative: 978, True positive: 21, False negative: 3, False positive: 960\n",
      "Epoch 790: Accuracy 50.92%\n",
      "Average true similarity: -2.16725790010194, Average false similarity: -2.1822028542303786\n",
      "True negative: 977, True positive: 22, False negative: 4, False positive: 959\n",
      "Epoch 791: Accuracy 51.02%\n",
      "Average true similarity: -2.1678593272171245, Average false similarity: -2.1826085626911316\n",
      "True negative: 977, True positive: 24, False negative: 4, False positive: 957\n",
      "Epoch 792: Accuracy 50.97%\n",
      "Average true similarity: -2.167296636085624, Average false similarity: -2.181801223241587\n",
      "True negative: 978, True positive: 22, False negative: 3, False positive: 959\n",
      "Epoch 793: Accuracy 50.82%\n",
      "Average true similarity: -2.1671478083588145, Average false similarity: -2.1822466870540245\n",
      "True negative: 978, True positive: 19, False negative: 3, False positive: 962\n",
      "Epoch 794: Accuracy 50.66%\n",
      "Average true similarity: -2.167742099898064, Average false similarity: -2.183138634046893\n",
      "True negative: 976, True positive: 18, False negative: 5, False positive: 963\n",
      "Epoch 795: Accuracy 51.12%\n",
      "Average true similarity: -2.1670835881753328, Average false similarity: -2.181865443425078\n",
      "True negative: 978, True positive: 25, False negative: 3, False positive: 956\n",
      "Epoch 796: Accuracy 50.87%\n",
      "Average true similarity: -2.1670112130479064, Average false similarity: -2.1820214067278267\n",
      "True negative: 977, True positive: 21, False negative: 4, False positive: 960\n",
      "Epoch 797: Accuracy 50.76%\n",
      "Average true similarity: -2.1663710499490296, Average false similarity: -2.1818338430173316\n",
      "True negative: 976, True positive: 20, False negative: 5, False positive: 961\n",
      "Epoch 798: Accuracy 50.87%\n",
      "Average true similarity: -2.1657268093781883, Average false similarity: -2.180403669724773\n",
      "True negative: 975, True positive: 23, False negative: 6, False positive: 958\n",
      "Epoch 799: Accuracy 50.76%\n",
      "Average true similarity: -2.166538226299692, Average false similarity: -2.1810560652395523\n",
      "True negative: 976, True positive: 20, False negative: 5, False positive: 961\n",
      "Epoch 800: Accuracy 50.92%\n",
      "Average true similarity: -2.16664525993884, Average false similarity: -2.1816717635066265\n",
      "True negative: 978, True positive: 21, False negative: 3, False positive: 960\n",
      "Epoch 801: Accuracy 50.92%\n",
      "Average true similarity: -2.164560652395517, Average false similarity: -2.179593272171251\n",
      "True negative: 978, True positive: 21, False negative: 3, False positive: 960\n",
      "Epoch 802: Accuracy 50.82%\n",
      "Average true similarity: -2.165338430173291, Average false similarity: -2.1797624872579\n",
      "True negative: 978, True positive: 19, False negative: 3, False positive: 962\n",
      "Epoch 803: Accuracy 50.87%\n",
      "Average true similarity: -2.165186544342508, Average false similarity: -2.180058103975536\n",
      "True negative: 978, True positive: 20, False negative: 3, False positive: 961\n",
      "Epoch 804: Accuracy 50.92%\n",
      "Average true similarity: -2.165307849133539, Average false similarity: -2.1806167176350635\n",
      "True negative: 979, True positive: 20, False negative: 2, False positive: 961\n",
      "Epoch 805: Accuracy 50.76%\n",
      "Average true similarity: -2.1648358817533144, Average false similarity: -2.1797074413863413\n",
      "True negative: 975, True positive: 21, False negative: 6, False positive: 960\n",
      "Epoch 806: Accuracy 50.82%\n",
      "Average true similarity: -2.164968399592253, Average false similarity: -2.1802283384301755\n",
      "True negative: 977, True positive: 20, False negative: 4, False positive: 961\n",
      "Epoch 807: Accuracy 51.07%\n",
      "Average true similarity: -2.165202854230375, Average false similarity: -2.179749235474006\n",
      "True negative: 977, True positive: 25, False negative: 4, False positive: 956\n",
      "Epoch 808: Accuracy 50.82%\n",
      "Average true similarity: -2.165687054026504, Average false similarity: -2.1807227319062177\n",
      "True negative: 978, True positive: 19, False negative: 3, False positive: 962\n",
      "Epoch 809: Accuracy 50.71%\n",
      "Average true similarity: -2.16624260958206, Average false similarity: -2.180965341488278\n",
      "True negative: 975, True positive: 20, False negative: 6, False positive: 961\n",
      "Epoch 810: Accuracy 50.66%\n",
      "Average true similarity: -2.1663914373088673, Average false similarity: -2.1809174311926594\n",
      "True negative: 977, True positive: 17, False negative: 4, False positive: 964\n",
      "Epoch 811: Accuracy 50.87%\n",
      "Average true similarity: -2.1650682976554543, Average false similarity: -2.1806738022426098\n",
      "True negative: 978, True positive: 20, False negative: 3, False positive: 961\n",
      "Epoch 812: Accuracy 50.76%\n",
      "Average true similarity: -2.165634046890931, Average false similarity: -2.1809174311926625\n",
      "True negative: 977, True positive: 19, False negative: 4, False positive: 962\n",
      "Epoch 813: Accuracy 51.02%\n",
      "Average true similarity: -2.166275229357801, Average false similarity: -2.181678899082568\n",
      "True negative: 978, True positive: 23, False negative: 3, False positive: 958\n",
      "Epoch 814: Accuracy 51.07%\n",
      "Average true similarity: -2.165543323139651, Average false similarity: -2.181162079510702\n",
      "True negative: 978, True positive: 24, False negative: 3, False positive: 957\n",
      "Epoch 815: Accuracy 51.22%\n",
      "Average true similarity: -2.166418960244649, Average false similarity: -2.1806992864424077\n",
      "True negative: 979, True positive: 26, False negative: 2, False positive: 955\n",
      "Epoch 816: Accuracy 50.71%\n",
      "Average true similarity: -2.1666106014271134, Average false similarity: -2.181389398572884\n",
      "True negative: 979, True positive: 16, False negative: 2, False positive: 965\n",
      "Epoch 817: Accuracy 50.56%\n",
      "Average true similarity: -2.165974515800205, Average false similarity: -2.181328236493373\n",
      "True negative: 975, True positive: 17, False negative: 6, False positive: 964\n",
      "Epoch 818: Accuracy 50.82%\n",
      "Average true similarity: -2.1660366972477068, Average false similarity: -2.1816187563710474\n",
      "True negative: 978, True positive: 19, False negative: 3, False positive: 962\n",
      "Epoch 819: Accuracy 51.02%\n",
      "Average true similarity: -2.1659673802242643, Average false similarity: -2.1806941896024457\n",
      "True negative: 975, True positive: 26, False negative: 6, False positive: 955\n",
      "Epoch 820: Accuracy 50.76%\n",
      "Average true similarity: -2.1655280326197746, Average false similarity: -2.1811559633027535\n",
      "True negative: 975, True positive: 21, False negative: 6, False positive: 960\n",
      "Epoch 821: Accuracy 50.87%\n",
      "Average true similarity: -2.1653058103975518, Average false similarity: -2.1801926605504574\n",
      "True negative: 979, True positive: 19, False negative: 2, False positive: 962\n",
      "Epoch 822: Accuracy 50.56%\n",
      "Average true similarity: -2.165226299694188, Average false similarity: -2.180422018348622\n",
      "True negative: 975, True positive: 17, False negative: 6, False positive: 964\n",
      "Epoch 823: Accuracy 51.02%\n",
      "Average true similarity: -2.1648746177370026, Average false similarity: -2.1795392456676845\n",
      "True negative: 979, True positive: 22, False negative: 2, False positive: 959\n",
      "Epoch 824: Accuracy 50.56%\n",
      "Average true similarity: -2.1649592252803274, Average false similarity: -2.1803017329255856\n",
      "True negative: 979, True positive: 13, False negative: 2, False positive: 968\n",
      "Epoch 825: Accuracy 50.66%\n",
      "Average true similarity: -2.1649959225280324, Average false similarity: -2.180179408766564\n",
      "True negative: 978, True positive: 16, False negative: 3, False positive: 965\n",
      "Epoch 826: Accuracy 50.87%\n",
      "Average true similarity: -2.164945973496432, Average false similarity: -2.1804057084607518\n",
      "True negative: 980, True positive: 18, False negative: 1, False positive: 963\n",
      "Epoch 827: Accuracy 50.87%\n",
      "Average true similarity: -2.164199796126402, Average false similarity: -2.179449541284406\n",
      "True negative: 978, True positive: 20, False negative: 3, False positive: 961\n",
      "Epoch 828: Accuracy 50.56%\n",
      "Average true similarity: -2.1648603465851184, Average false similarity: -2.1799847094801224\n",
      "True negative: 977, True positive: 15, False negative: 4, False positive: 966\n",
      "Epoch 829: Accuracy 50.87%\n",
      "Average true similarity: -2.164982670744135, Average false similarity: -2.1800122324159026\n",
      "True negative: 978, True positive: 20, False negative: 3, False positive: 961\n",
      "Epoch 830: Accuracy 51.17%\n",
      "Average true similarity: -2.1647747196738063, Average false similarity: -2.180083588175332\n",
      "True negative: 979, True positive: 25, False negative: 2, False positive: 956\n",
      "Epoch 831: Accuracy 51.02%\n",
      "Average true similarity: -2.165174311926608, Average false similarity: -2.1800927624872624\n",
      "True negative: 978, True positive: 23, False negative: 3, False positive: 958\n",
      "Epoch 832: Accuracy 50.87%\n",
      "Average true similarity: -2.164963302752295, Average false similarity: -2.1810071355759417\n",
      "True negative: 978, True positive: 20, False negative: 3, False positive: 961\n",
      "Epoch 833: Accuracy 50.76%\n",
      "Average true similarity: -2.1652426095820627, Average false similarity: -2.180420998980633\n",
      "True negative: 978, True positive: 18, False negative: 3, False positive: 963\n",
      "Epoch 834: Accuracy 50.87%\n",
      "Average true similarity: -2.1663353720693164, Average false similarity: -2.18154128440367\n",
      "True negative: 978, True positive: 20, False negative: 3, False positive: 961\n",
      "Epoch 835: Accuracy 50.92%\n",
      "Average true similarity: -2.165764525993886, Average false similarity: -2.1811386340468926\n",
      "True negative: 980, True positive: 19, False negative: 1, False positive: 962\n",
      "Epoch 836: Accuracy 50.82%\n",
      "Average true similarity: -2.1657155963302754, Average false similarity: -2.1815239551478074\n",
      "True negative: 978, True positive: 19, False negative: 3, False positive: 962\n",
      "Epoch 837: Accuracy 50.51%\n",
      "Average true similarity: -2.166114169215088, Average false similarity: -2.1802579001019318\n",
      "True negative: 976, True positive: 15, False negative: 5, False positive: 966\n",
      "Epoch 838: Accuracy 50.76%\n",
      "Average true similarity: -2.1669398572884817, Average false similarity: -2.1816962283384265\n",
      "True negative: 978, True positive: 18, False negative: 3, False positive: 963\n",
      "Epoch 839: Accuracy 50.87%\n",
      "Average true similarity: -2.1662803261977555, Average false similarity: -2.182101936799185\n",
      "True negative: 978, True positive: 20, False negative: 3, False positive: 961\n",
      "Epoch 840: Accuracy 50.76%\n",
      "Average true similarity: -2.1666860346585137, Average false similarity: -2.1820509683995923\n",
      "True negative: 977, True positive: 19, False negative: 4, False positive: 962\n",
      "Epoch 841: Accuracy 50.97%\n",
      "Average true similarity: -2.16667787971458, Average false similarity: -2.182518858307849\n",
      "True negative: 979, True positive: 21, False negative: 2, False positive: 960\n",
      "Epoch 842: Accuracy 51.02%\n",
      "Average true similarity: -2.1665392456676846, Average false similarity: -2.1810866462793066\n",
      "True negative: 978, True positive: 23, False negative: 3, False positive: 958\n",
      "Epoch 843: Accuracy 50.87%\n",
      "Average true similarity: -2.167022426095822, Average false similarity: -2.1816829765545345\n",
      "True negative: 980, True positive: 18, False negative: 1, False positive: 963\n",
      "Epoch 844: Accuracy 50.71%\n",
      "Average true similarity: -2.1665412844036713, Average false similarity: -2.1820081549439396\n",
      "True negative: 978, True positive: 17, False negative: 3, False positive: 964\n",
      "Epoch 845: Accuracy 50.92%\n",
      "Average true similarity: -2.166699286442403, Average false similarity: -2.181603465851169\n",
      "True negative: 979, True positive: 20, False negative: 2, False positive: 961\n",
      "Epoch 846: Accuracy 50.97%\n",
      "Average true similarity: -2.1669949031600386, Average false similarity: -2.181957186544344\n",
      "True negative: 977, True positive: 23, False negative: 4, False positive: 958\n",
      "Epoch 847: Accuracy 50.41%\n",
      "Average true similarity: -2.1670774719673815, Average false similarity: -2.181850152905199\n",
      "True negative: 976, True positive: 13, False negative: 5, False positive: 968\n",
      "Epoch 848: Accuracy 50.92%\n",
      "Average true similarity: -2.166587155963298, Average false similarity: -2.1813129459734952\n",
      "True negative: 978, True positive: 21, False negative: 3, False positive: 960\n",
      "Epoch 849: Accuracy 50.61%\n",
      "Average true similarity: -2.166930682976554, Average false similarity: -2.1816055045871585\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 850: Accuracy 50.76%\n",
      "Average true similarity: -2.166779816513763, Average false similarity: -2.1814566768603463\n",
      "True negative: 978, True positive: 18, False negative: 3, False positive: 963\n",
      "Epoch 851: Accuracy 50.87%\n",
      "Average true similarity: -2.1672538226299687, Average false similarity: -2.181669724770643\n",
      "True negative: 978, True positive: 20, False negative: 3, False positive: 961\n",
      "Epoch 852: Accuracy 50.71%\n",
      "Average true similarity: -2.1677951070336414, Average false similarity: -2.1824250764526\n",
      "True negative: 979, True positive: 16, False negative: 2, False positive: 965\n",
      "Epoch 853: Accuracy 50.61%\n",
      "Average true similarity: -2.1681223241590217, Average false similarity: -2.1827767584097884\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 854: Accuracy 50.82%\n",
      "Average true similarity: -2.1676452599388374, Average false similarity: -2.183131498470946\n",
      "True negative: 980, True positive: 17, False negative: 1, False positive: 964\n",
      "Epoch 855: Accuracy 50.71%\n",
      "Average true similarity: -2.168288481141691, Average false similarity: -2.18387869520897\n",
      "True negative: 980, True positive: 15, False negative: 1, False positive: 966\n",
      "Epoch 856: Accuracy 50.66%\n",
      "Average true similarity: -2.169496432212025, Average false similarity: -2.18351987767584\n",
      "True negative: 978, True positive: 16, False negative: 3, False positive: 965\n",
      "Epoch 857: Accuracy 50.51%\n",
      "Average true similarity: -2.1696941896024486, Average false similarity: -2.184041794087667\n",
      "True negative: 980, True positive: 11, False negative: 1, False positive: 970\n",
      "Epoch 858: Accuracy 50.87%\n",
      "Average true similarity: -2.1693139653414915, Average false similarity: -2.183854230377164\n",
      "True negative: 978, True positive: 20, False negative: 3, False positive: 961\n",
      "Epoch 859: Accuracy 50.56%\n",
      "Average true similarity: -2.1692935779816507, Average false similarity: -2.1839276248725796\n",
      "True negative: 978, True positive: 14, False negative: 3, False positive: 967\n",
      "Epoch 860: Accuracy 50.46%\n",
      "Average true similarity: -2.1691284403669737, Average false similarity: -2.18440265035678\n",
      "True negative: 978, True positive: 12, False negative: 3, False positive: 969\n",
      "Epoch 861: Accuracy 50.56%\n",
      "Average true similarity: -2.169112130479104, Average false similarity: -2.1842854230377173\n",
      "True negative: 977, True positive: 15, False negative: 4, False positive: 966\n",
      "Epoch 862: Accuracy 50.76%\n",
      "Average true similarity: -2.16880733944954, Average false similarity: -2.1838960244648296\n",
      "True negative: 978, True positive: 18, False negative: 3, False positive: 963\n",
      "Epoch 863: Accuracy 50.71%\n",
      "Average true similarity: -2.1691580020387375, Average false similarity: -2.18400713557594\n",
      "True negative: 980, True positive: 15, False negative: 1, False positive: 966\n",
      "Epoch 864: Accuracy 50.66%\n",
      "Average true similarity: -2.16914882772681, Average false similarity: -2.1832915392456655\n",
      "True negative: 980, True positive: 14, False negative: 1, False positive: 967\n",
      "Epoch 865: Accuracy 50.71%\n",
      "Average true similarity: -2.1692925586136615, Average false similarity: -2.1849949031600384\n",
      "True negative: 980, True positive: 15, False negative: 1, False positive: 966\n",
      "Epoch 866: Accuracy 50.66%\n",
      "Average true similarity: -2.169423037716616, Average false similarity: -2.183708460754331\n",
      "True negative: 978, True positive: 16, False negative: 3, False positive: 965\n",
      "Epoch 867: Accuracy 50.87%\n",
      "Average true similarity: -2.1684250764525985, Average false similarity: -2.18387155963303\n",
      "True negative: 979, True positive: 19, False negative: 2, False positive: 962\n",
      "Epoch 868: Accuracy 50.82%\n",
      "Average true similarity: -2.1694658511722738, Average false similarity: -2.1842150866462786\n",
      "True negative: 981, True positive: 16, False negative: 0, False positive: 965\n",
      "Epoch 869: Accuracy 50.82%\n",
      "Average true similarity: -2.169527013251786, Average false similarity: -2.1838450560652394\n",
      "True negative: 978, True positive: 19, False negative: 3, False positive: 962\n",
      "Epoch 870: Accuracy 50.51%\n",
      "Average true similarity: -2.169854230377164, Average false similarity: -2.184788990825688\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 871: Accuracy 50.41%\n",
      "Average true similarity: -2.1698644240570806, Average false similarity: -2.1845779816513766\n",
      "True negative: 978, True positive: 11, False negative: 3, False positive: 970\n",
      "Epoch 872: Accuracy 50.56%\n",
      "Average true similarity: -2.169306829765546, Average false similarity: -2.184223241590215\n",
      "True negative: 980, True positive: 12, False negative: 1, False positive: 969\n",
      "Epoch 873: Accuracy 50.51%\n",
      "Average true similarity: -2.1699602446483155, Average false similarity: -2.1845820591233442\n",
      "True negative: 980, True positive: 11, False negative: 1, False positive: 970\n",
      "Epoch 874: Accuracy 50.71%\n",
      "Average true similarity: -2.169655453618755, Average false similarity: -2.184543323139651\n",
      "True negative: 978, True positive: 17, False negative: 3, False positive: 964\n",
      "Epoch 875: Accuracy 50.82%\n",
      "Average true similarity: -2.1696972477064236, Average false similarity: -2.1843577981651348\n",
      "True negative: 979, True positive: 18, False negative: 2, False positive: 963\n",
      "Epoch 876: Accuracy 50.61%\n",
      "Average true similarity: -2.1694954128440376, Average false similarity: -2.1847981651376145\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 877: Accuracy 50.41%\n",
      "Average true similarity: -2.17009480122324, Average false similarity: -2.1846707441386353\n",
      "True negative: 979, True positive: 10, False negative: 2, False positive: 971\n",
      "Epoch 878: Accuracy 50.76%\n",
      "Average true similarity: -2.170077471967382, Average false similarity: -2.184811416921507\n",
      "True negative: 977, True positive: 19, False negative: 4, False positive: 962\n",
      "Epoch 879: Accuracy 50.71%\n",
      "Average true similarity: -2.1709235474006148, Average false similarity: -2.185810397553518\n",
      "True negative: 981, True positive: 14, False negative: 0, False positive: 967\n",
      "Epoch 880: Accuracy 50.41%\n",
      "Average true similarity: -2.170888888888888, Average false similarity: -2.18630377166157\n",
      "True negative: 977, True positive: 12, False negative: 4, False positive: 969\n",
      "Epoch 881: Accuracy 50.71%\n",
      "Average true similarity: -2.170940876656474, Average false similarity: -2.1863690112130496\n",
      "True negative: 979, True positive: 16, False negative: 2, False positive: 965\n",
      "Epoch 882: Accuracy 50.51%\n",
      "Average true similarity: -2.171042813455657, Average false similarity: -2.185304791029564\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 883: Accuracy 50.51%\n",
      "Average true similarity: -2.1715790010193703, Average false similarity: -2.1865331294597334\n",
      "True negative: 980, True positive: 11, False negative: 1, False positive: 970\n",
      "Epoch 884: Accuracy 50.66%\n",
      "Average true similarity: -2.170406727828748, Average false similarity: -2.18494291539246\n",
      "True negative: 980, True positive: 14, False negative: 1, False positive: 967\n",
      "Epoch 885: Accuracy 50.46%\n",
      "Average true similarity: -2.1704006116207943, Average false similarity: -2.185159021406727\n",
      "True negative: 976, True positive: 14, False negative: 5, False positive: 967\n",
      "Epoch 886: Accuracy 50.36%\n",
      "Average true similarity: -2.1704403669724766, Average false similarity: -2.184601427115189\n",
      "True negative: 979, True positive: 9, False negative: 2, False positive: 972\n",
      "Epoch 887: Accuracy 50.92%\n",
      "Average true similarity: -2.1699367991845055, Average false similarity: -2.1846371049949025\n",
      "True negative: 981, True positive: 18, False negative: 0, False positive: 963\n",
      "Epoch 888: Accuracy 50.56%\n",
      "Average true similarity: -2.17045769622834, Average false similarity: -2.1849469928644245\n",
      "True negative: 977, True positive: 15, False negative: 4, False positive: 966\n",
      "Epoch 889: Accuracy 50.61%\n",
      "Average true similarity: -2.170255861365952, Average false similarity: -2.1856177370030583\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 890: Accuracy 50.76%\n",
      "Average true similarity: -2.1704994903160064, Average false similarity: -2.1845270132517833\n",
      "True negative: 979, True positive: 17, False negative: 2, False positive: 964\n",
      "Epoch 891: Accuracy 50.61%\n",
      "Average true similarity: -2.1712069317023475, Average false similarity: -2.1857390417940867\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 892: Accuracy 50.36%\n",
      "Average true similarity: -2.1706472986748215, Average false similarity: -2.1859785932721705\n",
      "True negative: 978, True positive: 10, False negative: 3, False positive: 971\n",
      "Epoch 893: Accuracy 50.82%\n",
      "Average true similarity: -2.170903160040774, Average false similarity: -2.1862731906218165\n",
      "True negative: 980, True positive: 17, False negative: 1, False positive: 964\n",
      "Epoch 894: Accuracy 50.61%\n",
      "Average true similarity: -2.170982670744139, Average false similarity: -2.186192660550459\n",
      "True negative: 980, True positive: 13, False negative: 1, False positive: 968\n",
      "Epoch 895: Accuracy 50.71%\n",
      "Average true similarity: -2.1712619775739026, Average false similarity: -2.185908256880733\n",
      "True negative: 977, True positive: 18, False negative: 4, False positive: 963\n",
      "Epoch 896: Accuracy 51.02%\n",
      "Average true similarity: -2.1718837920489302, Average false similarity: -2.1862191641182442\n",
      "True negative: 979, True positive: 22, False negative: 2, False positive: 959\n",
      "Epoch 897: Accuracy 50.66%\n",
      "Average true similarity: -2.17218246687054, Average false similarity: -2.1871610601427105\n",
      "True negative: 978, True positive: 16, False negative: 3, False positive: 965\n",
      "Epoch 898: Accuracy 50.76%\n",
      "Average true similarity: -2.1720560652395506, Average false similarity: -2.187297655453619\n",
      "True negative: 980, True positive: 16, False negative: 1, False positive: 965\n",
      "Epoch 899: Accuracy 50.51%\n",
      "Average true similarity: -2.1716126401630995, Average false similarity: -2.186808358817532\n",
      "True negative: 978, True positive: 13, False negative: 3, False positive: 968\n",
      "Epoch 900: Accuracy 50.66%\n",
      "Average true similarity: -2.1724332313965333, Average false similarity: -2.1874128440366962\n",
      "True negative: 979, True positive: 15, False negative: 2, False positive: 966\n",
      "Epoch 901: Accuracy 50.46%\n",
      "Average true similarity: -2.172420998980628, Average false similarity: -2.1874923547400598\n",
      "True negative: 979, True positive: 11, False negative: 2, False positive: 970\n",
      "Epoch 902: Accuracy 50.71%\n",
      "Average true similarity: -2.1727747196738023, Average false similarity: -2.187694189602446\n",
      "True negative: 980, True positive: 15, False negative: 1, False positive: 966\n",
      "Epoch 903: Accuracy 50.56%\n",
      "Average true similarity: -2.172516819571867, Average false similarity: -2.187749235474007\n",
      "True negative: 977, True positive: 15, False negative: 4, False positive: 966\n",
      "Epoch 904: Accuracy 50.36%\n",
      "Average true similarity: -2.1741335372069326, Average false similarity: -2.1890020387359805\n",
      "True negative: 978, True positive: 10, False negative: 3, False positive: 971\n",
      "Epoch 905: Accuracy 50.56%\n",
      "Average true similarity: -2.174201834862384, Average false similarity: -2.1887787971457704\n",
      "True negative: 980, True positive: 12, False negative: 1, False positive: 969\n",
      "Epoch 906: Accuracy 50.51%\n",
      "Average true similarity: -2.1732935779816485, Average false similarity: -2.1877502548419963\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 907: Accuracy 50.46%\n",
      "Average true similarity: -2.17280122324159, Average false similarity: -2.187822629969419\n",
      "True negative: 980, True positive: 10, False negative: 1, False positive: 971\n",
      "Epoch 908: Accuracy 50.76%\n",
      "Average true similarity: -2.173476044852191, Average false similarity: -2.1894495412844046\n",
      "True negative: 981, True positive: 15, False negative: 0, False positive: 966\n",
      "Epoch 909: Accuracy 50.61%\n",
      "Average true similarity: -2.1742619775739023, Average false similarity: -2.1898491335372103\n",
      "True negative: 981, True positive: 12, False negative: 0, False positive: 969\n",
      "Epoch 910: Accuracy 50.71%\n",
      "Average true similarity: -2.1739306829765535, Average false similarity: -2.1884413863404735\n",
      "True negative: 980, True positive: 15, False negative: 1, False positive: 966\n",
      "Epoch 911: Accuracy 50.51%\n",
      "Average true similarity: -2.1730988786952117, Average false similarity: -2.1887808358817535\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 912: Accuracy 50.61%\n",
      "Average true similarity: -2.173975535168198, Average false similarity: -2.189277268093779\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 913: Accuracy 50.56%\n",
      "Average true similarity: -2.174715596330278, Average false similarity: -2.1896870540265025\n",
      "True negative: 980, True positive: 12, False negative: 1, False positive: 969\n",
      "Epoch 914: Accuracy 50.76%\n",
      "Average true similarity: -2.172864424057087, Average false similarity: -2.1881345565749224\n",
      "True negative: 979, True positive: 17, False negative: 2, False positive: 964\n",
      "Epoch 915: Accuracy 50.76%\n",
      "Average true similarity: -2.173575942915391, Average false similarity: -2.187906218144751\n",
      "True negative: 980, True positive: 16, False negative: 1, False positive: 965\n",
      "Epoch 916: Accuracy 50.66%\n",
      "Average true similarity: -2.1741416921508674, Average false similarity: -2.1897991845056057\n",
      "True negative: 979, True positive: 15, False negative: 2, False positive: 966\n",
      "Epoch 917: Accuracy 50.61%\n",
      "Average true similarity: -2.1743700305810423, Average false similarity: -2.1903669724770656\n",
      "True negative: 977, True positive: 16, False negative: 4, False positive: 965\n",
      "Epoch 918: Accuracy 50.82%\n",
      "Average true similarity: -2.1753098878695196, Average false similarity: -2.1907145769622804\n",
      "True negative: 979, True positive: 18, False negative: 2, False positive: 963\n",
      "Epoch 919: Accuracy 50.61%\n",
      "Average true similarity: -2.1761590214067255, Average false similarity: -2.1911294597349653\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 920: Accuracy 50.76%\n",
      "Average true similarity: -2.1755249745158003, Average false similarity: -2.1909072375127425\n",
      "True negative: 980, True positive: 16, False negative: 1, False positive: 965\n",
      "Epoch 921: Accuracy 50.41%\n",
      "Average true similarity: -2.175585117227323, Average false similarity: -2.190211009174314\n",
      "True negative: 980, True positive: 9, False negative: 1, False positive: 972\n",
      "Epoch 922: Accuracy 50.56%\n",
      "Average true similarity: -2.175695208970438, Average false similarity: -2.1910805300713534\n",
      "True negative: 978, True positive: 14, False negative: 3, False positive: 967\n",
      "Epoch 923: Accuracy 50.51%\n",
      "Average true similarity: -2.1757645259938845, Average false similarity: -2.19093272171254\n",
      "True negative: 981, True positive: 10, False negative: 0, False positive: 971\n",
      "Epoch 924: Accuracy 50.46%\n",
      "Average true similarity: -2.1763985728848105, Average false similarity: -2.1915331294597356\n",
      "True negative: 980, True positive: 10, False negative: 1, False positive: 971\n",
      "Epoch 925: Accuracy 50.71%\n",
      "Average true similarity: -2.1755922528032627, Average false similarity: -2.1906788990825694\n",
      "True negative: 980, True positive: 15, False negative: 1, False positive: 966\n",
      "Epoch 926: Accuracy 50.51%\n",
      "Average true similarity: -2.176928644240571, Average false similarity: -2.191790010193677\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 927: Accuracy 50.56%\n",
      "Average true similarity: -2.175670744138631, Average false similarity: -2.191429153924569\n",
      "True negative: 979, True positive: 13, False negative: 2, False positive: 968\n",
      "Epoch 928: Accuracy 50.36%\n",
      "Average true similarity: -2.1761314984709474, Average false similarity: -2.191239551478079\n",
      "True negative: 980, True positive: 8, False negative: 1, False positive: 973\n",
      "Epoch 929: Accuracy 50.61%\n",
      "Average true similarity: -2.176079510703367, Average false similarity: -2.190661569826705\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 930: Accuracy 50.36%\n",
      "Average true similarity: -2.1764311926605475, Average false similarity: -2.191779816513761\n",
      "True negative: 980, True positive: 8, False negative: 1, False positive: 973\n",
      "Epoch 931: Accuracy 50.56%\n",
      "Average true similarity: -2.175050968399593, Average false similarity: -2.1908664627930663\n",
      "True negative: 980, True positive: 12, False negative: 1, False positive: 969\n",
      "Epoch 932: Accuracy 50.36%\n",
      "Average true similarity: -2.1760795107033646, Average false similarity: -2.1908511722731885\n",
      "True negative: 978, True positive: 10, False negative: 3, False positive: 971\n",
      "Epoch 933: Accuracy 50.61%\n",
      "Average true similarity: -2.1757288481141708, Average false similarity: -2.190898063200811\n",
      "True negative: 980, True positive: 13, False negative: 1, False positive: 968\n",
      "Epoch 934: Accuracy 50.61%\n",
      "Average true similarity: -2.1756106014271164, Average false similarity: -2.1904464831804265\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 935: Accuracy 50.46%\n",
      "Average true similarity: -2.175629969418962, Average false similarity: -2.190328236493377\n",
      "True negative: 980, True positive: 10, False negative: 1, False positive: 971\n",
      "Epoch 936: Accuracy 50.51%\n",
      "Average true similarity: -2.1751569826707433, Average false similarity: -2.1912568807339445\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 937: Accuracy 50.41%\n",
      "Average true similarity: -2.176178389398573, Average false similarity: -2.191135575942916\n",
      "True negative: 980, True positive: 9, False negative: 1, False positive: 972\n",
      "Epoch 938: Accuracy 50.56%\n",
      "Average true similarity: -2.1756024464831776, Average false similarity: -2.1908124362894985\n",
      "True negative: 978, True positive: 14, False negative: 3, False positive: 967\n",
      "Epoch 939: Accuracy 50.71%\n",
      "Average true similarity: -2.1747471967380214, Average false similarity: -2.190585117227322\n",
      "True negative: 979, True positive: 16, False negative: 2, False positive: 965\n",
      "Epoch 940: Accuracy 50.87%\n",
      "Average true similarity: -2.1749826707441375, Average false similarity: -2.1895769622833856\n",
      "True negative: 980, True positive: 18, False negative: 1, False positive: 963\n",
      "Epoch 941: Accuracy 50.41%\n",
      "Average true similarity: -2.1743323139653414, Average false similarity: -2.1890132517838956\n",
      "True negative: 978, True positive: 11, False negative: 3, False positive: 970\n",
      "Epoch 942: Accuracy 50.41%\n",
      "Average true similarity: -2.17464118246687, Average false similarity: -2.1901131498470945\n",
      "True negative: 979, True positive: 10, False negative: 2, False positive: 971\n",
      "Epoch 943: Accuracy 50.87%\n",
      "Average true similarity: -2.1733414882772673, Average false similarity: -2.188892966360857\n",
      "True negative: 980, True positive: 18, False negative: 1, False positive: 963\n",
      "Epoch 944: Accuracy 50.2%\n",
      "Average true similarity: -2.1743088685015306, Average false similarity: -2.189064220183487\n",
      "True negative: 978, True positive: 7, False negative: 3, False positive: 974\n",
      "Epoch 945: Accuracy 50.71%\n",
      "Average true similarity: -2.1734281345565747, Average false similarity: -2.188227319062183\n",
      "True negative: 977, True positive: 18, False negative: 4, False positive: 963\n",
      "Epoch 946: Accuracy 50.51%\n",
      "Average true similarity: -2.1749459734964316, Average false similarity: -2.1895861365953126\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 947: Accuracy 50.25%\n",
      "Average true similarity: -2.174259938837919, Average false similarity: -2.1893058103975545\n",
      "True negative: 977, True positive: 9, False negative: 4, False positive: 972\n",
      "Epoch 948: Accuracy 50.66%\n",
      "Average true similarity: -2.174273190621812, Average false similarity: -2.1895963302752297\n",
      "True negative: 979, True positive: 15, False negative: 2, False positive: 966\n",
      "Epoch 949: Accuracy 50.56%\n",
      "Average true similarity: -2.1739011213047914, Average false similarity: -2.189085626911318\n",
      "True negative: 980, True positive: 12, False negative: 1, False positive: 969\n",
      "Epoch 950: Accuracy 50.56%\n",
      "Average true similarity: -2.1741977573904183, Average false similarity: -2.1893149847094757\n",
      "True negative: 978, True positive: 14, False negative: 3, False positive: 967\n",
      "Epoch 951: Accuracy 50.66%\n",
      "Average true similarity: -2.174173292558613, Average false similarity: -2.189158002038734\n",
      "True negative: 978, True positive: 16, False negative: 3, False positive: 965\n",
      "Epoch 952: Accuracy 50.82%\n",
      "Average true similarity: -2.1742466870540236, Average false similarity: -2.1890458715596326\n",
      "True negative: 979, True positive: 18, False negative: 2, False positive: 963\n",
      "Epoch 953: Accuracy 50.76%\n",
      "Average true similarity: -2.1740356778797167, Average false similarity: -2.1891111111111075\n",
      "True negative: 979, True positive: 17, False negative: 2, False positive: 964\n",
      "Epoch 954: Accuracy 50.61%\n",
      "Average true similarity: -2.1744760448521907, Average false similarity: -2.189363914373085\n",
      "True negative: 981, True positive: 12, False negative: 0, False positive: 969\n",
      "Epoch 955: Accuracy 50.66%\n",
      "Average true similarity: -2.174953109072376, Average false similarity: -2.1903751274209973\n",
      "True negative: 981, True positive: 13, False negative: 0, False positive: 968\n",
      "Epoch 956: Accuracy 50.31%\n",
      "Average true similarity: -2.1750856269113163, Average false similarity: -2.18991029561672\n",
      "True negative: 979, True positive: 8, False negative: 2, False positive: 973\n",
      "Epoch 957: Accuracy 50.46%\n",
      "Average true similarity: -2.1753160040774744, Average false similarity: -2.1900122324159\n",
      "True negative: 980, True positive: 10, False negative: 1, False positive: 971\n",
      "Epoch 958: Accuracy 50.41%\n",
      "Average true similarity: -2.175269113149845, Average false similarity: -2.1907675840978595\n",
      "True negative: 979, True positive: 10, False negative: 2, False positive: 971\n",
      "Epoch 959: Accuracy 50.71%\n",
      "Average true similarity: -2.174675840978594, Average false similarity: -2.1893435270132535\n",
      "True negative: 979, True positive: 16, False negative: 2, False positive: 965\n",
      "Epoch 960: Accuracy 50.51%\n",
      "Average true similarity: -2.173533129459734, Average false similarity: -2.188381243628954\n",
      "True negative: 978, True positive: 13, False negative: 3, False positive: 968\n",
      "Epoch 961: Accuracy 50.56%\n",
      "Average true similarity: -2.1735453618756364, Average false similarity: -2.189430173292559\n",
      "True negative: 977, True positive: 15, False negative: 4, False positive: 966\n",
      "Epoch 962: Accuracy 50.51%\n",
      "Average true similarity: -2.173774719673807, Average false similarity: -2.1894841997961265\n",
      "True negative: 978, True positive: 13, False negative: 3, False positive: 968\n",
      "Epoch 963: Accuracy 50.46%\n",
      "Average true similarity: -2.1737257900101934, Average false similarity: -2.1883017329255847\n",
      "True negative: 977, True positive: 13, False negative: 4, False positive: 968\n",
      "Epoch 964: Accuracy 50.82%\n",
      "Average true similarity: -2.1741029561671787, Average false similarity: -2.1890499490316007\n",
      "True negative: 980, True positive: 17, False negative: 1, False positive: 964\n",
      "Epoch 965: Accuracy 50.87%\n",
      "Average true similarity: -2.173646279306829, Average false similarity: -2.1886146788990852\n",
      "True negative: 980, True positive: 18, False negative: 1, False positive: 963\n",
      "Epoch 966: Accuracy 50.66%\n",
      "Average true similarity: -2.1736646279306826, Average false similarity: -2.18822120285423\n",
      "True negative: 978, True positive: 16, False negative: 3, False positive: 965\n",
      "Epoch 967: Accuracy 50.66%\n",
      "Average true similarity: -2.173814475025484, Average false similarity: -2.188931702344544\n",
      "True negative: 978, True positive: 16, False negative: 3, False positive: 965\n",
      "Epoch 968: Accuracy 50.46%\n",
      "Average true similarity: -2.1744740061162076, Average false similarity: -2.189748216106015\n",
      "True negative: 978, True positive: 12, False negative: 3, False positive: 969\n",
      "Epoch 969: Accuracy 50.71%\n",
      "Average true similarity: -2.1737492354740047, Average false similarity: -2.188668705402652\n",
      "True negative: 980, True positive: 15, False negative: 1, False positive: 966\n",
      "Epoch 970: Accuracy 50.46%\n",
      "Average true similarity: -2.173853211009173, Average false similarity: -2.1887227319062177\n",
      "True negative: 978, True positive: 12, False negative: 3, False positive: 969\n",
      "Epoch 971: Accuracy 50.71%\n",
      "Average true similarity: -2.1747461773700305, Average false similarity: -2.1900356778797114\n",
      "True negative: 980, True positive: 15, False negative: 1, False positive: 966\n",
      "Epoch 972: Accuracy 50.61%\n",
      "Average true similarity: -2.1746401630988785, Average false similarity: -2.1900907237512763\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 973: Accuracy 50.56%\n",
      "Average true similarity: -2.1739938837920487, Average false similarity: -2.1886330275229335\n",
      "True negative: 977, True positive: 15, False negative: 4, False positive: 966\n",
      "Epoch 974: Accuracy 50.66%\n",
      "Average true similarity: -2.1745484199796117, Average false similarity: -2.190229357798166\n",
      "True negative: 978, True positive: 16, False negative: 3, False positive: 965\n",
      "Epoch 975: Accuracy 50.66%\n",
      "Average true similarity: -2.1739001019368, Average false similarity: -2.189411824668705\n",
      "True negative: 980, True positive: 14, False negative: 1, False positive: 967\n",
      "Epoch 976: Accuracy 50.76%\n",
      "Average true similarity: -2.1743119266055047, Average false similarity: -2.189218144750254\n",
      "True negative: 980, True positive: 16, False negative: 1, False positive: 965\n",
      "Epoch 977: Accuracy 50.46%\n",
      "Average true similarity: -2.1749001019367977, Average false similarity: -2.190002038735986\n",
      "True negative: 978, True positive: 12, False negative: 3, False positive: 969\n",
      "Epoch 978: Accuracy 50.56%\n",
      "Average true similarity: -2.175147808358817, Average false similarity: -2.1902650356778812\n",
      "True negative: 978, True positive: 14, False negative: 3, False positive: 967\n",
      "Epoch 979: Accuracy 50.56%\n",
      "Average true similarity: -2.175602446483178, Average false similarity: -2.190681957186543\n",
      "True negative: 981, True positive: 11, False negative: 0, False positive: 970\n",
      "Epoch 980: Accuracy 50.61%\n",
      "Average true similarity: -2.174954128440369, Average false similarity: -2.190625891946992\n",
      "True negative: 979, True positive: 14, False negative: 2, False positive: 967\n",
      "Epoch 981: Accuracy 50.71%\n",
      "Average true similarity: -2.17520387359837, Average false similarity: -2.190822629969421\n",
      "True negative: 979, True positive: 16, False negative: 2, False positive: 965\n",
      "Epoch 982: Accuracy 50.31%\n",
      "Average true similarity: -2.1750295616717636, Average false similarity: -2.1898205912334343\n",
      "True negative: 980, True positive: 7, False negative: 1, False positive: 974\n",
      "Epoch 983: Accuracy 50.61%\n",
      "Average true similarity: -2.1747991845056087, Average false similarity: -2.190624872579001\n",
      "True negative: 980, True positive: 13, False negative: 1, False positive: 968\n",
      "Epoch 984: Accuracy 50.51%\n",
      "Average true similarity: -2.1753190621814467, Average false similarity: -2.1903221202854235\n",
      "True negative: 981, True positive: 10, False negative: 0, False positive: 971\n",
      "Epoch 985: Accuracy 50.61%\n",
      "Average true similarity: -2.175148827726809, Average false similarity: -2.190257900101934\n",
      "True negative: 978, True positive: 15, False negative: 3, False positive: 966\n",
      "Epoch 986: Accuracy 50.51%\n",
      "Average true similarity: -2.1752558613659545, Average false similarity: -2.1904740061162102\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 987: Accuracy 50.41%\n",
      "Average true similarity: -2.1753781855249756, Average false similarity: -2.190447502548419\n",
      "True negative: 979, True positive: 10, False negative: 2, False positive: 971\n",
      "Epoch 988: Accuracy 50.51%\n",
      "Average true similarity: -2.1745779816513786, Average false similarity: -2.189845056065236\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 989: Accuracy 50.41%\n",
      "Average true similarity: -2.174227319062179, Average false similarity: -2.189318042813461\n",
      "True negative: 978, True positive: 11, False negative: 3, False positive: 970\n",
      "Epoch 990: Accuracy 50.71%\n",
      "Average true similarity: -2.1745616717635086, Average false similarity: -2.1891416921508706\n",
      "True negative: 978, True positive: 17, False negative: 3, False positive: 964\n",
      "Epoch 991: Accuracy 50.66%\n",
      "Average true similarity: -2.1745657492354717, Average false similarity: -2.1889592252803256\n",
      "True negative: 980, True positive: 14, False negative: 1, False positive: 967\n",
      "Epoch 992: Accuracy 50.71%\n",
      "Average true similarity: -2.174060142711521, Average false similarity: -2.189369011213046\n",
      "True negative: 979, True positive: 16, False negative: 2, False positive: 965\n",
      "Epoch 993: Accuracy 50.51%\n",
      "Average true similarity: -2.1741304791029554, Average false similarity: -2.189873598369013\n",
      "True negative: 980, True positive: 11, False negative: 1, False positive: 970\n",
      "Epoch 994: Accuracy 50.51%\n",
      "Average true similarity: -2.1747492354740077, Average false similarity: -2.189354740061163\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 995: Accuracy 50.2%\n",
      "Average true similarity: -2.1746585117227326, Average false similarity: -2.189460754332315\n",
      "True negative: 976, True positive: 9, False negative: 5, False positive: 972\n",
      "Epoch 996: Accuracy 50.51%\n",
      "Average true similarity: -2.175415902140672, Average false similarity: -2.190749235474006\n",
      "True negative: 979, True positive: 12, False negative: 2, False positive: 969\n",
      "Epoch 997: Accuracy 50.46%\n",
      "Average true similarity: -2.174555555555555, Average false similarity: -2.189451580020389\n",
      "True negative: 978, True positive: 12, False negative: 3, False positive: 969\n",
      "Epoch 998: Accuracy 50.56%\n",
      "Average true similarity: -2.1747930682976575, Average false similarity: -2.1896982670744194\n",
      "True negative: 978, True positive: 14, False negative: 3, False positive: 967\n",
      "Epoch 999: Accuracy 50.66%\n",
      "Average true similarity: -2.175201834862386, Average false similarity: -2.190240570846072\n",
      "True negative: 981, True positive: 13, False negative: 0, False positive: 968\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0.98, 'Similarity Distribution Graph')"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Experiment 11: Similarity Distribution of 4-bit precision bound hypervector\n",
    "\n",
    "import dna\n",
    "import hdgim\n",
    "import torch\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "torch.set_printoptions(sci_mode=False)\n",
    "\n",
    "# [begin] hyperparameters\n",
    "HYPERVECTOR_DIMENSION = 1000  # denoted D in paper\n",
    "DNA_SEQUENCE_LENGTH = 1000\n",
    "DNA_SUBSEQUENCE_LENGTH = 20  # denoted n in paper\n",
    "BIT_PRECISION = 4 # denoted B in paper\n",
    "NOISE_PROBABILTY = 0.1  # denoted p in paper, probability: 0 ~ 1, simulate FeFET noise\n",
    "EPOCHES = 1000\n",
    "FUNCTION = 'hamming'\n",
    "LEARNING_RATE = 10\n",
    "THRESHOLD = -2.02\n",
    "\n",
    "# create dataset\n",
    "dataset = dna.DNADataset(DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH)\n",
    "\n",
    "# create model\n",
    "model = hdgim.HDGIM(HYPERVECTOR_DIMENSION, DNA_SEQUENCE_LENGTH, DNA_SUBSEQUENCE_LENGTH, BIT_PRECISION, NOISE_PROBABILTY)\n",
    "model.set_dataset(dataset)\n",
    "model.create_base_hypervectors()\n",
    "model.create_dna_sequence()\n",
    "model.create_dna_subsequences()\n",
    "model.bind()\n",
    "model.quantize_cdf()\n",
    "model.noise()\n",
    "\n",
    "# train\n",
    "accuracies, true_similarities, false_similarities = model.train(EPOCHES, LEARNING_RATE, THRESHOLD, FUNCTION, False, True, True)\n",
    "\n",
    "# draw histogram which shows true similarity distribution and false similarity distribution\n",
    "fig, axes = plt.subplots(1, 2, figsize=(10, 5))\n",
    "\n",
    "axes[0].hist(true_similarities[0], bins=30, density=True, alpha=0.5, color='blue', edgecolor='black', label='true')\n",
    "axes[0].hist(false_similarities[0], bins=30, density=True, alpha=0.5, color='red', edgecolor='black', label='false')\n",
    "axes[0].legend(prop={'size': 10})\n",
    "axes[0].set_title('Epoch 0')\n",
    "axes[0].set_xlabel('Similarity')\n",
    "axes[0].set_ylabel('Density')\n",
    "\n",
    "axes[1].hist(true_similarities[EPOCHES-1], bins=30, density=True, alpha=0.5, color='blue', edgecolor='black', label='true')\n",
    "axes[1].hist(false_similarities[EPOCHES-1], bins=30, density=True, alpha=0.5, color='red', edgecolor='black', label='false')\n",
    "axes[1].legend(prop={'size': 10})\n",
    "axes[1].set_title('Final Epoch')\n",
    "axes[1].set_xlabel('Similarity')\n",
    "axes[1].set_ylabel('Density')\n",
    "\n",
    "plt.suptitle('Similarity Distribution Graph')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}