Ru_catboost.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "import seaborn as sns\n",
    "from matplotlib import __version__ as ma_ver\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "\n",
    "from sklearn import __version__ as sk_ver\n",
    "from sklearn.model_selection import GridSearchCV, StratifiedKFold, train_test_split\n",
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.preprocessing import StandardScaler, normalize\n",
    "from sklearn.decomposition import TruncatedSVD\n",
    "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, fbeta_score, roc_auc_score, silhouette_score, ConfusionMatrixDisplay, confusion_matrix\n",
    "from sklearn.metrics import make_scorer, fbeta_score\n",
    "from catboost import CatBoostClassifier, Pool, __version__ as catb_v\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "\n",
    "from plotly import __version__ as pl_ver\n",
    "from plotly.tools import FigureFactory as ff\n",
    "from plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot\n",
    "from plotly.colors import n_colors\n",
    "import plotly.express as px\n",
    "import plotly.graph_objs as go\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "from joblib import dump, load, __version__ as joblib_ver\n",
    "\n",
    "# import util"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "pandas version: 1.3.4\n",
      "seaborn version: 0.12.2\n",
      "numpy version: 1.24.3\n",
      "matplotlib version: 3.7.2\n",
      "sklearn version: 1.3.0\n",
      "plotly version: 5.19.0\n",
      "catboost version: 1.2.3\n",
      "joblib version: 1.2.0\n"
     ]
    }
   ],
   "source": [
    "print(\"pandas version:\", pd.__version__)\n",
    "print(\"seaborn version:\", sns.__version__)\n",
    "print(\"numpy version:\", np.__version__)\n",
    "print(\"matplotlib version:\", ma_ver)\n",
    "print(\"sklearn version:\", sk_ver)\n",
    "print(\"plotly version:\", pl_ver)\n",
    "print(\"catboost version:\", catb_v)\n",
    "print(\"joblib version:\", joblib_ver)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "# Загрузка файлов"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>prompt</th>\n",
       "      <th>canonical_solution</th>\n",
       "      <th>model_response</th>\n",
       "      <th>target</th>\n",
       "      <th>processed_prompt</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>&lt;file_sep&gt;&lt;fim_prefix&gt;// Copyright 2020 The Go...</td>\n",
       "      <td>\\t\\tswitch x.Interface().(type) {</td>\n",
       "      <td>\\t\\t// False sorts before true.</td>\n",
       "      <td>0</td>\n",
       "      <td>copyright 2020 go author right reserved use so...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>&lt;file_sep&gt;&lt;fim_prefix&gt;import re\\nimport uuid\\n...</td>\n",
       "      <td>st.markdown(\"---\")</td>\n",
       "      <td>if st.session_state[\"mdl_json\"]:</td>\n",
       "      <td>0</td>\n",
       "      <td>import import uuid import orjson import stream...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>&lt;file_sep&gt;&lt;fim_prefix&gt;package tlsrpt\\nimport (...</td>\n",
       "      <td>func (p *parser) wsp() {</td>\n",
       "      <td>//../rfc/8460:374</td>\n",
       "      <td>0</td>\n",
       "      <td>package tlsrpt import fmt string extension add...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>&lt;file_sep&gt;&lt;fim_prefix&gt;import { describe, expec...</td>\n",
       "      <td>expect(() =&gt; (rect = new class_test.Rect()...</td>\n",
       "      <td>expect(() =&gt; (rect = new class_test.Rectan...</td>\n",
       "      <td>0</td>\n",
       "      <td>import describe expect test const require cons...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>&lt;file_sep&gt;&lt;fim_prefix&gt;import type {\\n  Default...</td>\n",
       "      <td>* Include or exclude stale queries</td>\n",
       "      <td>* Include queries matching this mutation key</td>\n",
       "      <td>0</td>\n",
       "      <td>import type defaulterror enabled fetchstatus m...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6657</th>\n",
       "      <td>&lt;file_sep&gt;&lt;fim_prefix&gt;import { describe, expec...</td>\n",
       "      <td>queryKey: ['key'],</td>\n",
       "      <td>queryKey: ['key'],</td>\n",
       "      <td>1</td>\n",
       "      <td>import describe expecttypeof import injectquer...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6658</th>\n",
       "      <td>&lt;file_sep&gt;&lt;fim_prefix&gt;import moment from'momen...</td>\n",
       "      <td>dateRangeType: DateRangeType;</td>\n",
       "      <td>dateRangeType: DateRangeType;</td>\n",
       "      <td>1</td>\n",
       "      <td>import moment import type moment import datemo...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6659</th>\n",
       "      <td>&lt;file_sep&gt;&lt;fim_prefix&gt;package dkim\\nimport (\\n...</td>\n",
       "      <td>func (p *parser) xalgorithm() (string, string) {</td>\n",
       "      <td>//../rfc/6376:658</td>\n",
       "      <td>0</td>\n",
       "      <td>package dkim import fmt strconv string pedanti...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6660</th>\n",
       "      <td>&lt;file_sep&gt;&lt;fim_prefix&gt;package com.tencent.supe...</td>\n",
       "      <td>String line;</td>\n",
       "      <td>String line = br.readLine();</td>\n",
       "      <td>0</td>\n",
       "      <td>package import static import import import imp...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6661</th>\n",
       "      <td>&lt;file_sep&gt;&lt;fim_prefix&gt;import pytest\\nfrom rest...</td>\n",
       "      <td>\"\"\"Perform tests on Projects API endpoint ...</td>\n",
       "      <td>\"\"\"Perform tests on Projects API endpoint ...</td>\n",
       "      <td>1</td>\n",
       "      <td>import pytest import apiclient import project ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>6662 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 prompt  \\\n",
       "0     <file_sep><fim_prefix>// Copyright 2020 The Go...   \n",
       "1     <file_sep><fim_prefix>import re\\nimport uuid\\n...   \n",
       "2     <file_sep><fim_prefix>package tlsrpt\\nimport (...   \n",
       "3     <file_sep><fim_prefix>import { describe, expec...   \n",
       "4     <file_sep><fim_prefix>import type {\\n  Default...   \n",
       "...                                                 ...   \n",
       "6657  <file_sep><fim_prefix>import { describe, expec...   \n",
       "6658  <file_sep><fim_prefix>import moment from'momen...   \n",
       "6659  <file_sep><fim_prefix>package dkim\\nimport (\\n...   \n",
       "6660  <file_sep><fim_prefix>package com.tencent.supe...   \n",
       "6661  <file_sep><fim_prefix>import pytest\\nfrom rest...   \n",
       "\n",
       "                                     canonical_solution  \\\n",
       "0                     \\t\\tswitch x.Interface().(type) {   \n",
       "1                                    st.markdown(\"---\")   \n",
       "2                              func (p *parser) wsp() {   \n",
       "3         expect(() => (rect = new class_test.Rect()...   \n",
       "4                    * Include or exclude stale queries   \n",
       "...                                                 ...   \n",
       "6657                                 queryKey: ['key'],   \n",
       "6658                      dateRangeType: DateRangeType;   \n",
       "6659   func (p *parser) xalgorithm() (string, string) {   \n",
       "6660                                       String line;   \n",
       "6661      \"\"\"Perform tests on Projects API endpoint ...   \n",
       "\n",
       "                                         model_response target  \\\n",
       "0                       \\t\\t// False sorts before true.      0   \n",
       "1                      if st.session_state[\"mdl_json\"]:      0   \n",
       "2                                     //../rfc/8460:374      0   \n",
       "3         expect(() => (rect = new class_test.Rectan...      0   \n",
       "4          * Include queries matching this mutation key      0   \n",
       "...                                                 ...    ...   \n",
       "6657                                 queryKey: ['key'],      1   \n",
       "6658                      dateRangeType: DateRangeType;      1   \n",
       "6659                                  //../rfc/6376:658      0   \n",
       "6660                       String line = br.readLine();      0   \n",
       "6661      \"\"\"Perform tests on Projects API endpoint ...      1   \n",
       "\n",
       "                                       processed_prompt  \n",
       "0     copyright 2020 go author right reserved use so...  \n",
       "1     import import uuid import orjson import stream...  \n",
       "2     package tlsrpt import fmt string extension add...  \n",
       "3     import describe expect test const require cons...  \n",
       "4     import type defaulterror enabled fetchstatus m...  \n",
       "...                                                 ...  \n",
       "6657  import describe expecttypeof import injectquer...  \n",
       "6658  import moment import type moment import datemo...  \n",
       "6659  package dkim import fmt strconv string pedanti...  \n",
       "6660  package import static import import import imp...  \n",
       "6661  import pytest import apiclient import project ...  \n",
       "\n",
       "[6662 rows x 5 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv(\"dataset_bmm.csv\")\n",
    "data['processed_prompt'] = pd.read_csv('processed_prompt_data.csv')\n",
    "data['target'] = data['target'].astype(int).astype('category') # для катбуста\n",
    "data['model_response'] = data['model_response'].astype(str)\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "File \u001b[1;32m<timed exec>:4\u001b[0m\n",
      "File \u001b[1;32mc:\\Users\\light\\anaconda3\\lib\\site-packages\\sklearn\\feature_extraction\\text.py:2126\u001b[0m, in \u001b[0;36mTfidfVectorizer.fit_transform\u001b[1;34m(self, raw_documents, y)\u001b[0m\n\u001b[0;32m   2119\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_check_params()\n\u001b[0;32m   2120\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_tfidf \u001b[38;5;241m=\u001b[39m TfidfTransformer(\n\u001b[0;32m   2121\u001b[0m     norm\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mnorm,\n\u001b[0;32m   2122\u001b[0m     use_idf\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39muse_idf,\n\u001b[0;32m   2123\u001b[0m     smooth_idf\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39msmooth_idf,\n\u001b[0;32m   2124\u001b[0m     sublinear_tf\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39msublinear_tf,\n\u001b[0;32m   2125\u001b[0m )\n\u001b[1;32m-> 2126\u001b[0m X \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43msuper\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfit_transform\u001b[49m\u001b[43m(\u001b[49m\u001b[43mraw_documents\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m   2127\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_tfidf\u001b[38;5;241m.\u001b[39mfit(X)\n\u001b[0;32m   2128\u001b[0m \u001b[38;5;66;03m# X is already a transformed view of raw_documents so\u001b[39;00m\n\u001b[0;32m   2129\u001b[0m \u001b[38;5;66;03m# we set copy to False\u001b[39;00m\n",
      "File \u001b[1;32mc:\\Users\\light\\anaconda3\\lib\\site-packages\\sklearn\\base.py:1151\u001b[0m, in \u001b[0;36m_fit_context.<locals>.decorator.<locals>.wrapper\u001b[1;34m(estimator, *args, **kwargs)\u001b[0m\n\u001b[0;32m   1144\u001b[0m     estimator\u001b[38;5;241m.\u001b[39m_validate_params()\n\u001b[0;32m   1146\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m config_context(\n\u001b[0;32m   1147\u001b[0m     skip_parameter_validation\u001b[38;5;241m=\u001b[39m(\n\u001b[0;32m   1148\u001b[0m         prefer_skip_nested_validation \u001b[38;5;129;01mor\u001b[39;00m global_skip_validation\n\u001b[0;32m   1149\u001b[0m     )\n\u001b[0;32m   1150\u001b[0m ):\n\u001b[1;32m-> 1151\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m fit_method(estimator, \u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n",
      "File \u001b[1;32mc:\\Users\\light\\anaconda3\\lib\\site-packages\\sklearn\\feature_extraction\\text.py:1383\u001b[0m, in \u001b[0;36mCountVectorizer.fit_transform\u001b[1;34m(self, raw_documents, y)\u001b[0m\n\u001b[0;32m   1375\u001b[0m             warnings\u001b[38;5;241m.\u001b[39mwarn(\n\u001b[0;32m   1376\u001b[0m                 \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mUpper case characters found in\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m   1377\u001b[0m                 \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m vocabulary while \u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mlowercase\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m   1378\u001b[0m                 \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m is True. These entries will not\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m   1379\u001b[0m                 \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m be matched with any documents\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m   1380\u001b[0m             )\n\u001b[0;32m   1381\u001b[0m             \u001b[38;5;28;01mbreak\u001b[39;00m\n\u001b[1;32m-> 1383\u001b[0m vocabulary, X \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_count_vocab\u001b[49m\u001b[43m(\u001b[49m\u001b[43mraw_documents\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfixed_vocabulary_\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m   1385\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mbinary:\n\u001b[0;32m   1386\u001b[0m     X\u001b[38;5;241m.\u001b[39mdata\u001b[38;5;241m.\u001b[39mfill(\u001b[38;5;241m1\u001b[39m)\n",
      "File \u001b[1;32mc:\\Users\\light\\anaconda3\\lib\\site-packages\\sklearn\\feature_extraction\\text.py:1270\u001b[0m, in \u001b[0;36mCountVectorizer._count_vocab\u001b[1;34m(self, raw_documents, fixed_vocab)\u001b[0m\n\u001b[0;32m   1268\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m doc \u001b[38;5;129;01min\u001b[39;00m raw_documents:\n\u001b[0;32m   1269\u001b[0m     feature_counter \u001b[38;5;241m=\u001b[39m {}\n\u001b[1;32m-> 1270\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m feature \u001b[38;5;129;01min\u001b[39;00m \u001b[43manalyze\u001b[49m\u001b[43m(\u001b[49m\u001b[43mdoc\u001b[49m\u001b[43m)\u001b[49m:\n\u001b[0;32m   1271\u001b[0m         \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m   1272\u001b[0m             feature_idx \u001b[38;5;241m=\u001b[39m vocabulary[feature]\n",
      "File \u001b[1;32mc:\\Users\\light\\anaconda3\\lib\\site-packages\\sklearn\\feature_extraction\\text.py:117\u001b[0m, in \u001b[0;36m_analyze\u001b[1;34m(doc, analyzer, tokenizer, ngrams, preprocessor, decoder, stop_words)\u001b[0m\n\u001b[0;32m    115\u001b[0m             doc \u001b[38;5;241m=\u001b[39m ngrams(doc, stop_words)\n\u001b[0;32m    116\u001b[0m         \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m--> 117\u001b[0m             doc \u001b[38;5;241m=\u001b[39m \u001b[43mngrams\u001b[49m\u001b[43m(\u001b[49m\u001b[43mdoc\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    118\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m doc\n",
      "File \u001b[1;32mc:\\Users\\light\\anaconda3\\lib\\site-packages\\sklearn\\feature_extraction\\text.py:270\u001b[0m, in \u001b[0;36m_VectorizerMixin._word_ngrams\u001b[1;34m(self, tokens, stop_words)\u001b[0m\n\u001b[0;32m    268\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m n \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mrange\u001b[39m(min_n, \u001b[38;5;28mmin\u001b[39m(max_n \u001b[38;5;241m+\u001b[39m \u001b[38;5;241m1\u001b[39m, n_original_tokens \u001b[38;5;241m+\u001b[39m \u001b[38;5;241m1\u001b[39m)):\n\u001b[0;32m    269\u001b[0m         \u001b[38;5;28;01mfor\u001b[39;00m i \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mrange\u001b[39m(n_original_tokens \u001b[38;5;241m-\u001b[39m n \u001b[38;5;241m+\u001b[39m \u001b[38;5;241m1\u001b[39m):\n\u001b[1;32m--> 270\u001b[0m             \u001b[43mtokens_append\u001b[49m\u001b[43m(\u001b[49m\u001b[43mspace_join\u001b[49m\u001b[43m(\u001b[49m\u001b[43moriginal_tokens\u001b[49m\u001b[43m[\u001b[49m\u001b[43mi\u001b[49m\u001b[43m \u001b[49m\u001b[43m:\u001b[49m\u001b[43m \u001b[49m\u001b[43mi\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m+\u001b[39;49m\u001b[43m \u001b[49m\u001b[43mn\u001b[49m\u001b[43m]\u001b[49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    272\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m tokens\n",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "%%time\n",
    "tfidf_vectorizer = TfidfVectorizer(ngram_range=(1, 1))\n",
    "\n",
    "# Применение TF-IDF к текстовым данным\n",
    "tfidf_matrix = tfidf_vectorizer.fit_transform(data['processed_prompt'])\n",
    "\n",
    "tfidf_matrix = normalize(tfidf_matrix, norm='l2', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: total: 531 ms\n",
      "Wall time: 248 ms\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PC1_3d</th>\n",
       "      <th>PC2_3d</th>\n",
       "      <th>PC3_3d</th>\n",
       "      <th>prompt_length</th>\n",
       "      <th>prompt_unique</th>\n",
       "      <th>prompt_{</th>\n",
       "      <th>prompt_}</th>\n",
       "      <th>prompt_;</th>\n",
       "      <th>prompt_&lt;</th>\n",
       "      <th>prompt_&gt;</th>\n",
       "      <th>prompt_#</th>\n",
       "      <th>prompt_:</th>\n",
       "      <th>prompt_&amp;</th>\n",
       "      <th>prompt_nil</th>\n",
       "      <th>prompt_#include</th>\n",
       "      <th>prompt_import</th>\n",
       "      <th>prompt_java</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.660227</td>\n",
       "      <td>-0.429241</td>\n",
       "      <td>-0.058012</td>\n",
       "      <td>-0.552438</td>\n",
       "      <td>-0.322668</td>\n",
       "      <td>-0.583318</td>\n",
       "      <td>-0.569906</td>\n",
       "      <td>-0.562931</td>\n",
       "      <td>-0.206516</td>\n",
       "      <td>-0.519429</td>\n",
       "      <td>-0.325307</td>\n",
       "      <td>-0.729119</td>\n",
       "      <td>1.448684</td>\n",
       "      <td>-0.010529</td>\n",
       "      <td>-0.279701</td>\n",
       "      <td>-0.597592</td>\n",
       "      <td>-0.309105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.594210</td>\n",
       "      <td>0.444247</td>\n",
       "      <td>-0.538410</td>\n",
       "      <td>-0.217155</td>\n",
       "      <td>-0.308665</td>\n",
       "      <td>-0.773261</td>\n",
       "      <td>-0.729447</td>\n",
       "      <td>-0.562931</td>\n",
       "      <td>-0.479397</td>\n",
       "      <td>-0.519429</td>\n",
       "      <td>-0.325307</td>\n",
       "      <td>-0.320404</td>\n",
       "      <td>-0.506400</td>\n",
       "      <td>-0.342431</td>\n",
       "      <td>-0.279701</td>\n",
       "      <td>-0.209537</td>\n",
       "      <td>-0.309105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.113586</td>\n",
       "      <td>-0.977653</td>\n",
       "      <td>-0.020422</td>\n",
       "      <td>-0.168538</td>\n",
       "      <td>0.699551</td>\n",
       "      <td>0.619651</td>\n",
       "      <td>0.674510</td>\n",
       "      <td>-0.350309</td>\n",
       "      <td>0.011789</td>\n",
       "      <td>-0.186250</td>\n",
       "      <td>-0.325307</td>\n",
       "      <td>-0.171781</td>\n",
       "      <td>2.705524</td>\n",
       "      <td>-0.010529</td>\n",
       "      <td>-0.279701</td>\n",
       "      <td>-0.597592</td>\n",
       "      <td>-0.309105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.848903</td>\n",
       "      <td>0.255661</td>\n",
       "      <td>1.615738</td>\n",
       "      <td>-0.582666</td>\n",
       "      <td>-0.798770</td>\n",
       "      <td>-0.583318</td>\n",
       "      <td>-0.601814</td>\n",
       "      <td>0.712800</td>\n",
       "      <td>-0.479397</td>\n",
       "      <td>0.396813</td>\n",
       "      <td>-0.325307</td>\n",
       "      <td>-0.673385</td>\n",
       "      <td>-0.506400</td>\n",
       "      <td>-0.342431</td>\n",
       "      <td>-0.279701</td>\n",
       "      <td>-0.597592</td>\n",
       "      <td>-0.309105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.312973</td>\n",
       "      <td>0.479682</td>\n",
       "      <td>0.992593</td>\n",
       "      <td>-1.063297</td>\n",
       "      <td>-0.994812</td>\n",
       "      <td>-0.773261</td>\n",
       "      <td>-0.761355</td>\n",
       "      <td>-0.562931</td>\n",
       "      <td>-0.479397</td>\n",
       "      <td>-0.477782</td>\n",
       "      <td>-0.325307</td>\n",
       "      <td>-0.673385</td>\n",
       "      <td>-0.506400</td>\n",
       "      <td>-0.342431</td>\n",
       "      <td>-0.279701</td>\n",
       "      <td>-0.403564</td>\n",
       "      <td>-0.309105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6657</th>\n",
       "      <td>-0.197924</td>\n",
       "      <td>0.574632</td>\n",
       "      <td>1.310852</td>\n",
       "      <td>-0.949185</td>\n",
       "      <td>-1.015817</td>\n",
       "      <td>-0.425033</td>\n",
       "      <td>-0.474182</td>\n",
       "      <td>-0.562931</td>\n",
       "      <td>-0.151940</td>\n",
       "      <td>0.105281</td>\n",
       "      <td>-0.325307</td>\n",
       "      <td>-0.654807</td>\n",
       "      <td>-0.506400</td>\n",
       "      <td>-0.342431</td>\n",
       "      <td>-0.279701</td>\n",
       "      <td>-0.306550</td>\n",
       "      <td>-0.309105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6658</th>\n",
       "      <td>-0.798751</td>\n",
       "      <td>0.027933</td>\n",
       "      <td>0.561856</td>\n",
       "      <td>2.335883</td>\n",
       "      <td>1.049626</td>\n",
       "      <td>2.645705</td>\n",
       "      <td>2.684721</td>\n",
       "      <td>2.272027</td>\n",
       "      <td>-0.479397</td>\n",
       "      <td>0.355165</td>\n",
       "      <td>-0.325307</td>\n",
       "      <td>2.689225</td>\n",
       "      <td>-0.227102</td>\n",
       "      <td>-0.342431</td>\n",
       "      <td>-0.279701</td>\n",
       "      <td>-0.209537</td>\n",
       "      <td>-0.309105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6659</th>\n",
       "      <td>1.211024</td>\n",
       "      <td>-0.924240</td>\n",
       "      <td>0.000028</td>\n",
       "      <td>-0.545888</td>\n",
       "      <td>-0.098620</td>\n",
       "      <td>0.081481</td>\n",
       "      <td>0.132072</td>\n",
       "      <td>-0.562931</td>\n",
       "      <td>-0.424821</td>\n",
       "      <td>-0.352840</td>\n",
       "      <td>-0.325307</td>\n",
       "      <td>-0.469028</td>\n",
       "      <td>-0.227102</td>\n",
       "      <td>-0.342431</td>\n",
       "      <td>-0.279701</td>\n",
       "      <td>-0.597592</td>\n",
       "      <td>-0.309105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6660</th>\n",
       "      <td>1.359338</td>\n",
       "      <td>1.885335</td>\n",
       "      <td>-1.946448</td>\n",
       "      <td>-0.591987</td>\n",
       "      <td>-0.665742</td>\n",
       "      <td>-0.646632</td>\n",
       "      <td>-0.761355</td>\n",
       "      <td>0.452929</td>\n",
       "      <td>-0.206516</td>\n",
       "      <td>-0.311193</td>\n",
       "      <td>-0.325307</td>\n",
       "      <td>-0.766275</td>\n",
       "      <td>-0.506400</td>\n",
       "      <td>-0.342431</td>\n",
       "      <td>-0.279701</td>\n",
       "      <td>2.312825</td>\n",
       "      <td>6.797554</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6661</th>\n",
       "      <td>-0.662310</td>\n",
       "      <td>0.312985</td>\n",
       "      <td>-0.150783</td>\n",
       "      <td>-0.660756</td>\n",
       "      <td>-0.840779</td>\n",
       "      <td>-0.741604</td>\n",
       "      <td>-0.697538</td>\n",
       "      <td>-0.562931</td>\n",
       "      <td>-0.479397</td>\n",
       "      <td>-0.519429</td>\n",
       "      <td>-0.212420</td>\n",
       "      <td>-0.394716</td>\n",
       "      <td>-0.506400</td>\n",
       "      <td>-0.342431</td>\n",
       "      <td>-0.279701</td>\n",
       "      <td>-0.209537</td>\n",
       "      <td>-0.309105</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>6662 rows × 17 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        PC1_3d    PC2_3d    PC3_3d  prompt_length  prompt_unique  prompt_{  \\\n",
       "0    -0.660227 -0.429241 -0.058012      -0.552438      -0.322668 -0.583318   \n",
       "1    -0.594210  0.444247 -0.538410      -0.217155      -0.308665 -0.773261   \n",
       "2     1.113586 -0.977653 -0.020422      -0.168538       0.699551  0.619651   \n",
       "3    -0.848903  0.255661  1.615738      -0.582666      -0.798770 -0.583318   \n",
       "4    -0.312973  0.479682  0.992593      -1.063297      -0.994812 -0.773261   \n",
       "...        ...       ...       ...            ...            ...       ...   \n",
       "6657 -0.197924  0.574632  1.310852      -0.949185      -1.015817 -0.425033   \n",
       "6658 -0.798751  0.027933  0.561856       2.335883       1.049626  2.645705   \n",
       "6659  1.211024 -0.924240  0.000028      -0.545888      -0.098620  0.081481   \n",
       "6660  1.359338  1.885335 -1.946448      -0.591987      -0.665742 -0.646632   \n",
       "6661 -0.662310  0.312985 -0.150783      -0.660756      -0.840779 -0.741604   \n",
       "\n",
       "      prompt_}  prompt_;  prompt_<  prompt_>  prompt_#  prompt_:  prompt_&  \\\n",
       "0    -0.569906 -0.562931 -0.206516 -0.519429 -0.325307 -0.729119  1.448684   \n",
       "1    -0.729447 -0.562931 -0.479397 -0.519429 -0.325307 -0.320404 -0.506400   \n",
       "2     0.674510 -0.350309  0.011789 -0.186250 -0.325307 -0.171781  2.705524   \n",
       "3    -0.601814  0.712800 -0.479397  0.396813 -0.325307 -0.673385 -0.506400   \n",
       "4    -0.761355 -0.562931 -0.479397 -0.477782 -0.325307 -0.673385 -0.506400   \n",
       "...        ...       ...       ...       ...       ...       ...       ...   \n",
       "6657 -0.474182 -0.562931 -0.151940  0.105281 -0.325307 -0.654807 -0.506400   \n",
       "6658  2.684721  2.272027 -0.479397  0.355165 -0.325307  2.689225 -0.227102   \n",
       "6659  0.132072 -0.562931 -0.424821 -0.352840 -0.325307 -0.469028 -0.227102   \n",
       "6660 -0.761355  0.452929 -0.206516 -0.311193 -0.325307 -0.766275 -0.506400   \n",
       "6661 -0.697538 -0.562931 -0.479397 -0.519429 -0.212420 -0.394716 -0.506400   \n",
       "\n",
       "      prompt_nil  prompt_#include  prompt_import  prompt_java  \n",
       "0      -0.010529        -0.279701      -0.597592    -0.309105  \n",
       "1      -0.342431        -0.279701      -0.209537    -0.309105  \n",
       "2      -0.010529        -0.279701      -0.597592    -0.309105  \n",
       "3      -0.342431        -0.279701      -0.597592    -0.309105  \n",
       "4      -0.342431        -0.279701      -0.403564    -0.309105  \n",
       "...          ...              ...            ...          ...  \n",
       "6657   -0.342431        -0.279701      -0.306550    -0.309105  \n",
       "6658   -0.342431        -0.279701      -0.209537    -0.309105  \n",
       "6659   -0.342431        -0.279701      -0.597592    -0.309105  \n",
       "6660   -0.342431        -0.279701       2.312825     6.797554  \n",
       "6661   -0.342431        -0.279701      -0.209537    -0.309105  \n",
       "\n",
       "[6662 rows x 17 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%time\n",
    "svd = TruncatedSVD(n_components=3, random_state=42)\n",
    "PCs_3d = pd.DataFrame(svd.fit_transform(\n",
    "    tfidf_matrix))\n",
    "PCs_3d.columns = [\"PC1_3d\", \"PC2_3d\", \"PC3_3d\"]\n",
    "\n",
    "new_features = pd.read_csv('new_features.csv')\n",
    "\n",
    "join_df = pd.concat([PCs_3d, new_features.drop(columns=['target', 'model_response_length', 'model_response_unique'])], axis=1, join='inner')\n",
    "join_df_cols = join_df.columns\n",
    "std_scaler = StandardScaler()\n",
    "join_df = pd.DataFrame(std_scaler.fit_transform(join_df), columns=join_df_cols)\n",
    "join_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "# Catboost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Learning rate set to 0.047906\n",
      "0:\tlearn: 0.0471638\ttest: 0.0270706\tbest: 0.0270706 (0)\ttotal: 165ms\tremaining: 2m 45s\n",
      "1:\tlearn: 0.1404110\ttest: 0.1190624\tbest: 0.1190624 (1)\ttotal: 169ms\tremaining: 1m 24s\n",
      "2:\tlearn: 0.1306480\ttest: 0.1140235\tbest: 0.1190624 (1)\ttotal: 173ms\tremaining: 57.4s\n",
      "3:\tlearn: 0.1525534\ttest: 0.1295419\tbest: 0.1295419 (3)\ttotal: 176ms\tremaining: 43.8s\n",
      "4:\tlearn: 0.1108550\ttest: 0.0868705\tbest: 0.1295419 (3)\ttotal: 179ms\tremaining: 35.6s\n",
      "5:\tlearn: 0.1088443\ttest: 0.0729693\tbest: 0.1295419 (3)\ttotal: 182ms\tremaining: 30.1s\n",
      "6:\tlearn: 0.0927499\ttest: 0.0502487\tbest: 0.1295419 (3)\ttotal: 185ms\tremaining: 26.2s\n",
      "7:\tlearn: 0.1063573\ttest: 0.0646549\tbest: 0.1295419 (3)\ttotal: 191ms\tremaining: 23.6s\n",
      "8:\tlearn: 0.1084775\ttest: 0.0647291\tbest: 0.1295419 (3)\ttotal: 195ms\tremaining: 21.4s\n",
      "9:\tlearn: 0.1013811\ttest: 0.0732195\tbest: 0.1295419 (3)\ttotal: 198ms\tremaining: 19.6s\n",
      "10:\tlearn: 0.1012327\ttest: 0.0703712\tbest: 0.1295419 (3)\ttotal: 200ms\tremaining: 18s\n",
      "11:\tlearn: 0.1103995\ttest: 0.0872650\tbest: 0.1295419 (3)\ttotal: 204ms\tremaining: 16.8s\n",
      "12:\tlearn: 0.1083510\ttest: 0.0874636\tbest: 0.1295419 (3)\ttotal: 208ms\tremaining: 15.8s\n",
      "13:\tlearn: 0.1139303\ttest: 0.0985647\tbest: 0.1295419 (3)\ttotal: 211ms\tremaining: 14.8s\n",
      "14:\tlearn: 0.1219441\ttest: 0.0982313\tbest: 0.1295419 (3)\ttotal: 214ms\tremaining: 14s\n",
      "15:\tlearn: 0.1240309\ttest: 0.1037756\tbest: 0.1295419 (3)\ttotal: 217ms\tremaining: 13.3s\n",
      "16:\tlearn: 0.1260763\ttest: 0.1036590\tbest: 0.1295419 (3)\ttotal: 220ms\tremaining: 12.7s\n",
      "17:\tlearn: 0.1336924\ttest: 0.1010065\tbest: 0.1295419 (3)\ttotal: 224ms\tremaining: 12.2s\n",
      "18:\tlearn: 0.1337696\ttest: 0.1117947\tbest: 0.1295419 (3)\ttotal: 226ms\tremaining: 11.7s\n",
      "19:\tlearn: 0.1326741\ttest: 0.1034265\tbest: 0.1295419 (3)\ttotal: 229ms\tremaining: 11.2s\n",
      "20:\tlearn: 0.1311538\ttest: 0.1090502\tbest: 0.1295419 (3)\ttotal: 232ms\tremaining: 10.8s\n",
      "21:\tlearn: 0.1323678\ttest: 0.1172654\tbest: 0.1295419 (3)\ttotal: 236ms\tremaining: 10.5s\n",
      "22:\tlearn: 0.1397761\ttest: 0.1254258\tbest: 0.1295419 (3)\ttotal: 239ms\tremaining: 10.2s\n",
      "23:\tlearn: 0.1444282\ttest: 0.1228486\tbest: 0.1295419 (3)\ttotal: 242ms\tremaining: 9.85s\n",
      "24:\tlearn: 0.1511541\ttest: 0.1330893\tbest: 0.1330893 (24)\ttotal: 245ms\tremaining: 9.56s\n",
      "25:\tlearn: 0.1570547\ttest: 0.1357710\tbest: 0.1357710 (25)\ttotal: 248ms\tremaining: 9.29s\n",
      "26:\tlearn: 0.1568170\ttest: 0.1384467\tbest: 0.1384467 (26)\ttotal: 252ms\tremaining: 9.08s\n",
      "27:\tlearn: 0.1648657\ttest: 0.1275669\tbest: 0.1384467 (26)\ttotal: 255ms\tremaining: 8.85s\n",
      "28:\tlearn: 0.1648657\ttest: 0.1277082\tbest: 0.1384467 (26)\ttotal: 258ms\tremaining: 8.63s\n",
      "29:\tlearn: 0.1647257\ttest: 0.1275669\tbest: 0.1384467 (26)\ttotal: 261ms\tremaining: 8.43s\n",
      "30:\tlearn: 0.1700014\ttest: 0.1274259\tbest: 0.1384467 (26)\ttotal: 264ms\tremaining: 8.26s\n",
      "31:\tlearn: 0.1746708\ttest: 0.1354719\tbest: 0.1384467 (26)\ttotal: 269ms\tremaining: 8.13s\n",
      "32:\tlearn: 0.1792688\ttest: 0.1381421\tbest: 0.1384467 (26)\ttotal: 272ms\tremaining: 7.96s\n",
      "33:\tlearn: 0.1793192\ttest: 0.1379902\tbest: 0.1384467 (26)\ttotal: 275ms\tremaining: 7.81s\n",
      "34:\tlearn: 0.1805565\ttest: 0.1406518\tbest: 0.1406518 (34)\ttotal: 278ms\tremaining: 7.66s\n",
      "35:\tlearn: 0.1797867\ttest: 0.1482770\tbest: 0.1482770 (35)\ttotal: 282ms\tremaining: 7.54s\n",
      "36:\tlearn: 0.1806072\ttest: 0.1453207\tbest: 0.1482770 (35)\ttotal: 286ms\tremaining: 7.43s\n",
      "37:\tlearn: 0.1859459\ttest: 0.1505815\tbest: 0.1505815 (37)\ttotal: 288ms\tremaining: 7.3s\n",
      "38:\tlearn: 0.1937469\ttest: 0.1505815\tbest: 0.1505815 (37)\ttotal: 291ms\tremaining: 7.18s\n",
      "39:\tlearn: 0.1932477\ttest: 0.1507456\tbest: 0.1507456 (39)\ttotal: 294ms\tremaining: 7.07s\n",
      "40:\tlearn: 0.2045880\ttest: 0.1554818\tbest: 0.1554818 (40)\ttotal: 297ms\tremaining: 6.96s\n",
      "41:\tlearn: 0.2065473\ttest: 0.1554818\tbest: 0.1554818 (40)\ttotal: 301ms\tremaining: 6.87s\n",
      "42:\tlearn: 0.2128199\ttest: 0.1629285\tbest: 0.1629285 (42)\ttotal: 304ms\tremaining: 6.77s\n",
      "43:\tlearn: 0.2121714\ttest: 0.1629285\tbest: 0.1629285 (42)\ttotal: 307ms\tremaining: 6.67s\n",
      "44:\tlearn: 0.2164053\ttest: 0.1603398\tbest: 0.1629285 (42)\ttotal: 311ms\tremaining: 6.59s\n",
      "45:\tlearn: 0.2169909\ttest: 0.1603398\tbest: 0.1629285 (42)\ttotal: 313ms\tremaining: 6.5s\n",
      "46:\tlearn: 0.2219557\ttest: 0.1655117\tbest: 0.1655117 (46)\ttotal: 318ms\tremaining: 6.45s\n",
      "47:\tlearn: 0.2286167\ttest: 0.1629285\tbest: 0.1655117 (46)\ttotal: 321ms\tremaining: 6.36s\n",
      "48:\tlearn: 0.2316069\ttest: 0.1655117\tbest: 0.1655117 (46)\ttotal: 324ms\tremaining: 6.28s\n",
      "49:\tlearn: 0.2323678\ttest: 0.1656900\tbest: 0.1656900 (49)\ttotal: 327ms\tremaining: 6.21s\n",
      "50:\tlearn: 0.2346478\ttest: 0.1656900\tbest: 0.1656900 (49)\ttotal: 329ms\tremaining: 6.13s\n",
      "51:\tlearn: 0.2371108\ttest: 0.1629285\tbest: 0.1656900 (49)\ttotal: 333ms\tremaining: 6.08s\n",
      "52:\tlearn: 0.2380493\ttest: 0.1653337\tbest: 0.1656900 (49)\ttotal: 336ms\tremaining: 6.01s\n",
      "53:\tlearn: 0.2442618\ttest: 0.1704782\tbest: 0.1704782 (53)\ttotal: 339ms\tremaining: 5.95s\n",
      "54:\tlearn: 0.2453145\ttest: 0.1704782\tbest: 0.1704782 (53)\ttotal: 342ms\tremaining: 5.88s\n",
      "55:\tlearn: 0.2502354\ttest: 0.1702954\tbest: 0.1704782 (53)\ttotal: 345ms\tremaining: 5.82s\n",
      "56:\tlearn: 0.2502354\ttest: 0.1702954\tbest: 0.1704782 (53)\ttotal: 349ms\tremaining: 5.77s\n",
      "57:\tlearn: 0.2487229\ttest: 0.1724875\tbest: 0.1724875 (57)\ttotal: 352ms\tremaining: 5.71s\n",
      "58:\tlearn: 0.2511412\ttest: 0.1724875\tbest: 0.1724875 (57)\ttotal: 356ms\tremaining: 5.68s\n",
      "59:\tlearn: 0.2547906\ttest: 0.1724875\tbest: 0.1724875 (57)\ttotal: 360ms\tremaining: 5.64s\n",
      "60:\tlearn: 0.2565759\ttest: 0.1724875\tbest: 0.1724875 (57)\ttotal: 364ms\tremaining: 5.61s\n",
      "61:\tlearn: 0.2581502\ttest: 0.1724875\tbest: 0.1724875 (57)\ttotal: 367ms\tremaining: 5.55s\n",
      "62:\tlearn: 0.2558898\ttest: 0.1746655\tbest: 0.1746655 (62)\ttotal: 371ms\tremaining: 5.52s\n",
      "63:\tlearn: 0.2600654\ttest: 0.1746655\tbest: 0.1746655 (62)\ttotal: 374ms\tremaining: 5.47s\n",
      "64:\tlearn: 0.2599960\ttest: 0.1746655\tbest: 0.1746655 (62)\ttotal: 377ms\tremaining: 5.42s\n",
      "65:\tlearn: 0.2604716\ttest: 0.1746655\tbest: 0.1746655 (62)\ttotal: 381ms\tremaining: 5.39s\n",
      "66:\tlearn: 0.2650921\ttest: 0.1746655\tbest: 0.1746655 (62)\ttotal: 384ms\tremaining: 5.35s\n",
      "67:\tlearn: 0.2648808\ttest: 0.1741092\tbest: 0.1746655 (62)\ttotal: 387ms\tremaining: 5.3s\n",
      "68:\tlearn: 0.2661012\ttest: 0.1741092\tbest: 0.1746655 (62)\ttotal: 390ms\tremaining: 5.26s\n",
      "69:\tlearn: 0.2697547\ttest: 0.1811169\tbest: 0.1811169 (69)\ttotal: 393ms\tremaining: 5.22s\n",
      "70:\tlearn: 0.2696832\ttest: 0.1813082\tbest: 0.1813082 (70)\ttotal: 397ms\tremaining: 5.19s\n",
      "71:\tlearn: 0.2779409\ttest: 0.1861307\tbest: 0.1861307 (71)\ttotal: 401ms\tremaining: 5.17s\n",
      "72:\tlearn: 0.2773378\ttest: 0.1861307\tbest: 0.1861307 (71)\ttotal: 404ms\tremaining: 5.13s\n",
      "73:\tlearn: 0.2797482\ttest: 0.1838201\tbest: 0.1861307 (71)\ttotal: 408ms\tremaining: 5.1s\n",
      "74:\tlearn: 0.2816268\ttest: 0.1882338\tbest: 0.1882338 (74)\ttotal: 411ms\tremaining: 5.07s\n",
      "75:\tlearn: 0.2823766\ttest: 0.1857396\tbest: 0.1882338 (74)\ttotal: 415ms\tremaining: 5.04s\n",
      "76:\tlearn: 0.2865012\ttest: 0.1884315\tbest: 0.1884315 (76)\ttotal: 418ms\tremaining: 5.01s\n",
      "77:\tlearn: 0.2870997\ttest: 0.1884315\tbest: 0.1884315 (76)\ttotal: 421ms\tremaining: 4.97s\n",
      "78:\tlearn: 0.2882202\ttest: 0.1909228\tbest: 0.1909228 (78)\ttotal: 423ms\tremaining: 4.94s\n",
      "79:\tlearn: 0.2883718\ttest: 0.1934089\tbest: 0.1934089 (79)\ttotal: 427ms\tremaining: 4.92s\n",
      "80:\tlearn: 0.2882960\ttest: 0.1934089\tbest: 0.1934089 (79)\ttotal: 432ms\tremaining: 4.89s\n",
      "81:\tlearn: 0.2882960\ttest: 0.1958898\tbest: 0.1958898 (81)\ttotal: 434ms\tremaining: 4.86s\n",
      "82:\tlearn: 0.2918772\ttest: 0.2010461\tbest: 0.2010461 (82)\ttotal: 437ms\tremaining: 4.83s\n",
      "83:\tlearn: 0.2917240\ttest: 0.2008360\tbest: 0.2010461 (82)\ttotal: 440ms\tremaining: 4.8s\n",
      "84:\tlearn: 0.2892630\ttest: 0.2008360\tbest: 0.2010461 (82)\ttotal: 445ms\tremaining: 4.79s\n",
      "85:\tlearn: 0.2928380\ttest: 0.2010461\tbest: 0.2010461 (82)\ttotal: 448ms\tremaining: 4.76s\n",
      "86:\tlearn: 0.2957311\ttest: 0.2030894\tbest: 0.2030894 (86)\ttotal: 451ms\tremaining: 4.73s\n",
      "87:\tlearn: 0.2964017\ttest: 0.2006263\tbest: 0.2030894 (86)\ttotal: 453ms\tremaining: 4.7s\n",
      "88:\tlearn: 0.2997193\ttest: 0.2077838\tbest: 0.2077838 (88)\ttotal: 456ms\tremaining: 4.67s\n",
      "89:\tlearn: 0.3019225\ttest: 0.2053333\tbest: 0.2077838 (88)\ttotal: 460ms\tremaining: 4.65s\n",
      "90:\tlearn: 0.3024327\ttest: 0.2077838\tbest: 0.2077838 (88)\ttotal: 463ms\tremaining: 4.63s\n",
      "91:\tlearn: 0.3059602\ttest: 0.2124490\tbest: 0.2124490 (91)\ttotal: 466ms\tremaining: 4.6s\n",
      "92:\tlearn: 0.3084906\ttest: 0.2122291\tbest: 0.2124490 (91)\ttotal: 469ms\tremaining: 4.57s\n",
      "93:\tlearn: 0.3086508\ttest: 0.2173098\tbest: 0.2173098 (93)\ttotal: 472ms\tremaining: 4.55s\n",
      "94:\tlearn: 0.3094991\ttest: 0.2197327\tbest: 0.2197327 (94)\ttotal: 476ms\tremaining: 4.53s\n",
      "95:\tlearn: 0.3121726\ttest: 0.2197327\tbest: 0.2197327 (94)\ttotal: 479ms\tremaining: 4.51s\n",
      "96:\tlearn: 0.3136605\ttest: 0.2175348\tbest: 0.2197327 (94)\ttotal: 482ms\tremaining: 4.48s\n",
      "97:\tlearn: 0.3155780\ttest: 0.2195059\tbest: 0.2197327 (94)\ttotal: 485ms\tremaining: 4.46s\n",
      "98:\tlearn: 0.3178961\ttest: 0.2216930\tbest: 0.2216930 (98)\ttotal: 488ms\tremaining: 4.44s\n",
      "99:\tlearn: 0.3190533\ttest: 0.2216930\tbest: 0.2216930 (98)\ttotal: 491ms\tremaining: 4.42s\n",
      "100:\tlearn: 0.3198797\ttest: 0.2360693\tbest: 0.2360693 (100)\ttotal: 494ms\tremaining: 4.4s\n",
      "101:\tlearn: 0.3197151\ttest: 0.2334464\tbest: 0.2360693 (100)\ttotal: 497ms\tremaining: 4.38s\n",
      "102:\tlearn: 0.3199620\ttest: 0.2355871\tbest: 0.2360693 (100)\ttotal: 500ms\tremaining: 4.36s\n",
      "103:\tlearn: 0.3200444\ttest: 0.2332076\tbest: 0.2360693 (100)\ttotal: 503ms\tremaining: 4.33s\n",
      "104:\tlearn: 0.3211988\ttest: 0.2332076\tbest: 0.2360693 (100)\ttotal: 507ms\tremaining: 4.32s\n",
      "105:\tlearn: 0.3211161\ttest: 0.2332076\tbest: 0.2360693 (100)\ttotal: 510ms\tremaining: 4.3s\n",
      "106:\tlearn: 0.3197974\ttest: 0.2310599\tbest: 0.2360693 (100)\ttotal: 513ms\tremaining: 4.28s\n",
      "107:\tlearn: 0.3210335\ttest: 0.2426962\tbest: 0.2426962 (107)\ttotal: 516ms\tremaining: 4.26s\n",
      "108:\tlearn: 0.3239961\ttest: 0.2450563\tbest: 0.2450563 (108)\ttotal: 520ms\tremaining: 4.25s\n",
      "109:\tlearn: 0.3248957\ttest: 0.2431915\tbest: 0.2450563 (108)\ttotal: 524ms\tremaining: 4.24s\n",
      "110:\tlearn: 0.3259601\ttest: 0.2426962\tbest: 0.2450563 (108)\ttotal: 527ms\tremaining: 4.22s\n",
      "111:\tlearn: 0.3291616\ttest: 0.2495087\tbest: 0.2495087 (111)\ttotal: 530ms\tremaining: 4.2s\n",
      "112:\tlearn: 0.3309629\ttest: 0.2547068\tbest: 0.2547068 (112)\ttotal: 534ms\tremaining: 4.19s\n",
      "113:\tlearn: 0.3303909\ttest: 0.2570452\tbest: 0.2570452 (113)\ttotal: 537ms\tremaining: 4.17s\n",
      "114:\tlearn: 0.3297341\ttest: 0.2617076\tbest: 0.2617076 (114)\ttotal: 540ms\tremaining: 4.16s\n",
      "115:\tlearn: 0.3308781\ttest: 0.2593788\tbest: 0.2617076 (114)\ttotal: 543ms\tremaining: 4.14s\n",
      "116:\tlearn: 0.3314497\ttest: 0.2614433\tbest: 0.2617076 (114)\ttotal: 547ms\tremaining: 4.13s\n",
      "117:\tlearn: 0.3342178\ttest: 0.2617076\tbest: 0.2617076 (114)\ttotal: 551ms\tremaining: 4.12s\n",
      "118:\tlearn: 0.3376317\ttest: 0.2614433\tbest: 0.2617076 (114)\ttotal: 555ms\tremaining: 4.11s\n",
      "119:\tlearn: 0.3376317\ttest: 0.2614433\tbest: 0.2617076 (114)\ttotal: 558ms\tremaining: 4.09s\n",
      "120:\tlearn: 0.3369773\ttest: 0.2640318\tbest: 0.2640318 (120)\ttotal: 561ms\tremaining: 4.07s\n",
      "121:\tlearn: 0.3377179\ttest: 0.2591165\tbest: 0.2640318 (120)\ttotal: 564ms\tremaining: 4.06s\n",
      "122:\tlearn: 0.3381996\ttest: 0.2614433\tbest: 0.2640318 (120)\ttotal: 568ms\tremaining: 4.05s\n",
      "123:\tlearn: 0.3404686\ttest: 0.2614433\tbest: 0.2640318 (120)\ttotal: 571ms\tremaining: 4.03s\n",
      "124:\tlearn: 0.3418189\ttest: 0.2634995\tbest: 0.2640318 (120)\ttotal: 574ms\tremaining: 4.02s\n",
      "125:\tlearn: 0.3422099\ttest: 0.2632341\tbest: 0.2640318 (120)\ttotal: 577ms\tremaining: 4s\n",
      "126:\tlearn: 0.3422969\ttest: 0.2655474\tbest: 0.2655474 (126)\ttotal: 580ms\tremaining: 3.98s\n",
      "127:\tlearn: 0.3433388\ttest: 0.2655474\tbest: 0.2655474 (126)\ttotal: 583ms\tremaining: 3.97s\n",
      "128:\tlearn: 0.3449425\ttest: 0.2655474\tbest: 0.2655474 (126)\ttotal: 586ms\tremaining: 3.96s\n",
      "129:\tlearn: 0.3443792\ttest: 0.2673187\tbest: 0.2673187 (129)\ttotal: 589ms\tremaining: 3.94s\n",
      "130:\tlearn: 0.3455934\ttest: 0.2673187\tbest: 0.2673187 (129)\ttotal: 592ms\tremaining: 3.93s\n",
      "131:\tlearn: 0.3456812\ttest: 0.2673187\tbest: 0.2673187 (129)\ttotal: 596ms\tremaining: 3.92s\n",
      "132:\tlearn: 0.3490551\ttest: 0.2673187\tbest: 0.2673187 (129)\ttotal: 599ms\tremaining: 3.9s\n",
      "133:\tlearn: 0.3524188\ttest: 0.2690793\tbest: 0.2690793 (133)\ttotal: 602ms\tremaining: 3.89s\n",
      "134:\tlearn: 0.3518589\ttest: 0.2713706\tbest: 0.2713706 (134)\ttotal: 605ms\tremaining: 3.88s\n",
      "135:\tlearn: 0.3518589\ttest: 0.2688105\tbest: 0.2713706 (134)\ttotal: 608ms\tremaining: 3.86s\n",
      "136:\tlearn: 0.3529784\ttest: 0.2688105\tbest: 0.2713706 (134)\ttotal: 611ms\tremaining: 3.85s\n",
      "137:\tlearn: 0.3552141\ttest: 0.2733844\tbest: 0.2733844 (137)\ttotal: 614ms\tremaining: 3.83s\n",
      "138:\tlearn: 0.3557723\ttest: 0.2736573\tbest: 0.2736573 (138)\ttotal: 617ms\tremaining: 3.82s\n",
      "139:\tlearn: 0.3563302\ttest: 0.2736573\tbest: 0.2736573 (138)\ttotal: 620ms\tremaining: 3.81s\n",
      "140:\tlearn: 0.3568878\ttest: 0.2736573\tbest: 0.2736573 (138)\ttotal: 623ms\tremaining: 3.79s\n",
      "141:\tlearn: 0.3554838\ttest: 0.2736573\tbest: 0.2736573 (138)\ttotal: 626ms\tremaining: 3.78s\n",
      "142:\tlearn: 0.3553039\ttest: 0.2782170\tbest: 0.2782170 (142)\ttotal: 629ms\tremaining: 3.77s\n",
      "143:\tlearn: 0.3553039\ttest: 0.2756646\tbest: 0.2782170 (142)\ttotal: 633ms\tremaining: 3.76s\n",
      "144:\tlearn: 0.3576259\ttest: 0.2759394\tbest: 0.2782170 (142)\ttotal: 636ms\tremaining: 3.75s\n",
      "145:\tlearn: 0.3620766\ttest: 0.2776639\tbest: 0.2782170 (142)\ttotal: 639ms\tremaining: 3.74s\n",
      "146:\tlearn: 0.3625402\ttest: 0.2753903\tbest: 0.2782170 (142)\ttotal: 642ms\tremaining: 3.72s\n",
      "147:\tlearn: 0.3635576\ttest: 0.2753903\tbest: 0.2782170 (142)\ttotal: 645ms\tremaining: 3.71s\n",
      "148:\tlearn: 0.3646653\ttest: 0.2756646\tbest: 0.2782170 (142)\ttotal: 648ms\tremaining: 3.7s\n",
      "149:\tlearn: 0.3659564\ttest: 0.2799330\tbest: 0.2799330 (149)\ttotal: 651ms\tremaining: 3.69s\n",
      "150:\tlearn: 0.3666929\ttest: 0.2802112\tbest: 0.2802112 (150)\ttotal: 654ms\tremaining: 3.68s\n",
      "151:\tlearn: 0.3662323\ttest: 0.2838949\tbest: 0.2838949 (151)\ttotal: 657ms\tremaining: 3.67s\n",
      "152:\tlearn: 0.3696300\ttest: 0.2816387\tbest: 0.2838949 (151)\ttotal: 660ms\tremaining: 3.65s\n",
      "153:\tlearn: 0.3690797\ttest: 0.2813601\tbest: 0.2838949 (151)\ttotal: 664ms\tremaining: 3.65s\n",
      "154:\tlearn: 0.3694446\ttest: 0.2813601\tbest: 0.2838949 (151)\ttotal: 667ms\tremaining: 3.64s\n",
      "155:\tlearn: 0.3721906\ttest: 0.2816387\tbest: 0.2838949 (151)\ttotal: 670ms\tremaining: 3.63s\n",
      "156:\tlearn: 0.3742888\ttest: 0.2838949\tbest: 0.2838949 (151)\ttotal: 674ms\tremaining: 3.62s\n",
      "157:\tlearn: 0.3738348\ttest: 0.2836144\tbest: 0.2838949 (151)\ttotal: 677ms\tremaining: 3.61s\n",
      "158:\tlearn: 0.3769467\ttest: 0.2878256\tbest: 0.2878256 (158)\ttotal: 680ms\tremaining: 3.6s\n",
      "159:\tlearn: 0.3759467\ttest: 0.2858641\tbest: 0.2878256 (158)\ttotal: 683ms\tremaining: 3.59s\n",
      "160:\tlearn: 0.3751173\ttest: 0.2881095\tbest: 0.2881095 (160)\ttotal: 686ms\tremaining: 3.58s\n",
      "161:\tlearn: 0.3740220\ttest: 0.2881095\tbest: 0.2881095 (160)\ttotal: 689ms\tremaining: 3.56s\n",
      "162:\tlearn: 0.3736611\ttest: 0.2881095\tbest: 0.2881095 (160)\ttotal: 692ms\tremaining: 3.56s\n",
      "163:\tlearn: 0.3747574\ttest: 0.2881095\tbest: 0.2881095 (160)\ttotal: 697ms\tremaining: 3.55s\n",
      "164:\tlearn: 0.3765698\ttest: 0.2897794\tbest: 0.2897794 (164)\ttotal: 700ms\tremaining: 3.54s\n",
      "165:\tlearn: 0.3783965\ttest: 0.2875423\tbest: 0.2897794 (164)\ttotal: 704ms\tremaining: 3.54s\n",
      "166:\tlearn: 0.3794874\ttest: 0.2894947\tbest: 0.2897794 (164)\ttotal: 707ms\tremaining: 3.53s\n",
      "167:\tlearn: 0.3792033\ttest: 0.2936639\tbest: 0.2936639 (167)\ttotal: 712ms\tremaining: 3.52s\n",
      "168:\tlearn: 0.3808366\ttest: 0.2936639\tbest: 0.2936639 (167)\ttotal: 715ms\tremaining: 3.52s\n",
      "169:\tlearn: 0.3847339\ttest: 0.2961739\tbest: 0.2961739 (169)\ttotal: 719ms\tremaining: 3.51s\n",
      "170:\tlearn: 0.3847339\ttest: 0.2939518\tbest: 0.2961739 (169)\ttotal: 722ms\tremaining: 3.5s\n",
      "171:\tlearn: 0.3850843\ttest: 0.2936639\tbest: 0.2961739 (169)\ttotal: 727ms\tremaining: 3.5s\n",
      "172:\tlearn: 0.3874417\ttest: 0.2955948\tbest: 0.2961739 (169)\ttotal: 731ms\tremaining: 3.49s\n",
      "173:\tlearn: 0.3888699\ttest: 0.2936639\tbest: 0.2961739 (169)\ttotal: 733ms\tremaining: 3.48s\n",
      "174:\tlearn: 0.3902946\ttest: 0.2936639\tbest: 0.2961739 (169)\ttotal: 736ms\tremaining: 3.47s\n",
      "175:\tlearn: 0.3910272\ttest: 0.2933766\tbest: 0.2961739 (169)\ttotal: 740ms\tremaining: 3.46s\n",
      "176:\tlearn: 0.3925451\ttest: 0.2955948\tbest: 0.2961739 (169)\ttotal: 744ms\tremaining: 3.46s\n",
      "177:\tlearn: 0.3945484\ttest: 0.2911540\tbest: 0.2961739 (169)\ttotal: 747ms\tremaining: 3.45s\n",
      "178:\tlearn: 0.3938155\ttest: 0.2933766\tbest: 0.2961739 (169)\ttotal: 750ms\tremaining: 3.44s\n",
      "179:\tlearn: 0.3958656\ttest: 0.2953061\tbest: 0.2961739 (169)\ttotal: 754ms\tremaining: 3.43s\n",
      "180:\tlearn: 0.3995151\ttest: 0.2997259\tbest: 0.2997259 (180)\ttotal: 757ms\tremaining: 3.43s\n",
      "181:\tlearn: 0.3986849\ttest: 0.2997259\tbest: 0.2997259 (180)\ttotal: 761ms\tremaining: 3.42s\n",
      "182:\tlearn: 0.3976157\ttest: 0.2997259\tbest: 0.2997259 (180)\ttotal: 764ms\tremaining: 3.41s\n",
      "183:\tlearn: 0.3988820\ttest: 0.3000183\tbest: 0.3000183 (183)\ttotal: 767ms\tremaining: 3.4s\n",
      "184:\tlearn: 0.4003857\ttest: 0.3000183\tbest: 0.3000183 (183)\ttotal: 770ms\tremaining: 3.39s\n",
      "185:\tlearn: 0.4015516\ttest: 0.3044246\tbest: 0.3044246 (185)\ttotal: 774ms\tremaining: 3.38s\n",
      "186:\tlearn: 0.4020847\ttest: 0.3066213\tbest: 0.3066213 (186)\ttotal: 777ms\tremaining: 3.38s\n",
      "187:\tlearn: 0.4034487\ttest: 0.3066213\tbest: 0.3066213 (186)\ttotal: 780ms\tremaining: 3.37s\n",
      "188:\tlearn: 0.4039814\ttest: 0.3063233\tbest: 0.3066213 (186)\ttotal: 783ms\tremaining: 3.36s\n",
      "189:\tlearn: 0.4045138\ttest: 0.3044246\tbest: 0.3066213 (186)\ttotal: 786ms\tremaining: 3.35s\n",
      "190:\tlearn: 0.4053458\ttest: 0.3044246\tbest: 0.3066213 (186)\ttotal: 789ms\tremaining: 3.34s\n",
      "191:\tlearn: 0.4070716\ttest: 0.3066213\tbest: 0.3066213 (186)\ttotal: 792ms\tremaining: 3.33s\n",
      "192:\tlearn: 0.4087636\ttest: 0.3060259\tbest: 0.3066213 (186)\ttotal: 795ms\tremaining: 3.32s\n",
      "193:\tlearn: 0.4103530\ttest: 0.3057290\tbest: 0.3066213 (186)\ttotal: 798ms\tremaining: 3.32s\n",
      "194:\tlearn: 0.4103530\ttest: 0.3054327\tbest: 0.3066213 (186)\ttotal: 801ms\tremaining: 3.31s\n",
      "195:\tlearn: 0.4109833\ttest: 0.3073203\tbest: 0.3073203 (195)\ttotal: 805ms\tremaining: 3.3s\n",
      "196:\tlearn: 0.4136263\ttest: 0.3094994\tbest: 0.3094994 (196)\ttotal: 808ms\tremaining: 3.29s\n",
      "197:\tlearn: 0.4138492\ttest: 0.3113736\tbest: 0.3113736 (197)\ttotal: 811ms\tremaining: 3.28s\n",
      "198:\tlearn: 0.4154301\ttest: 0.3113736\tbest: 0.3113736 (197)\ttotal: 816ms\tremaining: 3.28s\n",
      "199:\tlearn: 0.4168045\ttest: 0.3132407\tbest: 0.3132407 (199)\ttotal: 819ms\tremaining: 3.27s\n",
      "200:\tlearn: 0.4163807\ttest: 0.3132407\tbest: 0.3132407 (199)\ttotal: 823ms\tremaining: 3.27s\n",
      "201:\tlearn: 0.4153283\ttest: 0.3154036\tbest: 0.3154036 (201)\ttotal: 825ms\tremaining: 3.26s\n",
      "202:\tlearn: 0.4174322\ttest: 0.3147982\tbest: 0.3154036 (201)\ttotal: 829ms\tremaining: 3.25s\n",
      "203:\tlearn: 0.4185851\ttest: 0.3117404\tbest: 0.3154036 (201)\ttotal: 832ms\tremaining: 3.25s\n",
      "204:\tlearn: 0.4186876\ttest: 0.3120393\tbest: 0.3154036 (201)\ttotal: 835ms\tremaining: 3.24s\n",
      "205:\tlearn: 0.4203648\ttest: 0.3141951\tbest: 0.3154036 (201)\ttotal: 839ms\tremaining: 3.23s\n",
      "206:\tlearn: 0.4219371\ttest: 0.3141951\tbest: 0.3154036 (201)\ttotal: 842ms\tremaining: 3.23s\n",
      "207:\tlearn: 0.4208892\ttest: 0.3141951\tbest: 0.3154036 (201)\ttotal: 845ms\tremaining: 3.22s\n",
      "208:\tlearn: 0.4240299\ttest: 0.3163467\tbest: 0.3163467 (208)\ttotal: 848ms\tremaining: 3.21s\n",
      "209:\tlearn: 0.4240299\ttest: 0.3160442\tbest: 0.3163467 (208)\ttotal: 851ms\tremaining: 3.2s\n",
      "210:\tlearn: 0.4236107\ttest: 0.3160442\tbest: 0.3163467 (208)\ttotal: 855ms\tremaining: 3.19s\n",
      "211:\tlearn: 0.4246563\ttest: 0.3166497\tbest: 0.3166497 (211)\ttotal: 858ms\tremaining: 3.19s\n",
      "212:\tlearn: 0.4270573\ttest: 0.3187990\tbest: 0.3187990 (212)\ttotal: 861ms\tremaining: 3.18s\n",
      "213:\tlearn: 0.4292441\ttest: 0.3187990\tbest: 0.3187990 (212)\ttotal: 864ms\tremaining: 3.17s\n",
      "214:\tlearn: 0.4314214\ttest: 0.3209441\tbest: 0.3209441 (214)\ttotal: 867ms\tremaining: 3.17s\n",
      "215:\tlearn: 0.4313165\ttest: 0.3206376\tbest: 0.3209441 (214)\ttotal: 871ms\tremaining: 3.16s\n",
      "216:\tlearn: 0.4333880\ttest: 0.3203316\tbest: 0.3209441 (214)\ttotal: 874ms\tremaining: 3.15s\n",
      "217:\tlearn: 0.4334933\ttest: 0.3178863\tbest: 0.3209441 (214)\ttotal: 877ms\tremaining: 3.15s\n",
      "218:\tlearn: 0.4374279\ttest: 0.3200262\tbest: 0.3209441 (214)\ttotal: 880ms\tremaining: 3.14s\n",
      "219:\tlearn: 0.4409270\ttest: 0.3261115\tbest: 0.3261115 (219)\ttotal: 884ms\tremaining: 3.13s\n",
      "220:\tlearn: 0.4426828\ttest: 0.3279223\tbest: 0.3279223 (220)\ttotal: 888ms\tremaining: 3.13s\n",
      "221:\tlearn: 0.4439245\ttest: 0.3279223\tbest: 0.3279223 (220)\ttotal: 891ms\tremaining: 3.12s\n",
      "222:\tlearn: 0.4460562\ttest: 0.3279223\tbest: 0.3279223 (220)\ttotal: 894ms\tremaining: 3.12s\n",
      "223:\tlearn: 0.4472953\ttest: 0.3336282\tbest: 0.3336282 (223)\ttotal: 897ms\tremaining: 3.11s\n",
      "224:\tlearn: 0.4474820\ttest: 0.3357290\tbest: 0.3357290 (224)\ttotal: 902ms\tremaining: 3.1s\n",
      "225:\tlearn: 0.4483950\ttest: 0.3354129\tbest: 0.3357290 (224)\ttotal: 905ms\tremaining: 3.1s\n",
      "226:\tlearn: 0.4489051\ttest: 0.3354129\tbest: 0.3357290 (224)\ttotal: 907ms\tremaining: 3.09s\n",
      "227:\tlearn: 0.4485032\ttest: 0.3354129\tbest: 0.3357290 (224)\ttotal: 911ms\tremaining: 3.08s\n",
      "228:\tlearn: 0.4478846\ttest: 0.3371909\tbest: 0.3371909 (228)\ttotal: 914ms\tremaining: 3.08s\n",
      "229:\tlearn: 0.4483950\ttest: 0.3350973\tbest: 0.3371909 (228)\ttotal: 918ms\tremaining: 3.07s\n",
      "230:\tlearn: 0.4494150\ttest: 0.3354129\tbest: 0.3371909 (228)\ttotal: 921ms\tremaining: 3.06s\n",
      "231:\tlearn: 0.4519607\ttest: 0.3402390\tbest: 0.3402390 (231)\ttotal: 924ms\tremaining: 3.06s\n",
      "232:\tlearn: 0.4520697\ttest: 0.3423296\tbest: 0.3423296 (232)\ttotal: 927ms\tremaining: 3.05s\n",
      "233:\tlearn: 0.4546793\ttest: 0.3437702\tbest: 0.3437702 (233)\ttotal: 931ms\tremaining: 3.05s\n",
      "234:\tlearn: 0.4563082\ttest: 0.3461741\tbest: 0.3461741 (234)\ttotal: 935ms\tremaining: 3.04s\n",
      "235:\tlearn: 0.4583308\ttest: 0.3437702\tbest: 0.3461741 (234)\ttotal: 938ms\tremaining: 3.04s\n",
      "236:\tlearn: 0.4584409\ttest: 0.3444164\tbest: 0.3461741 (234)\ttotal: 941ms\tremaining: 3.03s\n",
      "237:\tlearn: 0.4599556\ttest: 0.3440930\tbest: 0.3461741 (234)\ttotal: 944ms\tremaining: 3.02s\n",
      "238:\tlearn: 0.4593407\ttest: 0.3440930\tbest: 0.3461741 (234)\ttotal: 949ms\tremaining: 3.02s\n",
      "239:\tlearn: 0.4613575\ttest: 0.3440930\tbest: 0.3461741 (234)\ttotal: 952ms\tremaining: 3.01s\n",
      "240:\tlearn: 0.4613575\ttest: 0.3461741\tbest: 0.3461741 (234)\ttotal: 955ms\tremaining: 3.01s\n",
      "241:\tlearn: 0.4608537\ttest: 0.3447404\tbest: 0.3461741 (234)\ttotal: 958ms\tremaining: 3s\n",
      "242:\tlearn: 0.4613575\ttest: 0.3464992\tbest: 0.3464992 (242)\ttotal: 961ms\tremaining: 2.99s\n",
      "243:\tlearn: 0.4625862\ttest: 0.3452026\tbest: 0.3464992 (242)\ttotal: 965ms\tremaining: 2.99s\n",
      "244:\tlearn: 0.4629785\ttest: 0.3452026\tbest: 0.3464992 (242)\ttotal: 968ms\tremaining: 2.98s\n",
      "245:\tlearn: 0.4651003\ttest: 0.3458497\tbest: 0.3464992 (242)\ttotal: 971ms\tremaining: 2.98s\n",
      "246:\tlearn: 0.4644867\ttest: 0.3437702\tbest: 0.3464992 (242)\ttotal: 974ms\tremaining: 2.97s\n",
      "247:\tlearn: 0.4647092\ttest: 0.3434480\tbest: 0.3464992 (242)\ttotal: 978ms\tremaining: 2.96s\n",
      "248:\tlearn: 0.4672190\ttest: 0.3413662\tbest: 0.3464992 (242)\ttotal: 981ms\tremaining: 2.96s\n",
      "249:\tlearn: 0.4687220\ttest: 0.3413662\tbest: 0.3464992 (242)\ttotal: 984ms\tremaining: 2.95s\n",
      "250:\tlearn: 0.4687220\ttest: 0.3395994\tbest: 0.3464992 (242)\ttotal: 987ms\tremaining: 2.95s\n",
      "251:\tlearn: 0.4699474\ttest: 0.3395994\tbest: 0.3464992 (242)\ttotal: 991ms\tremaining: 2.94s\n",
      "252:\tlearn: 0.4718341\ttest: 0.3378259\tbest: 0.3464992 (242)\ttotal: 997ms\tremaining: 2.94s\n",
      "253:\tlearn: 0.4719468\ttest: 0.3381443\tbest: 0.3464992 (242)\ttotal: 1s\tremaining: 2.94s\n",
      "254:\tlearn: 0.4713347\ttest: 0.3405596\tbest: 0.3464992 (242)\ttotal: 1s\tremaining: 2.94s\n",
      "255:\tlearn: 0.4737832\ttest: 0.3444164\tbest: 0.3464992 (242)\ttotal: 1.01s\tremaining: 2.93s\n",
      "256:\tlearn: 0.4787608\ttest: 0.3429750\tbest: 0.3464992 (242)\ttotal: 1.01s\tremaining: 2.93s\n",
      "257:\tlearn: 0.4794857\ttest: 0.3447404\tbest: 0.3464992 (242)\ttotal: 1.01s\tremaining: 2.92s\n",
      "258:\tlearn: 0.4804783\ttest: 0.3447404\tbest: 0.3464992 (242)\ttotal: 1.02s\tremaining: 2.91s\n",
      "259:\tlearn: 0.4818506\ttest: 0.3444164\tbest: 0.3464992 (242)\ttotal: 1.02s\tremaining: 2.91s\n",
      "260:\tlearn: 0.4828407\ttest: 0.3440930\tbest: 0.3464992 (242)\ttotal: 1.02s\tremaining: 2.9s\n",
      "261:\tlearn: 0.4829907\ttest: 0.3426520\tbest: 0.3464992 (242)\ttotal: 1.03s\tremaining: 2.9s\n",
      "262:\tlearn: 0.4834848\ttest: 0.3420079\tbest: 0.3464992 (242)\ttotal: 1.03s\tremaining: 2.89s\n",
      "263:\tlearn: 0.4847027\ttest: 0.3381443\tbest: 0.3464992 (242)\ttotal: 1.03s\tremaining: 2.88s\n",
      "264:\tlearn: 0.4858051\ttest: 0.3360458\tbest: 0.3464992 (242)\ttotal: 1.04s\tremaining: 2.88s\n",
      "265:\tlearn: 0.4864448\ttest: 0.3381443\tbest: 0.3464992 (242)\ttotal: 1.04s\tremaining: 2.87s\n",
      "266:\tlearn: 0.4871384\ttest: 0.3381443\tbest: 0.3464992 (242)\ttotal: 1.04s\tremaining: 2.87s\n",
      "267:\tlearn: 0.4869923\ttest: 0.3381443\tbest: 0.3464992 (242)\ttotal: 1.05s\tremaining: 2.86s\n",
      "268:\tlearn: 0.4899501\ttest: 0.3381443\tbest: 0.3464992 (242)\ttotal: 1.05s\tremaining: 2.85s\n",
      "269:\tlearn: 0.4908177\ttest: 0.3405596\tbest: 0.3464992 (242)\ttotal: 1.05s\tremaining: 2.85s\n",
      "270:\tlearn: 0.4916592\ttest: 0.3405596\tbest: 0.3464992 (242)\ttotal: 1.05s\tremaining: 2.84s\n",
      "271:\tlearn: 0.4909342\ttest: 0.3405596\tbest: 0.3464992 (242)\ttotal: 1.06s\tremaining: 2.83s\n",
      "272:\tlearn: 0.4913092\ttest: 0.3405596\tbest: 0.3464992 (242)\ttotal: 1.06s\tremaining: 2.83s\n",
      "273:\tlearn: 0.4937415\ttest: 0.3405596\tbest: 0.3464992 (242)\ttotal: 1.06s\tremaining: 2.82s\n",
      "274:\tlearn: 0.4937415\ttest: 0.3423296\tbest: 0.3464992 (242)\ttotal: 1.07s\tremaining: 2.82s\n",
      "275:\tlearn: 0.4973079\ttest: 0.3450650\tbest: 0.3464992 (242)\ttotal: 1.07s\tremaining: 2.81s\n",
      "276:\tlearn: 0.4981671\ttest: 0.3447404\tbest: 0.3464992 (242)\ttotal: 1.07s\tremaining: 2.81s\n",
      "277:\tlearn: 0.4998819\ttest: 0.3468249\tbest: 0.3468249 (277)\ttotal: 1.08s\tremaining: 2.8s\n",
      "278:\tlearn: 0.5013423\ttest: 0.3461741\tbest: 0.3468249 (277)\ttotal: 1.08s\tremaining: 2.79s\n",
      "279:\tlearn: 0.5018286\ttest: 0.3482514\tbest: 0.3482514 (279)\ttotal: 1.08s\tremaining: 2.79s\n",
      "280:\tlearn: 0.5024332\ttest: 0.3482514\tbest: 0.3482514 (279)\ttotal: 1.09s\tremaining: 2.78s\n",
      "281:\tlearn: 0.5057029\ttest: 0.3455258\tbest: 0.3482514 (279)\ttotal: 1.09s\tremaining: 2.78s\n",
      "282:\tlearn: 0.5058222\ttest: 0.3472750\tbest: 0.3482514 (279)\ttotal: 1.09s\tremaining: 2.77s\n",
      "283:\tlearn: 0.5073958\ttest: 0.3472750\tbest: 0.3482514 (279)\ttotal: 1.1s\tremaining: 2.76s\n",
      "284:\tlearn: 0.5067918\ttest: 0.3493434\tbest: 0.3493434 (284)\ttotal: 1.1s\tremaining: 2.76s\n",
      "285:\tlearn: 0.5087283\ttest: 0.3452026\tbest: 0.3493434 (284)\ttotal: 1.1s\tremaining: 2.75s\n",
      "286:\tlearn: 0.5101783\ttest: 0.3452026\tbest: 0.3493434 (284)\ttotal: 1.11s\tremaining: 2.75s\n",
      "287:\tlearn: 0.5117465\ttest: 0.3452026\tbest: 0.3493434 (284)\ttotal: 1.11s\tremaining: 2.74s\n",
      "288:\tlearn: 0.5127109\ttest: 0.3479253\tbest: 0.3493434 (284)\ttotal: 1.11s\tremaining: 2.74s\n",
      "289:\tlearn: 0.5106611\ttest: 0.3472750\tbest: 0.3493434 (284)\ttotal: 1.11s\tremaining: 2.73s\n",
      "290:\tlearn: 0.5119880\ttest: 0.3510806\tbest: 0.3510806 (290)\ttotal: 1.12s\tremaining: 2.73s\n",
      "291:\tlearn: 0.5139142\ttest: 0.3551951\tbest: 0.3551951 (291)\ttotal: 1.12s\tremaining: 2.72s\n",
      "292:\tlearn: 0.5148759\ttest: 0.3551951\tbest: 0.3551951 (291)\ttotal: 1.13s\tremaining: 2.71s\n",
      "293:\tlearn: 0.5155948\ttest: 0.3531397\tbest: 0.3551951 (291)\ttotal: 1.13s\tremaining: 2.71s\n",
      "294:\tlearn: 0.5171605\ttest: 0.3534689\tbest: 0.3551951 (291)\ttotal: 1.13s\tremaining: 2.7s\n",
      "295:\tlearn: 0.5177618\ttest: 0.3510806\tbest: 0.3551951 (291)\ttotal: 1.14s\tremaining: 2.7s\n",
      "296:\tlearn: 0.5196794\ttest: 0.3486923\tbest: 0.3551951 (291)\ttotal: 1.14s\tremaining: 2.7s\n",
      "297:\tlearn: 0.5207589\ttest: 0.3507537\tbest: 0.3551951 (291)\ttotal: 1.14s\tremaining: 2.69s\n",
      "298:\tlearn: 0.5212373\ttest: 0.3463040\tbest: 0.3551951 (291)\ttotal: 1.15s\tremaining: 2.69s\n",
      "299:\tlearn: 0.5218378\ttest: 0.3504274\tbest: 0.3551951 (291)\ttotal: 1.15s\tremaining: 2.68s\n",
      "300:\tlearn: 0.5218378\ttest: 0.3483676\tbest: 0.3551951 (291)\ttotal: 1.15s\tremaining: 2.67s\n",
      "301:\tlearn: 0.5237489\ttest: 0.3480435\tbest: 0.3551951 (291)\ttotal: 1.16s\tremaining: 2.67s\n",
      "302:\tlearn: 0.5242261\ttest: 0.3501017\tbest: 0.3551951 (291)\ttotal: 1.16s\tremaining: 2.66s\n",
      "303:\tlearn: 0.5236263\ttest: 0.3456596\tbest: 0.3551951 (291)\ttotal: 1.16s\tremaining: 2.66s\n",
      "304:\tlearn: 0.5245945\ttest: 0.3456596\tbest: 0.3551951 (291)\ttotal: 1.16s\tremaining: 2.65s\n",
      "305:\tlearn: 0.5249488\ttest: 0.3497766\tbest: 0.3551951 (291)\ttotal: 1.17s\tremaining: 2.65s\n",
      "306:\tlearn: 0.5269784\ttest: 0.3542065\tbest: 0.3551951 (291)\ttotal: 1.17s\tremaining: 2.65s\n",
      "307:\tlearn: 0.5282826\ttest: 0.3521560\tbest: 0.3551951 (291)\ttotal: 1.17s\tremaining: 2.64s\n",
      "308:\tlearn: 0.5281775\ttest: 0.3562533\tbest: 0.3562533 (308)\ttotal: 1.18s\tremaining: 2.63s\n",
      "309:\tlearn: 0.5303274\ttest: 0.3559234\tbest: 0.3562533 (308)\ttotal: 1.18s\tremaining: 2.63s\n",
      "310:\tlearn: 0.5309051\ttest: 0.3508528\tbest: 0.3562533 (308)\ttotal: 1.19s\tremaining: 2.63s\n",
      "311:\tlearn: 0.5313795\ttest: 0.3549374\tbest: 0.3562533 (308)\ttotal: 1.19s\tremaining: 2.62s\n",
      "312:\tlearn: 0.5313795\ttest: 0.3528970\tbest: 0.3562533 (308)\ttotal: 1.19s\tremaining: 2.61s\n",
      "313:\tlearn: 0.5315036\ttest: 0.3593381\tbest: 0.3593381 (313)\ttotal: 1.19s\tremaining: 2.61s\n",
      "314:\tlearn: 0.5317296\ttest: 0.3593381\tbest: 0.3593381 (313)\ttotal: 1.2s\tremaining: 2.61s\n",
      "315:\tlearn: 0.5340969\ttest: 0.3610360\tbest: 0.3610360 (315)\ttotal: 1.2s\tremaining: 2.6s\n",
      "316:\tlearn: 0.5345697\ttest: 0.3593381\tbest: 0.3610360 (315)\ttotal: 1.21s\tremaining: 2.6s\n",
      "317:\tlearn: 0.5355146\ttest: 0.3566450\tbest: 0.3610360 (315)\ttotal: 1.21s\tremaining: 2.59s\n",
      "318:\tlearn: 0.5336239\ttest: 0.3590069\tbest: 0.3610360 (315)\ttotal: 1.21s\tremaining: 2.59s\n",
      "319:\tlearn: 0.5351669\ttest: 0.3590069\tbest: 0.3610360 (315)\ttotal: 1.22s\tremaining: 2.58s\n",
      "320:\tlearn: 0.5348190\ttest: 0.3566450\tbest: 0.3610360 (315)\ttotal: 1.22s\tremaining: 2.58s\n",
      "321:\tlearn: 0.5357642\ttest: 0.3603723\tbest: 0.3610360 (315)\ttotal: 1.22s\tremaining: 2.57s\n",
      "322:\tlearn: 0.5363615\ttest: 0.3603723\tbest: 0.3610360 (315)\ttotal: 1.22s\tremaining: 2.57s\n",
      "323:\tlearn: 0.5365836\ttest: 0.3607038\tbest: 0.3610360 (315)\ttotal: 1.23s\tremaining: 2.56s\n",
      "324:\tlearn: 0.5362365\ttest: 0.3586763\tbest: 0.3610360 (315)\ttotal: 1.23s\tremaining: 2.56s\n",
      "325:\tlearn: 0.5369303\ttest: 0.3586763\tbest: 0.3610360 (315)\ttotal: 1.23s\tremaining: 2.55s\n",
      "326:\tlearn: 0.5378731\ttest: 0.3586763\tbest: 0.3610360 (315)\ttotal: 1.24s\tremaining: 2.55s\n",
      "327:\tlearn: 0.5402260\ttest: 0.3586763\tbest: 0.3610360 (315)\ttotal: 1.24s\tremaining: 2.54s\n",
      "328:\tlearn: 0.5425736\ttest: 0.3630614\tbest: 0.3630614 (328)\ttotal: 1.24s\tremaining: 2.54s\n",
      "329:\tlearn: 0.5436372\ttest: 0.3627276\tbest: 0.3630614 (328)\ttotal: 1.25s\tremaining: 2.53s\n",
      "330:\tlearn: 0.5438533\ttest: 0.3603723\tbest: 0.3630614 (328)\ttotal: 1.25s\tremaining: 2.53s\n",
      "331:\tlearn: 0.5439795\ttest: 0.3603723\tbest: 0.3630614 (328)\ttotal: 1.25s\tremaining: 2.52s\n",
      "332:\tlearn: 0.5447004\ttest: 0.3627276\tbest: 0.3630614 (328)\ttotal: 1.26s\tremaining: 2.52s\n",
      "333:\tlearn: 0.5456364\ttest: 0.3627276\tbest: 0.3630614 (328)\ttotal: 1.26s\tremaining: 2.52s\n",
      "334:\tlearn: 0.5492448\ttest: 0.3650830\tbest: 0.3650830 (334)\ttotal: 1.26s\tremaining: 2.51s\n",
      "335:\tlearn: 0.5489058\ttest: 0.3671010\tbest: 0.3671010 (335)\ttotal: 1.27s\tremaining: 2.51s\n",
      "336:\tlearn: 0.5485665\ttest: 0.3671010\tbest: 0.3671010 (335)\ttotal: 1.27s\tremaining: 2.5s\n",
      "337:\tlearn: 0.5499653\ttest: 0.3647477\tbest: 0.3671010 (335)\ttotal: 1.27s\tremaining: 2.5s\n",
      "338:\tlearn: 0.5516997\ttest: 0.3650830\tbest: 0.3671010 (335)\ttotal: 1.28s\tremaining: 2.49s\n",
      "339:\tlearn: 0.5512346\ttest: 0.3633957\tbest: 0.3671010 (335)\ttotal: 1.28s\tremaining: 2.49s\n",
      "340:\tlearn: 0.5512346\ttest: 0.3654189\tbest: 0.3671010 (335)\ttotal: 1.28s\tremaining: 2.48s\n",
      "341:\tlearn: 0.5535580\ttest: 0.3667641\tbest: 0.3671010 (335)\ttotal: 1.29s\tremaining: 2.48s\n",
      "342:\tlearn: 0.5533495\ttest: 0.3707859\tbest: 0.3707859 (342)\ttotal: 1.29s\tremaining: 2.47s\n",
      "343:\tlearn: 0.5528849\ttest: 0.3731326\tbest: 0.3731326 (343)\ttotal: 1.3s\tremaining: 2.47s\n",
      "344:\tlearn: 0.5525479\ttest: 0.3734746\tbest: 0.3734746 (344)\ttotal: 1.3s\tremaining: 2.47s\n",
      "345:\tlearn: 0.5538139\ttest: 0.3714663\tbest: 0.3734746 (344)\ttotal: 1.3s\tremaining: 2.46s\n",
      "346:\tlearn: 0.5552059\ttest: 0.3714663\tbest: 0.3734746 (344)\ttotal: 1.3s\tremaining: 2.46s\n",
      "347:\tlearn: 0.5561328\ttest: 0.3734746\tbest: 0.3734746 (344)\ttotal: 1.31s\tremaining: 2.45s\n",
      "348:\tlearn: 0.5552059\ttest: 0.3714663\tbest: 0.3734746 (344)\ttotal: 1.31s\tremaining: 2.45s\n",
      "349:\tlearn: 0.5565182\ttest: 0.3734746\tbest: 0.3734746 (344)\ttotal: 1.31s\tremaining: 2.44s\n",
      "350:\tlearn: 0.5560547\ttest: 0.3734746\tbest: 0.3734746 (344)\ttotal: 1.32s\tremaining: 2.44s\n",
      "351:\tlearn: 0.5574447\ttest: 0.3734746\tbest: 0.3734746 (344)\ttotal: 1.32s\tremaining: 2.43s\n",
      "352:\tlearn: 0.5574447\ttest: 0.3731326\tbest: 0.3734746 (344)\ttotal: 1.32s\tremaining: 2.43s\n",
      "353:\tlearn: 0.5579076\ttest: 0.3731326\tbest: 0.3734746 (344)\ttotal: 1.33s\tremaining: 2.42s\n",
      "354:\tlearn: 0.5577023\ttest: 0.3731326\tbest: 0.3734746 (344)\ttotal: 1.33s\tremaining: 2.42s\n",
      "355:\tlearn: 0.5572390\ttest: 0.3691152\tbest: 0.3734746 (344)\ttotal: 1.34s\tremaining: 2.42s\n",
      "356:\tlearn: 0.5584993\ttest: 0.3711258\tbest: 0.3734746 (344)\ttotal: 1.34s\tremaining: 2.41s\n",
      "357:\tlearn: 0.5575735\ttest: 0.3731326\tbest: 0.3734746 (344)\ttotal: 1.34s\tremaining: 2.41s\n",
      "358:\tlearn: 0.5579076\ttest: 0.3711258\tbest: 0.3734746 (344)\ttotal: 1.35s\tremaining: 2.4s\n",
      "359:\tlearn: 0.5594242\ttest: 0.3754794\tbest: 0.3754794 (359)\ttotal: 1.35s\tremaining: 2.4s\n",
      "360:\tlearn: 0.5592200\ttest: 0.3754794\tbest: 0.3754794 (359)\ttotal: 1.35s\tremaining: 2.4s\n",
      "361:\tlearn: 0.5623235\ttest: 0.3758235\tbest: 0.3758235 (361)\ttotal: 1.36s\tremaining: 2.39s\n",
      "362:\tlearn: 0.5621939\ttest: 0.3794778\tbest: 0.3794778 (362)\ttotal: 1.36s\tremaining: 2.39s\n",
      "363:\tlearn: 0.5632452\ttest: 0.3811235\tbest: 0.3811235 (363)\ttotal: 1.36s\tremaining: 2.38s\n",
      "364:\tlearn: 0.5647562\ttest: 0.3791312\tbest: 0.3811235 (363)\ttotal: 1.37s\tremaining: 2.38s\n",
      "365:\tlearn: 0.5640955\ttest: 0.3811235\tbest: 0.3811235 (363)\ttotal: 1.37s\tremaining: 2.37s\n",
      "366:\tlearn: 0.5640955\ttest: 0.3811235\tbest: 0.3811235 (363)\ttotal: 1.37s\tremaining: 2.37s\n",
      "367:\tlearn: 0.5639655\ttest: 0.3834616\tbest: 0.3834616 (367)\ttotal: 1.38s\tremaining: 2.36s\n",
      "368:\tlearn: 0.5656759\ttest: 0.3814715\tbest: 0.3834616 (367)\ttotal: 1.38s\tremaining: 2.36s\n",
      "369:\tlearn: 0.5682328\ttest: 0.3814715\tbest: 0.3834616 (367)\ttotal: 1.38s\tremaining: 2.35s\n",
      "370:\tlearn: 0.5680353\ttest: 0.3794778\tbest: 0.3834616 (367)\ttotal: 1.39s\tremaining: 2.35s\n",
      "371:\tlearn: 0.5681661\ttest: 0.3838119\tbest: 0.3838119 (371)\ttotal: 1.39s\tremaining: 2.35s\n",
      "372:\tlearn: 0.5681661\ttest: 0.3818203\tbest: 0.3838119 (371)\ttotal: 1.39s\tremaining: 2.34s\n",
      "373:\tlearn: 0.5682970\ttest: 0.3818203\tbest: 0.3838119 (371)\ttotal: 1.4s\tremaining: 2.34s\n",
      "374:\tlearn: 0.5690181\ttest: 0.3798250\tbest: 0.3838119 (371)\ttotal: 1.4s\tremaining: 2.33s\n",
      "375:\tlearn: 0.5696738\ttest: 0.3818203\tbest: 0.3838119 (371)\ttotal: 1.4s\tremaining: 2.33s\n",
      "376:\tlearn: 0.5690181\ttest: 0.3794778\tbest: 0.3838119 (371)\ttotal: 1.41s\tremaining: 2.32s\n",
      "377:\tlearn: 0.5719643\ttest: 0.3818203\tbest: 0.3838119 (371)\ttotal: 1.41s\tremaining: 2.32s\n",
      "378:\tlearn: 0.5722902\ttest: 0.3818203\tbest: 0.3838119 (371)\ttotal: 1.41s\tremaining: 2.31s\n",
      "379:\tlearn: 0.5745740\ttest: 0.3798250\tbest: 0.3838119 (371)\ttotal: 1.42s\tremaining: 2.31s\n",
      "380:\tlearn: 0.5745740\ttest: 0.3794778\tbest: 0.3838119 (371)\ttotal: 1.42s\tremaining: 2.31s\n",
      "381:\tlearn: 0.5748378\ttest: 0.3798250\tbest: 0.3838119 (371)\ttotal: 1.42s\tremaining: 2.3s\n",
      "382:\tlearn: 0.5754262\ttest: 0.3798250\tbest: 0.3838119 (371)\ttotal: 1.43s\tremaining: 2.3s\n",
      "383:\tlearn: 0.5770588\ttest: 0.3818203\tbest: 0.3838119 (371)\ttotal: 1.43s\tremaining: 2.29s\n",
      "384:\tlearn: 0.5782925\ttest: 0.3818203\tbest: 0.3838119 (371)\ttotal: 1.43s\tremaining: 2.29s\n",
      "385:\tlearn: 0.5787475\ttest: 0.3798250\tbest: 0.3838119 (371)\ttotal: 1.44s\tremaining: 2.29s\n",
      "386:\tlearn: 0.5796568\ttest: 0.3818203\tbest: 0.3838119 (371)\ttotal: 1.44s\tremaining: 2.28s\n",
      "387:\tlearn: 0.5802994\ttest: 0.3834616\tbest: 0.3838119 (371)\ttotal: 1.44s\tremaining: 2.28s\n",
      "388:\tlearn: 0.5799782\ttest: 0.3834616\tbest: 0.3838119 (371)\ttotal: 1.45s\tremaining: 2.27s\n",
      "389:\tlearn: 0.5798454\ttest: 0.3811235\tbest: 0.3838119 (371)\ttotal: 1.45s\tremaining: 2.27s\n",
      "390:\tlearn: 0.5806202\ttest: 0.3794778\tbest: 0.3838119 (371)\ttotal: 1.45s\tremaining: 2.26s\n",
      "391:\tlearn: 0.5812067\ttest: 0.3814715\tbest: 0.3838119 (371)\ttotal: 1.46s\tremaining: 2.26s\n",
      "392:\tlearn: 0.5810737\ttest: 0.3814715\tbest: 0.3838119 (371)\ttotal: 1.46s\tremaining: 2.25s\n",
      "393:\tlearn: 0.5817140\ttest: 0.3827631\tbest: 0.3838119 (371)\ttotal: 1.46s\tremaining: 2.25s\n",
      "394:\tlearn: 0.5821667\ttest: 0.3827631\tbest: 0.3838119 (371)\ttotal: 1.47s\tremaining: 2.24s\n",
      "395:\tlearn: 0.5812067\ttest: 0.3854481\tbest: 0.3854481 (395)\ttotal: 1.47s\tremaining: 2.24s\n",
      "396:\tlearn: 0.5825130\ttest: 0.3850971\tbest: 0.3854481 (395)\ttotal: 1.47s\tremaining: 2.23s\n",
      "397:\tlearn: 0.5832857\ttest: 0.3850971\tbest: 0.3854481 (395)\ttotal: 1.47s\tremaining: 2.23s\n",
      "398:\tlearn: 0.5849105\ttest: 0.3847466\tbest: 0.3854481 (395)\ttotal: 1.48s\tremaining: 2.23s\n",
      "399:\tlearn: 0.5852289\ttest: 0.3847466\tbest: 0.3854481 (395)\ttotal: 1.48s\tremaining: 2.22s\n",
      "400:\tlearn: 0.5859986\ttest: 0.3847466\tbest: 0.3854481 (395)\ttotal: 1.49s\tremaining: 2.22s\n",
      "401:\tlearn: 0.5873523\ttest: 0.3831121\tbest: 0.3854481 (395)\ttotal: 1.49s\tremaining: 2.21s\n",
      "402:\tlearn: 0.5858648\ttest: 0.3811235\tbest: 0.3854481 (395)\ttotal: 1.49s\tremaining: 2.21s\n",
      "403:\tlearn: 0.5858648\ttest: 0.3811235\tbest: 0.3854481 (395)\ttotal: 1.5s\tremaining: 2.21s\n",
      "404:\tlearn: 0.5867673\ttest: 0.3811235\tbest: 0.3854481 (395)\ttotal: 1.5s\tremaining: 2.2s\n",
      "405:\tlearn: 0.5873523\ttest: 0.3831121\tbest: 0.3854481 (395)\ttotal: 1.5s\tremaining: 2.2s\n",
      "406:\tlearn: 0.5879373\ttest: 0.3811235\tbest: 0.3854481 (395)\ttotal: 1.5s\tremaining: 2.19s\n",
      "407:\tlearn: 0.5889729\ttest: 0.3807760\tbest: 0.3854481 (395)\ttotal: 1.51s\tremaining: 2.19s\n",
      "408:\tlearn: 0.5887912\ttest: 0.3831121\tbest: 0.3854481 (395)\ttotal: 1.51s\tremaining: 2.18s\n",
      "409:\tlearn: 0.5909973\ttest: 0.3814715\tbest: 0.3854481 (395)\ttotal: 1.51s\tremaining: 2.18s\n",
      "410:\tlearn: 0.5909973\ttest: 0.3850971\tbest: 0.3854481 (395)\ttotal: 1.52s\tremaining: 2.17s\n",
      "411:\tlearn: 0.5917622\ttest: 0.3870784\tbest: 0.3870784 (411)\ttotal: 1.52s\tremaining: 2.17s\n",
      "412:\tlearn: 0.5918973\ttest: 0.3870784\tbest: 0.3870784 (411)\ttotal: 1.52s\tremaining: 2.17s\n",
      "413:\tlearn: 0.5938302\ttest: 0.3870784\tbest: 0.3870784 (411)\ttotal: 1.53s\tremaining: 2.16s\n",
      "414:\tlearn: 0.5936948\ttest: 0.3890562\tbest: 0.3890562 (414)\ttotal: 1.53s\tremaining: 2.16s\n",
      "415:\tlearn: 0.5963849\ttest: 0.3910304\tbest: 0.3910304 (415)\ttotal: 1.53s\tremaining: 2.15s\n",
      "416:\tlearn: 0.5968325\ttest: 0.3910304\tbest: 0.3910304 (415)\ttotal: 1.54s\tremaining: 2.15s\n",
      "417:\tlearn: 0.5980381\ttest: 0.3887028\tbest: 0.3910304 (415)\ttotal: 1.54s\tremaining: 2.15s\n",
      "418:\tlearn: 0.5980381\ttest: 0.3887028\tbest: 0.3910304 (415)\ttotal: 1.54s\tremaining: 2.14s\n",
      "419:\tlearn: 0.5987572\ttest: 0.3867265\tbest: 0.3910304 (415)\ttotal: 1.55s\tremaining: 2.14s\n",
      "420:\tlearn: 0.6008520\ttest: 0.3867265\tbest: 0.3910304 (415)\ttotal: 1.55s\tremaining: 2.13s\n",
      "421:\tlearn: 0.6005427\ttest: 0.3887028\tbest: 0.3910304 (415)\ttotal: 1.55s\tremaining: 2.13s\n",
      "422:\tlearn: 0.6003696\ttest: 0.3887028\tbest: 0.3910304 (415)\ttotal: 1.56s\tremaining: 2.13s\n",
      "423:\tlearn: 0.6024618\ttest: 0.3890562\tbest: 0.3910304 (415)\ttotal: 1.56s\tremaining: 2.12s\n",
      "424:\tlearn: 0.6027701\ttest: 0.3890562\tbest: 0.3910304 (415)\ttotal: 1.56s\tremaining: 2.12s\n",
      "425:\tlearn: 0.6014343\ttest: 0.3910304\tbest: 0.3910304 (415)\ttotal: 1.57s\tremaining: 2.11s\n",
      "426:\tlearn: 0.6033519\ttest: 0.3910304\tbest: 0.3910304 (415)\ttotal: 1.57s\tremaining: 2.11s\n",
      "427:\tlearn: 0.6037966\ttest: 0.3910304\tbest: 0.3910304 (415)\ttotal: 1.57s\tremaining: 2.1s\n",
      "428:\tlearn: 0.6060172\ttest: 0.3906755\tbest: 0.3910304 (415)\ttotal: 1.58s\tremaining: 2.1s\n",
      "429:\tlearn: 0.6051296\ttest: 0.3906755\tbest: 0.3910304 (415)\ttotal: 1.58s\tremaining: 2.09s\n",
      "430:\tlearn: 0.6061547\ttest: 0.3906755\tbest: 0.3910304 (415)\ttotal: 1.58s\tremaining: 2.09s\n",
      "431:\tlearn: 0.6065983\ttest: 0.3903213\tbest: 0.3910304 (415)\ttotal: 1.59s\tremaining: 2.09s\n",
      "432:\tlearn: 0.6065983\ttest: 0.3903213\tbest: 0.3910304 (415)\ttotal: 1.59s\tremaining: 2.08s\n",
      "433:\tlearn: 0.6088132\ttest: 0.3922890\tbest: 0.3922890 (433)\ttotal: 1.59s\tremaining: 2.08s\n",
      "434:\tlearn: 0.6088132\ttest: 0.3919339\tbest: 0.3922890 (433)\ttotal: 1.6s\tremaining: 2.07s\n",
      "435:\tlearn: 0.6092556\ttest: 0.3919339\tbest: 0.3922890 (433)\ttotal: 1.6s\tremaining: 2.07s\n",
      "436:\tlearn: 0.6089511\ttest: 0.3919339\tbest: 0.3922890 (433)\ttotal: 1.6s\tremaining: 2.06s\n",
      "437:\tlearn: 0.6116029\ttest: 0.3919339\tbest: 0.3922890 (433)\ttotal: 1.61s\tremaining: 2.06s\n",
      "438:\tlearn: 0.6130646\ttest: 0.3942531\tbest: 0.3942531 (438)\ttotal: 1.61s\tremaining: 2.06s\n",
      "439:\tlearn: 0.6121826\ttest: 0.3942531\tbest: 0.3942531 (438)\ttotal: 1.61s\tremaining: 2.05s\n",
      "440:\tlearn: 0.6126237\ttest: 0.3942531\tbest: 0.3942531 (438)\ttotal: 1.62s\tremaining: 2.05s\n",
      "441:\tlearn: 0.6121826\ttest: 0.3942531\tbest: 0.3942531 (438)\ttotal: 1.62s\tremaining: 2.04s\n",
      "442:\tlearn: 0.6120441\ttest: 0.3942531\tbest: 0.3942531 (438)\ttotal: 1.62s\tremaining: 2.04s\n",
      "443:\tlearn: 0.6139459\ttest: 0.3942531\tbest: 0.3942531 (438)\ttotal: 1.63s\tremaining: 2.04s\n",
      "444:\tlearn: 0.6148263\ttest: 0.3903213\tbest: 0.3942531 (438)\ttotal: 1.63s\tremaining: 2.03s\n",
      "445:\tlearn: 0.6154053\ttest: 0.3922890\tbest: 0.3942531 (438)\ttotal: 1.63s\tremaining: 2.03s\n",
      "446:\tlearn: 0.6146874\ttest: 0.3922890\tbest: 0.3942531 (438)\ttotal: 1.64s\tremaining: 2.02s\n",
      "447:\tlearn: 0.6151273\ttest: 0.3922890\tbest: 0.3942531 (438)\ttotal: 1.64s\tremaining: 2.02s\n",
      "448:\tlearn: 0.6165848\ttest: 0.3919339\tbest: 0.3942531 (438)\ttotal: 1.64s\tremaining: 2.02s\n",
      "449:\tlearn: 0.6167241\ttest: 0.3935407\tbest: 0.3942531 (438)\ttotal: 1.65s\tremaining: 2.01s\n",
      "450:\tlearn: 0.6165631\ttest: 0.3935407\tbest: 0.3942531 (438)\ttotal: 1.65s\tremaining: 2.01s\n",
      "451:\tlearn: 0.6168633\ttest: 0.3935407\tbest: 0.3942531 (438)\ttotal: 1.65s\tremaining: 2s\n",
      "452:\tlearn: 0.6168633\ttest: 0.3919339\tbest: 0.3942531 (438)\ttotal: 1.66s\tremaining: 2s\n",
      "453:\tlearn: 0.6184597\ttest: 0.3919339\tbest: 0.3942531 (438)\ttotal: 1.66s\tremaining: 2s\n",
      "454:\tlearn: 0.6184597\ttest: 0.3919339\tbest: 0.3942531 (438)\ttotal: 1.66s\tremaining: 1.99s\n",
      "455:\tlearn: 0.6206517\ttest: 0.3938966\tbest: 0.3942531 (438)\ttotal: 1.67s\tremaining: 1.99s\n",
      "456:\tlearn: 0.6202137\ttest: 0.3938966\tbest: 0.3942531 (438)\ttotal: 1.67s\tremaining: 1.99s\n",
      "457:\tlearn: 0.6218245\ttest: 0.3938966\tbest: 0.3942531 (438)\ttotal: 1.68s\tremaining: 1.99s\n",
      "458:\tlearn: 0.6225584\ttest: 0.3938966\tbest: 0.3942531 (438)\ttotal: 1.68s\tremaining: 1.98s\n",
      "459:\tlearn: 0.6237276\ttest: 0.3938966\tbest: 0.3942531 (438)\ttotal: 1.69s\tremaining: 1.98s\n",
      "460:\tlearn: 0.6247399\ttest: 0.3935407\tbest: 0.3942531 (438)\ttotal: 1.69s\tremaining: 1.98s\n",
      "461:\tlearn: 0.6247399\ttest: 0.3935407\tbest: 0.3942531 (438)\ttotal: 1.69s\tremaining: 1.97s\n",
      "462:\tlearn: 0.6250351\ttest: 0.3935407\tbest: 0.3942531 (438)\ttotal: 1.7s\tremaining: 1.97s\n",
      "463:\tlearn: 0.6273513\ttest: 0.3915795\tbest: 0.3942531 (438)\ttotal: 1.7s\tremaining: 1.97s\n",
      "464:\tlearn: 0.6273513\ttest: 0.3915795\tbest: 0.3942531 (438)\ttotal: 1.71s\tremaining: 1.96s\n",
      "465:\tlearn: 0.6285133\ttest: 0.3931855\tbest: 0.3942531 (438)\ttotal: 1.71s\tremaining: 1.96s\n",
      "466:\tlearn: 0.6302477\ttest: 0.3931855\tbest: 0.3942531 (438)\ttotal: 1.71s\tremaining: 1.96s\n",
      "467:\tlearn: 0.6311138\ttest: 0.3938966\tbest: 0.3942531 (438)\ttotal: 1.72s\tremaining: 1.95s\n",
      "468:\tlearn: 0.6311138\ttest: 0.3938966\tbest: 0.3942531 (438)\ttotal: 1.72s\tremaining: 1.95s\n",
      "469:\tlearn: 0.6313968\ttest: 0.3935407\tbest: 0.3942531 (438)\ttotal: 1.73s\tremaining: 1.95s\n",
      "470:\tlearn: 0.6321130\ttest: 0.3915795\tbest: 0.3942531 (438)\ttotal: 1.73s\tremaining: 1.94s\n",
      "471:\tlearn: 0.6331271\ttest: 0.3931855\tbest: 0.3942531 (438)\ttotal: 1.74s\tremaining: 1.94s\n",
      "472:\tlearn: 0.6331271\ttest: 0.3931855\tbest: 0.3942531 (438)\ttotal: 1.74s\tremaining: 1.94s\n",
      "473:\tlearn: 0.6331271\ttest: 0.3915795\tbest: 0.3942531 (438)\ttotal: 1.74s\tremaining: 1.93s\n",
      "474:\tlearn: 0.6322623\ttest: 0.3928310\tbest: 0.3942531 (438)\ttotal: 1.75s\tremaining: 1.93s\n",
      "475:\tlearn: 0.6321207\ttest: 0.3928310\tbest: 0.3942531 (438)\ttotal: 1.75s\tremaining: 1.93s\n",
      "476:\tlearn: 0.6312553\ttest: 0.3908727\tbest: 0.3942531 (438)\ttotal: 1.75s\tremaining: 1.92s\n",
      "477:\tlearn: 0.6315384\ttest: 0.3912258\tbest: 0.3942531 (438)\ttotal: 1.76s\tremaining: 1.92s\n",
      "478:\tlearn: 0.6315384\ttest: 0.3912258\tbest: 0.3942531 (438)\ttotal: 1.76s\tremaining: 1.92s\n",
      "479:\tlearn: 0.6324040\ttest: 0.3951417\tbest: 0.3951417 (479)\ttotal: 1.77s\tremaining: 1.91s\n",
      "480:\tlearn: 0.6341330\ttest: 0.3951417\tbest: 0.3951417 (479)\ttotal: 1.77s\tremaining: 1.91s\n",
      "481:\tlearn: 0.6342751\ttest: 0.4029314\tbest: 0.4029314 (481)\ttotal: 1.77s\tremaining: 1.91s\n",
      "482:\tlearn: 0.6339910\ttest: 0.4013502\tbest: 0.4029314 (481)\ttotal: 1.78s\tremaining: 1.9s\n",
      "483:\tlearn: 0.6342751\ttest: 0.3970944\tbest: 0.4029314 (481)\ttotal: 1.78s\tremaining: 1.9s\n",
      "484:\tlearn: 0.6345594\ttest: 0.3974525\tbest: 0.4029314 (481)\ttotal: 1.78s\tremaining: 1.89s\n",
      "485:\tlearn: 0.6354231\ttest: 0.3974525\tbest: 0.4029314 (481)\ttotal: 1.79s\tremaining: 1.89s\n",
      "486:\tlearn: 0.6357123\ttest: 0.3974525\tbest: 0.4029314 (481)\ttotal: 1.79s\tremaining: 1.89s\n",
      "487:\tlearn: 0.6352808\ttest: 0.3974525\tbest: 0.4029314 (481)\ttotal: 1.79s\tremaining: 1.88s\n",
      "488:\tlearn: 0.6357123\ttest: 0.3994031\tbest: 0.4029314 (481)\ttotal: 1.8s\tremaining: 1.88s\n",
      "489:\tlearn: 0.6361436\ttest: 0.4009893\tbest: 0.4029314 (481)\ttotal: 1.8s\tremaining: 1.87s\n",
      "490:\tlearn: 0.6371482\ttest: 0.4002693\tbest: 0.4029314 (481)\ttotal: 1.8s\tremaining: 1.87s\n",
      "491:\tlearn: 0.6375790\ttest: 0.4006290\tbest: 0.4029314 (481)\ttotal: 1.81s\tremaining: 1.86s\n",
      "492:\tlearn: 0.6375790\ttest: 0.3967370\tbest: 0.4029314 (481)\ttotal: 1.81s\tremaining: 1.86s\n",
      "493:\tlearn: 0.6393003\ttest: 0.3970944\tbest: 0.4029314 (481)\ttotal: 1.81s\tremaining: 1.86s\n",
      "494:\tlearn: 0.6398731\ttest: 0.3986847\tbest: 0.4029314 (481)\ttotal: 1.82s\tremaining: 1.85s\n",
      "495:\tlearn: 0.6394432\ttest: 0.4002693\tbest: 0.4029314 (481)\ttotal: 1.82s\tremaining: 1.85s\n",
      "496:\tlearn: 0.6401594\ttest: 0.3999103\tbest: 0.4029314 (481)\ttotal: 1.82s\tremaining: 1.84s\n",
      "497:\tlearn: 0.6415917\ttest: 0.3999103\tbest: 0.4029314 (481)\ttotal: 1.83s\tremaining: 1.84s\n",
      "498:\tlearn: 0.6418775\ttest: 0.3999103\tbest: 0.4029314 (481)\ttotal: 1.83s\tremaining: 1.84s\n",
      "499:\tlearn: 0.6418775\ttest: 0.3983265\tbest: 0.4029314 (481)\ttotal: 1.83s\tremaining: 1.83s\n",
      "500:\tlearn: 0.6420210\ttest: 0.3979689\tbest: 0.4029314 (481)\ttotal: 1.84s\tremaining: 1.83s\n",
      "501:\tlearn: 0.6424501\ttest: 0.4002693\tbest: 0.4029314 (481)\ttotal: 1.84s\tremaining: 1.83s\n",
      "502:\tlearn: 0.6424501\ttest: 0.4002693\tbest: 0.4029314 (481)\ttotal: 1.84s\tremaining: 1.82s\n",
      "503:\tlearn: 0.6433078\ttest: 0.3976120\tbest: 0.4029314 (481)\ttotal: 1.85s\tremaining: 1.82s\n",
      "504:\tlearn: 0.6441648\ttest: 0.3995519\tbest: 0.4029314 (481)\ttotal: 1.85s\tremaining: 1.81s\n",
      "505:\tlearn: 0.6445929\ttest: 0.4025697\tbest: 0.4029314 (481)\ttotal: 1.85s\tremaining: 1.81s\n",
      "506:\tlearn: 0.6451649\ttest: 0.4006290\tbest: 0.4029314 (481)\ttotal: 1.86s\tremaining: 1.81s\n",
      "507:\tlearn: 0.6458763\ttest: 0.4002693\tbest: 0.4029314 (481)\ttotal: 1.86s\tremaining: 1.8s\n",
      "508:\tlearn: 0.6460204\ttest: 0.3991942\tbest: 0.4029314 (481)\ttotal: 1.86s\tremaining: 1.8s\n",
      "509:\tlearn: 0.6468752\ttest: 0.4011293\tbest: 0.4029314 (481)\ttotal: 1.87s\tremaining: 1.79s\n",
      "510:\tlearn: 0.6474466\ttest: 0.3991942\tbest: 0.4029314 (481)\ttotal: 1.87s\tremaining: 1.79s\n",
      "511:\tlearn: 0.6474466\ttest: 0.3995519\tbest: 0.4029314 (481)\ttotal: 1.87s\tremaining: 1.79s\n",
      "512:\tlearn: 0.6477292\ttest: 0.3991942\tbest: 0.4029314 (481)\ttotal: 1.88s\tremaining: 1.78s\n",
      "513:\tlearn: 0.6478736\ttest: 0.3991942\tbest: 0.4029314 (481)\ttotal: 1.88s\tremaining: 1.78s\n",
      "514:\tlearn: 0.6485895\ttest: 0.4011293\tbest: 0.4029314 (481)\ttotal: 1.88s\tremaining: 1.77s\n",
      "515:\tlearn: 0.6484449\ttest: 0.4011293\tbest: 0.4029314 (481)\ttotal: 1.89s\tremaining: 1.77s\n",
      "516:\tlearn: 0.6485895\ttest: 0.4007708\tbest: 0.4029314 (481)\ttotal: 1.89s\tremaining: 1.77s\n",
      "517:\tlearn: 0.6490162\ttest: 0.4011293\tbest: 0.4029314 (481)\ttotal: 1.89s\tremaining: 1.76s\n",
      "518:\tlearn: 0.6495875\ttest: 0.3991942\tbest: 0.4029314 (481)\ttotal: 1.9s\tremaining: 1.76s\n",
      "519:\tlearn: 0.6507212\ttest: 0.4042672\tbest: 0.4042672 (519)\ttotal: 1.9s\tremaining: 1.75s\n",
      "520:\tlearn: 0.6508662\ttest: 0.4007708\tbest: 0.4042672 (519)\ttotal: 1.9s\tremaining: 1.75s\n",
      "521:\tlearn: 0.6512920\ttest: 0.4027010\tbest: 0.4042672 (519)\ttotal: 1.91s\tremaining: 1.75s\n",
      "522:\tlearn: 0.6515725\ttest: 0.4027010\tbest: 0.4042672 (519)\ttotal: 1.91s\tremaining: 1.74s\n",
      "523:\tlearn: 0.6510019\ttest: 0.4018482\tbest: 0.4042672 (519)\ttotal: 1.91s\tremaining: 1.74s\n",
      "524:\tlearn: 0.6517077\ttest: 0.4014884\tbest: 0.4042672 (519)\ttotal: 1.92s\tremaining: 1.74s\n",
      "525:\tlearn: 0.6515628\ttest: 0.4034215\tbest: 0.4042672 (519)\ttotal: 1.92s\tremaining: 1.73s\n",
      "526:\tlearn: 0.6534069\ttest: 0.4030609\tbest: 0.4042672 (519)\ttotal: 1.92s\tremaining: 1.73s\n",
      "527:\tlearn: 0.6532617\ttest: 0.4027010\tbest: 0.4042672 (519)\ttotal: 1.93s\tremaining: 1.72s\n",
      "528:\tlearn: 0.6532617\ttest: 0.4049891\tbest: 0.4049891 (528)\ttotal: 1.93s\tremaining: 1.72s\n",
      "529:\tlearn: 0.6538312\ttest: 0.4049891\tbest: 0.4049891 (528)\ttotal: 1.93s\tremaining: 1.72s\n",
      "530:\tlearn: 0.6534069\ttest: 0.4069139\tbest: 0.4069139 (530)\ttotal: 1.94s\tremaining: 1.71s\n",
      "531:\tlearn: 0.6529824\ttest: 0.4069139\tbest: 0.4069139 (530)\ttotal: 1.94s\tremaining: 1.71s\n",
      "532:\tlearn: 0.6525577\ttest: 0.4030609\tbest: 0.4069139 (530)\ttotal: 1.94s\tremaining: 1.7s\n",
      "533:\tlearn: 0.6534069\ttest: 0.4049891\tbest: 0.4069139 (530)\ttotal: 1.95s\tremaining: 1.7s\n",
      "534:\tlearn: 0.6556722\ttest: 0.4049891\tbest: 0.4069139 (530)\ttotal: 1.95s\tremaining: 1.7s\n",
      "535:\tlearn: 0.6565188\ttest: 0.4049891\tbest: 0.4069139 (530)\ttotal: 1.95s\tremaining: 1.69s\n",
      "536:\tlearn: 0.6577875\ttest: 0.4049891\tbest: 0.4069139 (530)\ttotal: 1.96s\tremaining: 1.69s\n",
      "537:\tlearn: 0.6594763\ttest: 0.4046278\tbest: 0.4069139 (530)\ttotal: 1.96s\tremaining: 1.68s\n",
      "538:\tlearn: 0.6598981\ttest: 0.4049891\tbest: 0.4069139 (530)\ttotal: 1.97s\tremaining: 1.68s\n",
      "539:\tlearn: 0.6597688\ttest: 0.4049891\tbest: 0.4069139 (530)\ttotal: 1.97s\tremaining: 1.68s\n",
      "540:\tlearn: 0.6601907\ttest: 0.4030609\tbest: 0.4069139 (530)\ttotal: 1.97s\tremaining: 1.67s\n",
      "541:\tlearn: 0.6597688\ttest: 0.4049891\tbest: 0.4069139 (530)\ttotal: 1.97s\tremaining: 1.67s\n",
      "542:\tlearn: 0.6597688\ttest: 0.4049891\tbest: 0.4069139 (530)\ttotal: 1.98s\tremaining: 1.66s\n",
      "543:\tlearn: 0.6597688\ttest: 0.4030609\tbest: 0.4069139 (530)\ttotal: 1.98s\tremaining: 1.66s\n",
      "544:\tlearn: 0.6608874\ttest: 0.4046278\tbest: 0.4069139 (530)\ttotal: 1.98s\tremaining: 1.66s\n",
      "545:\tlearn: 0.6607410\ttest: 0.4042672\tbest: 0.4069139 (530)\ttotal: 1.99s\tremaining: 1.65s\n",
      "546:\tlearn: 0.6617297\ttest: 0.4061891\tbest: 0.4069139 (530)\ttotal: 1.99s\tremaining: 1.65s\n",
      "547:\tlearn: 0.6620040\ttest: 0.4042672\tbest: 0.4069139 (530)\ttotal: 1.99s\tremaining: 1.64s\n",
      "548:\tlearn: 0.6620040\ttest: 0.4061891\tbest: 0.4069139 (530)\ttotal: 2s\tremaining: 1.64s\n",
      "549:\tlearn: 0.6625713\ttest: 0.4081077\tbest: 0.4081077 (549)\ttotal: 2s\tremaining: 1.64s\n",
      "550:\tlearn: 0.6615832\ttest: 0.4061891\tbest: 0.4081077 (549)\ttotal: 2s\tremaining: 1.63s\n",
      "551:\tlearn: 0.6620040\ttest: 0.4081077\tbest: 0.4081077 (549)\ttotal: 2.01s\tremaining: 1.63s\n",
      "552:\tlearn: 0.6628451\ttest: 0.4061891\tbest: 0.4081077 (549)\ttotal: 2.01s\tremaining: 1.62s\n",
      "553:\tlearn: 0.6624246\ttest: 0.4077449\tbest: 0.4081077 (549)\ttotal: 2.01s\tremaining: 1.62s\n",
      "554:\tlearn: 0.6628451\ttest: 0.4077449\tbest: 0.4081077 (549)\ttotal: 2.02s\tremaining: 1.62s\n",
      "555:\tlearn: 0.6632653\ttest: 0.4077449\tbest: 0.4081077 (549)\ttotal: 2.02s\tremaining: 1.61s\n",
      "556:\tlearn: 0.6632653\ttest: 0.4077449\tbest: 0.4081077 (549)\ttotal: 2.02s\tremaining: 1.61s\n",
      "557:\tlearn: 0.6621506\ttest: 0.4081077\tbest: 0.4081077 (549)\ttotal: 2.02s\tremaining: 1.6s\n",
      "558:\tlearn: 0.6604660\ttest: 0.4073827\tbest: 0.4081077 (549)\ttotal: 2.03s\tremaining: 1.6s\n",
      "559:\tlearn: 0.6608874\ttest: 0.4073827\tbest: 0.4081077 (549)\ttotal: 2.03s\tremaining: 1.6s\n",
      "560:\tlearn: 0.6617297\ttest: 0.4073827\tbest: 0.4081077 (549)\ttotal: 2.03s\tremaining: 1.59s\n",
      "561:\tlearn: 0.6629918\ttest: 0.4054670\tbest: 0.4081077 (549)\ttotal: 2.04s\tremaining: 1.59s\n",
      "562:\tlearn: 0.6629918\ttest: 0.4054670\tbest: 0.4081077 (549)\ttotal: 2.04s\tremaining: 1.58s\n",
      "563:\tlearn: 0.6634121\ttest: 0.4054670\tbest: 0.4081077 (549)\ttotal: 2.04s\tremaining: 1.58s\n",
      "564:\tlearn: 0.6634121\ttest: 0.4054670\tbest: 0.4081077 (549)\ttotal: 2.05s\tremaining: 1.58s\n",
      "565:\tlearn: 0.6638528\ttest: 0.4035478\tbest: 0.4081077 (549)\ttotal: 2.05s\tremaining: 1.57s\n",
      "566:\tlearn: 0.6638528\ttest: 0.4035478\tbest: 0.4081077 (549)\ttotal: 2.06s\tremaining: 1.57s\n",
      "567:\tlearn: 0.6645250\ttest: 0.4035478\tbest: 0.4081077 (549)\ttotal: 2.06s\tremaining: 1.56s\n",
      "568:\tlearn: 0.6653638\ttest: 0.4016252\tbest: 0.4081077 (549)\ttotal: 2.06s\tremaining: 1.56s\n",
      "569:\tlearn: 0.6666436\ttest: 0.4016252\tbest: 0.4081077 (549)\ttotal: 2.06s\tremaining: 1.56s\n",
      "570:\tlearn: 0.6670626\ttest: 0.3996992\tbest: 0.4081077 (549)\ttotal: 2.07s\tremaining: 1.55s\n",
      "571:\tlearn: 0.6662245\ttest: 0.3993432\tbest: 0.4081077 (549)\ttotal: 2.07s\tremaining: 1.55s\n",
      "572:\tlearn: 0.6662245\ttest: 0.3993432\tbest: 0.4081077 (549)\ttotal: 2.07s\tremaining: 1.55s\n",
      "573:\tlearn: 0.6673339\ttest: 0.4012679\tbest: 0.4081077 (549)\ttotal: 2.08s\tremaining: 1.54s\n",
      "574:\tlearn: 0.6684660\ttest: 0.3974152\tbest: 0.4081077 (549)\ttotal: 2.08s\tremaining: 1.54s\n",
      "575:\tlearn: 0.6684660\ttest: 0.3974152\tbest: 0.4081077 (549)\ttotal: 2.09s\tremaining: 1.53s\n",
      "576:\tlearn: 0.6687365\ttest: 0.3974152\tbest: 0.4081077 (549)\ttotal: 2.09s\tremaining: 1.53s\n",
      "577:\tlearn: 0.6676289\ttest: 0.3977697\tbest: 0.4081077 (549)\ttotal: 2.09s\tremaining: 1.53s\n",
      "578:\tlearn: 0.6681951\ttest: 0.3974152\tbest: 0.4081077 (549)\ttotal: 2.1s\tremaining: 1.52s\n",
      "579:\tlearn: 0.6680718\ttest: 0.3974152\tbest: 0.4081077 (549)\ttotal: 2.1s\tremaining: 1.52s\n",
      "580:\tlearn: 0.6701901\ttest: 0.3989879\tbest: 0.4081077 (549)\ttotal: 2.1s\tremaining: 1.52s\n",
      "581:\tlearn: 0.6689342\ttest: 0.4016252\tbest: 0.4081077 (549)\ttotal: 2.11s\tremaining: 1.51s\n",
      "582:\tlearn: 0.6710265\ttest: 0.3993432\tbest: 0.4081077 (549)\ttotal: 2.11s\tremaining: 1.51s\n",
      "583:\tlearn: 0.6696236\ttest: 0.3993432\tbest: 0.4081077 (549)\ttotal: 2.11s\tremaining: 1.5s\n",
      "584:\tlearn: 0.6700420\ttest: 0.3993432\tbest: 0.4081077 (549)\ttotal: 2.12s\tremaining: 1.5s\n",
      "585:\tlearn: 0.6692050\ttest: 0.3967080\tbest: 0.4081077 (549)\ttotal: 2.12s\tremaining: 1.5s\n",
      "586:\tlearn: 0.6708515\ttest: 0.3974152\tbest: 0.4081077 (549)\ttotal: 2.12s\tremaining: 1.49s\n",
      "587:\tlearn: 0.6715654\ttest: 0.3963554\tbest: 0.4081077 (549)\ttotal: 2.13s\tremaining: 1.49s\n",
      "588:\tlearn: 0.6715654\ttest: 0.3963554\tbest: 0.4081077 (549)\ttotal: 2.13s\tremaining: 1.49s\n",
      "589:\tlearn: 0.6719828\ttest: 0.3963554\tbest: 0.4081077 (549)\ttotal: 2.13s\tremaining: 1.48s\n",
      "590:\tlearn: 0.6732339\ttest: 0.3963554\tbest: 0.4081077 (549)\ttotal: 2.14s\tremaining: 1.48s\n",
      "591:\tlearn: 0.6740670\ttest: 0.3986333\tbest: 0.4081077 (549)\ttotal: 2.14s\tremaining: 1.48s\n",
      "592:\tlearn: 0.6744833\ttest: 0.4001997\tbest: 0.4081077 (549)\ttotal: 2.14s\tremaining: 1.47s\n",
      "593:\tlearn: 0.6753154\ttest: 0.4001997\tbest: 0.4081077 (549)\ttotal: 2.15s\tremaining: 1.47s\n",
      "594:\tlearn: 0.6768602\ttest: 0.3998448\tbest: 0.4081077 (549)\ttotal: 2.15s\tremaining: 1.46s\n",
      "595:\tlearn: 0.6764447\ttest: 0.4021167\tbest: 0.4081077 (549)\ttotal: 2.15s\tremaining: 1.46s\n",
      "596:\tlearn: 0.6778399\ttest: 0.4021167\tbest: 0.4081077 (549)\ttotal: 2.16s\tremaining: 1.46s\n",
      "597:\tlearn: 0.6778399\ttest: 0.4017605\tbest: 0.4081077 (549)\ttotal: 2.16s\tremaining: 1.45s\n",
      "598:\tlearn: 0.6779892\ttest: 0.4033158\tbest: 0.4081077 (549)\ttotal: 2.16s\tremaining: 1.45s\n",
      "599:\tlearn: 0.6782549\ttest: 0.4048657\tbest: 0.4081077 (549)\ttotal: 2.17s\tremaining: 1.45s\n",
      "600:\tlearn: 0.6782549\ttest: 0.4048657\tbest: 0.4081077 (549)\ttotal: 2.17s\tremaining: 1.44s\n",
      "601:\tlearn: 0.6792339\ttest: 0.4033158\tbest: 0.4081077 (549)\ttotal: 2.17s\tremaining: 1.44s\n",
      "602:\tlearn: 0.6804769\ttest: 0.4033158\tbest: 0.4081077 (549)\ttotal: 2.18s\tremaining: 1.43s\n",
      "603:\tlearn: 0.6806266\ttest: 0.4052233\tbest: 0.4081077 (549)\ttotal: 2.18s\tremaining: 1.43s\n",
      "604:\tlearn: 0.6814545\ttest: 0.4052233\tbest: 0.4081077 (549)\ttotal: 2.18s\tremaining: 1.43s\n",
      "605:\tlearn: 0.6810406\ttest: 0.4052233\tbest: 0.4081077 (549)\ttotal: 2.19s\tremaining: 1.42s\n",
      "606:\tlearn: 0.6810406\ttest: 0.4052233\tbest: 0.4081077 (549)\ttotal: 2.19s\tremaining: 1.42s\n",
      "607:\tlearn: 0.6811905\ttest: 0.4055817\tbest: 0.4081077 (549)\ttotal: 2.19s\tremaining: 1.42s\n",
      "608:\tlearn: 0.6819045\ttest: 0.4055817\tbest: 0.4081077 (549)\ttotal: 2.2s\tremaining: 1.41s\n",
      "609:\tlearn: 0.6834085\ttest: 0.4055817\tbest: 0.4081077 (549)\ttotal: 2.2s\tremaining: 1.41s\n",
      "610:\tlearn: 0.6842344\ttest: 0.4055817\tbest: 0.4081077 (549)\ttotal: 2.2s\tremaining: 1.4s\n",
      "611:\tlearn: 0.6842344\ttest: 0.4036728\tbest: 0.4081077 (549)\ttotal: 2.21s\tremaining: 1.4s\n",
      "612:\tlearn: 0.6875309\ttest: 0.4036728\tbest: 0.4081077 (549)\ttotal: 2.21s\tremaining: 1.4s\n",
      "613:\tlearn: 0.6887640\ttest: 0.4040303\tbest: 0.4081077 (549)\ttotal: 2.21s\tremaining: 1.39s\n",
      "614:\tlearn: 0.6901468\ttest: 0.4040303\tbest: 0.4081077 (549)\ttotal: 2.22s\tremaining: 1.39s\n",
      "615:\tlearn: 0.6920897\ttest: 0.4021167\tbest: 0.4081077 (549)\ttotal: 2.22s\tremaining: 1.38s\n",
      "616:\tlearn: 0.6916799\ttest: 0.4024735\tbest: 0.4081077 (549)\ttotal: 2.22s\tremaining: 1.38s\n",
      "617:\tlearn: 0.6918315\ttest: 0.4005551\tbest: 0.4081077 (549)\ttotal: 2.23s\tremaining: 1.38s\n",
      "618:\tlearn: 0.6922413\ttest: 0.4014049\tbest: 0.4081077 (549)\ttotal: 2.23s\tremaining: 1.37s\n",
      "619:\tlearn: 0.6929088\ttest: 0.4052233\tbest: 0.4081077 (549)\ttotal: 2.23s\tremaining: 1.37s\n",
      "620:\tlearn: 0.6929088\ttest: 0.4052233\tbest: 0.4081077 (549)\ttotal: 2.24s\tremaining: 1.36s\n",
      "621:\tlearn: 0.6950008\ttest: 0.4052233\tbest: 0.4081077 (549)\ttotal: 2.24s\tremaining: 1.36s\n",
      "622:\tlearn: 0.6940309\ttest: 0.4033158\tbest: 0.4081077 (549)\ttotal: 2.25s\tremaining: 1.36s\n",
      "623:\tlearn: 0.6952574\ttest: 0.4093893\tbest: 0.4093893 (623)\ttotal: 2.25s\tremaining: 1.35s\n",
      "624:\tlearn: 0.6966346\ttest: 0.4093893\tbest: 0.4093893 (623)\ttotal: 2.25s\tremaining: 1.35s\n",
      "625:\tlearn: 0.6970426\ttest: 0.4093893\tbest: 0.4093893 (623)\ttotal: 2.25s\tremaining: 1.35s\n",
      "626:\tlearn: 0.6967870\ttest: 0.4093893\tbest: 0.4093893 (623)\ttotal: 2.26s\tremaining: 1.34s\n",
      "627:\tlearn: 0.6967870\ttest: 0.4093893\tbest: 0.4093893 (623)\ttotal: 2.26s\tremaining: 1.34s\n",
      "628:\tlearn: 0.6958181\ttest: 0.4074872\tbest: 0.4093893 (623)\ttotal: 2.26s\tremaining: 1.33s\n",
      "629:\tlearn: 0.6962751\ttest: 0.4090283\tbest: 0.4093893 (623)\ttotal: 2.27s\tremaining: 1.33s\n",
      "630:\tlearn: 0.6965312\ttest: 0.4090283\tbest: 0.4093893 (623)\ttotal: 2.27s\tremaining: 1.33s\n",
      "631:\tlearn: 0.6970920\ttest: 0.4109257\tbest: 0.4109257 (631)\ttotal: 2.27s\tremaining: 1.32s\n",
      "632:\tlearn: 0.6977555\ttest: 0.4109257\tbest: 0.4109257 (631)\ttotal: 2.28s\tremaining: 1.32s\n",
      "633:\tlearn: 0.6988255\ttest: 0.4071275\tbest: 0.4109257 (631)\ttotal: 2.28s\tremaining: 1.32s\n",
      "634:\tlearn: 0.6985708\ttest: 0.4090283\tbest: 0.4109257 (631)\ttotal: 2.28s\tremaining: 1.31s\n",
      "635:\tlearn: 0.6991310\ttest: 0.4090283\tbest: 0.4109257 (631)\ttotal: 2.29s\tremaining: 1.31s\n",
      "636:\tlearn: 0.6995383\ttest: 0.4090283\tbest: 0.4109257 (631)\ttotal: 2.29s\tremaining: 1.3s\n",
      "637:\tlearn: 0.7011656\ttest: 0.4071275\tbest: 0.4109257 (631)\ttotal: 2.29s\tremaining: 1.3s\n",
      "638:\tlearn: 0.7011656\ttest: 0.4071275\tbest: 0.4109257 (631)\ttotal: 2.3s\tremaining: 1.3s\n",
      "639:\tlearn: 0.7013187\ttest: 0.4067685\tbest: 0.4109257 (631)\ttotal: 2.3s\tremaining: 1.29s\n",
      "640:\tlearn: 0.7014720\ttest: 0.4067685\tbest: 0.4109257 (631)\ttotal: 2.3s\tremaining: 1.29s\n",
      "641:\tlearn: 0.7021314\ttest: 0.4067685\tbest: 0.4109257 (631)\ttotal: 2.31s\tremaining: 1.29s\n",
      "642:\tlearn: 0.7025375\ttest: 0.4048657\tbest: 0.4109257 (631)\ttotal: 2.31s\tremaining: 1.28s\n",
      "643:\tlearn: 0.7041601\ttest: 0.4067685\tbest: 0.4109257 (631)\ttotal: 2.31s\tremaining: 1.28s\n",
      "644:\tlearn: 0.7039083\ttest: 0.4067685\tbest: 0.4109257 (631)\ttotal: 2.32s\tremaining: 1.27s\n",
      "645:\tlearn: 0.7040065\ttest: 0.4067685\tbest: 0.4109257 (631)\ttotal: 2.32s\tremaining: 1.27s\n",
      "646:\tlearn: 0.7045653\ttest: 0.4074872\tbest: 0.4109257 (631)\ttotal: 2.32s\tremaining: 1.27s\n",
      "647:\tlearn: 0.7055289\ttest: 0.4074872\tbest: 0.4109257 (631)\ttotal: 2.33s\tremaining: 1.26s\n",
      "648:\tlearn: 0.7056261\ttest: 0.4059406\tbest: 0.4109257 (631)\ttotal: 2.33s\tremaining: 1.26s\n",
      "649:\tlearn: 0.7052214\ttest: 0.4059406\tbest: 0.4109257 (631)\ttotal: 2.33s\tremaining: 1.26s\n",
      "650:\tlearn: 0.7056261\ttest: 0.4059406\tbest: 0.4109257 (631)\ttotal: 2.34s\tremaining: 1.25s\n",
      "651:\tlearn: 0.7061843\ttest: 0.4078475\tbest: 0.4109257 (631)\ttotal: 2.34s\tremaining: 1.25s\n",
      "652:\tlearn: 0.7061843\ttest: 0.4074872\tbest: 0.4109257 (631)\ttotal: 2.34s\tremaining: 1.25s\n",
      "653:\tlearn: 0.7065886\ttest: 0.4071275\tbest: 0.4109257 (631)\ttotal: 2.35s\tremaining: 1.24s\n",
      "654:\tlearn: 0.7061843\ttest: 0.4052233\tbest: 0.4109257 (631)\ttotal: 2.35s\tremaining: 1.24s\n",
      "655:\tlearn: 0.7065886\ttest: 0.4055817\tbest: 0.4109257 (631)\ttotal: 2.35s\tremaining: 1.23s\n",
      "656:\tlearn: 0.7073967\ttest: 0.4078475\tbest: 0.4109257 (631)\ttotal: 2.36s\tremaining: 1.23s\n",
      "657:\tlearn: 0.7069928\ttest: 0.4097510\tbest: 0.4109257 (631)\ttotal: 2.36s\tremaining: 1.23s\n",
      "658:\tlearn: 0.7075508\ttest: 0.4097510\tbest: 0.4109257 (631)\ttotal: 2.36s\tremaining: 1.22s\n",
      "659:\tlearn: 0.7081088\ttest: 0.4074872\tbest: 0.4109257 (631)\ttotal: 2.37s\tremaining: 1.22s\n",
      "660:\tlearn: 0.7085125\ttest: 0.4074872\tbest: 0.4109257 (631)\ttotal: 2.37s\tremaining: 1.22s\n",
      "661:\tlearn: 0.7089160\ttest: 0.4071275\tbest: 0.4109257 (631)\ttotal: 2.37s\tremaining: 1.21s\n",
      "662:\tlearn: 0.7092248\ttest: 0.4093893\tbest: 0.4109257 (631)\ttotal: 2.37s\tremaining: 1.21s\n",
      "663:\tlearn: 0.7100316\ttest: 0.4128198\tbest: 0.4128198 (663)\ttotal: 2.38s\tremaining: 1.2s\n",
      "664:\tlearn: 0.7104347\ttest: 0.4147106\tbest: 0.4147106 (664)\ttotal: 2.38s\tremaining: 1.2s\n",
      "665:\tlearn: 0.7134067\ttest: 0.4147106\tbest: 0.4147106 (664)\ttotal: 2.38s\tremaining: 1.2s\n",
      "666:\tlearn: 0.7130047\ttest: 0.4147106\tbest: 0.4147106 (664)\ttotal: 2.39s\tremaining: 1.19s\n",
      "667:\tlearn: 0.7142101\ttest: 0.4147106\tbest: 0.4147106 (664)\ttotal: 2.39s\tremaining: 1.19s\n",
      "668:\tlearn: 0.7154138\ttest: 0.4154414\tbest: 0.4154414 (668)\ttotal: 2.39s\tremaining: 1.18s\n",
      "669:\tlearn: 0.7150128\ttest: 0.4154414\tbest: 0.4154414 (668)\ttotal: 2.4s\tremaining: 1.18s\n",
      "670:\tlearn: 0.7147667\ttest: 0.4135479\tbest: 0.4154414 (668)\ttotal: 2.4s\tremaining: 1.18s\n",
      "671:\tlearn: 0.7151681\ttest: 0.4131835\tbest: 0.4154414 (668)\ttotal: 2.4s\tremaining: 1.17s\n",
      "672:\tlearn: 0.7151681\ttest: 0.4131835\tbest: 0.4154414 (668)\ttotal: 2.41s\tremaining: 1.17s\n",
      "673:\tlearn: 0.7159701\ttest: 0.4135479\tbest: 0.4154414 (668)\ttotal: 2.41s\tremaining: 1.17s\n",
      "674:\tlearn: 0.7164606\ttest: 0.4139129\tbest: 0.4154414 (668)\ttotal: 2.41s\tremaining: 1.16s\n",
      "675:\tlearn: 0.7172611\ttest: 0.4131835\tbest: 0.4154414 (668)\ttotal: 2.42s\tremaining: 1.16s\n",
      "676:\tlearn: 0.7196584\ttest: 0.4131835\tbest: 0.4154414 (668)\ttotal: 2.42s\tremaining: 1.15s\n",
      "677:\tlearn: 0.7194151\ttest: 0.4131835\tbest: 0.4154414 (668)\ttotal: 2.42s\tremaining: 1.15s\n",
      "678:\tlearn: 0.7200574\ttest: 0.4150757\tbest: 0.4154414 (668)\ttotal: 2.42s\tremaining: 1.15s\n",
      "679:\tlearn: 0.7200574\ttest: 0.4150757\tbest: 0.4154414 (668)\ttotal: 2.43s\tremaining: 1.14s\n",
      "680:\tlearn: 0.7216514\ttest: 0.4150757\tbest: 0.4154414 (668)\ttotal: 2.43s\tremaining: 1.14s\n",
      "681:\tlearn: 0.7218934\ttest: 0.4128198\tbest: 0.4154414 (668)\ttotal: 2.44s\tremaining: 1.14s\n",
      "682:\tlearn: 0.7226890\ttest: 0.4128198\tbest: 0.4154414 (668)\ttotal: 2.44s\tremaining: 1.13s\n",
      "683:\tlearn: 0.7230865\ttest: 0.4143461\tbest: 0.4154414 (668)\ttotal: 2.44s\tremaining: 1.13s\n",
      "684:\tlearn: 0.7224474\ttest: 0.4124567\tbest: 0.4154414 (668)\ttotal: 2.44s\tremaining: 1.12s\n",
      "685:\tlearn: 0.7237965\ttest: 0.4120942\tbest: 0.4154414 (668)\ttotal: 2.45s\tremaining: 1.12s\n",
      "686:\tlearn: 0.7233991\ttest: 0.4128198\tbest: 0.4154414 (668)\ttotal: 2.45s\tremaining: 1.12s\n",
      "687:\tlearn: 0.7245069\ttest: 0.4109257\tbest: 0.4154414 (668)\ttotal: 2.45s\tremaining: 1.11s\n",
      "688:\tlearn: 0.7241095\ttest: 0.4112881\tbest: 0.4154414 (668)\ttotal: 2.46s\tremaining: 1.11s\n",
      "689:\tlearn: 0.7241095\ttest: 0.4112881\tbest: 0.4154414 (668)\ttotal: 2.46s\tremaining: 1.1s\n",
      "690:\tlearn: 0.7245069\ttest: 0.4102028\tbest: 0.4154414 (668)\ttotal: 2.46s\tremaining: 1.1s\n",
      "691:\tlearn: 0.7260947\ttest: 0.4105640\tbest: 0.4154414 (668)\ttotal: 2.47s\tremaining: 1.1s\n",
      "692:\tlearn: 0.7256980\ttest: 0.4105640\tbest: 0.4154414 (668)\ttotal: 2.47s\tremaining: 1.09s\n",
      "693:\tlearn: 0.7264912\ttest: 0.4124567\tbest: 0.4154414 (668)\ttotal: 2.47s\tremaining: 1.09s\n",
      "694:\tlearn: 0.7267307\ttest: 0.4124567\tbest: 0.4154414 (668)\ttotal: 2.48s\tremaining: 1.09s\n",
      "695:\tlearn: 0.7271268\ttest: 0.4139823\tbest: 0.4154414 (668)\ttotal: 2.48s\tremaining: 1.08s\n",
      "696:\tlearn: 0.7275228\ttest: 0.4139823\tbest: 0.4154414 (668)\ttotal: 2.48s\tremaining: 1.08s\n",
      "697:\tlearn: 0.7279186\ttest: 0.4143461\tbest: 0.4154414 (668)\ttotal: 2.48s\tremaining: 1.07s\n",
      "698:\tlearn: 0.7285526\ttest: 0.4158671\tbest: 0.4158671 (698)\ttotal: 2.49s\tremaining: 1.07s\n",
      "699:\tlearn: 0.7289478\ttest: 0.4139823\tbest: 0.4158671 (698)\ttotal: 2.49s\tremaining: 1.07s\n",
      "700:\tlearn: 0.7302897\ttest: 0.4139823\tbest: 0.4158671 (698)\ttotal: 2.5s\tremaining: 1.06s\n",
      "701:\tlearn: 0.7302897\ttest: 0.4120942\tbest: 0.4158671 (698)\ttotal: 2.5s\tremaining: 1.06s\n",
      "702:\tlearn: 0.7306843\ttest: 0.4124567\tbest: 0.4158671 (698)\ttotal: 2.5s\tremaining: 1.06s\n",
      "703:\tlearn: 0.7306843\ttest: 0.4124567\tbest: 0.4158671 (698)\ttotal: 2.5s\tremaining: 1.05s\n",
      "704:\tlearn: 0.7297377\ttest: 0.4143461\tbest: 0.4158671 (698)\ttotal: 2.51s\tremaining: 1.05s\n",
      "705:\tlearn: 0.7301325\ttest: 0.4143461\tbest: 0.4158671 (698)\ttotal: 2.51s\tremaining: 1.04s\n",
      "706:\tlearn: 0.7305270\ttest: 0.4143461\tbest: 0.4158671 (698)\ttotal: 2.51s\tremaining: 1.04s\n",
      "707:\tlearn: 0.7313156\ttest: 0.4143461\tbest: 0.4158671 (698)\ttotal: 2.52s\tremaining: 1.04s\n",
      "708:\tlearn: 0.7313156\ttest: 0.4158671\tbest: 0.4158671 (698)\ttotal: 2.52s\tremaining: 1.03s\n",
      "709:\tlearn: 0.7318671\ttest: 0.4177486\tbest: 0.4177486 (709)\ttotal: 2.52s\tremaining: 1.03s\n",
      "710:\tlearn: 0.7326548\ttest: 0.4185272\tbest: 0.4185272 (710)\ttotal: 2.53s\tremaining: 1.03s\n",
      "711:\tlearn: 0.7322610\ttest: 0.4162891\tbest: 0.4185272 (710)\ttotal: 2.53s\tremaining: 1.02s\n",
      "712:\tlearn: 0.7324186\ttest: 0.4159258\tbest: 0.4185272 (710)\ttotal: 2.53s\tremaining: 1.02s\n",
      "713:\tlearn: 0.7316305\ttest: 0.4159258\tbest: 0.4185272 (710)\ttotal: 2.54s\tremaining: 1.02s\n",
      "714:\tlearn: 0.7317090\ttest: 0.4159258\tbest: 0.4185272 (710)\ttotal: 2.54s\tremaining: 1.01s\n",
      "715:\tlearn: 0.7323400\ttest: 0.4155632\tbest: 0.4185272 (710)\ttotal: 2.54s\tremaining: 1.01s\n",
      "716:\tlearn: 0.7327340\ttest: 0.4159258\tbest: 0.4185272 (710)\ttotal: 2.55s\tremaining: 1s\n",
      "717:\tlearn: 0.7343086\ttest: 0.4140510\tbest: 0.4185272 (710)\ttotal: 2.55s\tremaining: 1s\n",
      "718:\tlearn: 0.7356459\ttest: 0.4136897\tbest: 0.4185272 (710)\ttotal: 2.55s\tremaining: 998ms\n",
      "719:\tlearn: 0.7356459\ttest: 0.4136897\tbest: 0.4185272 (710)\ttotal: 2.56s\tremaining: 994ms\n",
      "720:\tlearn: 0.7350949\ttest: 0.4114535\tbest: 0.4185272 (710)\ttotal: 2.56s\tremaining: 991ms\n",
      "721:\tlearn: 0.7348599\ttest: 0.4095748\tbest: 0.4185272 (710)\ttotal: 2.56s\tremaining: 987ms\n",
      "722:\tlearn: 0.7354878\ttest: 0.4136897\tbest: 0.4185272 (710)\ttotal: 2.57s\tremaining: 983ms\n",
      "723:\tlearn: 0.7350949\ttest: 0.4155632\tbest: 0.4185272 (710)\ttotal: 2.57s\tremaining: 979ms\n",
      "724:\tlearn: 0.7352530\ttest: 0.4155632\tbest: 0.4185272 (710)\ttotal: 2.57s\tremaining: 976ms\n",
      "725:\tlearn: 0.7352530\ttest: 0.4155632\tbest: 0.4185272 (710)\ttotal: 2.58s\tremaining: 972ms\n",
      "726:\tlearn: 0.7356459\ttest: 0.4155632\tbest: 0.4185272 (710)\ttotal: 2.58s\tremaining: 968ms\n",
      "727:\tlearn: 0.7360387\ttest: 0.4170701\tbest: 0.4185272 (710)\ttotal: 2.58s\tremaining: 965ms\n",
      "728:\tlearn: 0.7360387\ttest: 0.4170701\tbest: 0.4185272 (710)\ttotal: 2.58s\tremaining: 961ms\n",
      "729:\tlearn: 0.7372160\ttest: 0.4170701\tbest: 0.4185272 (710)\ttotal: 2.59s\tremaining: 958ms\n",
      "730:\tlearn: 0.7368237\ttest: 0.4155632\tbest: 0.4185272 (710)\ttotal: 2.59s\tremaining: 954ms\n",
      "731:\tlearn: 0.7370576\ttest: 0.4155632\tbest: 0.4185272 (710)\ttotal: 2.6s\tremaining: 951ms\n",
      "732:\tlearn: 0.7374497\ttest: 0.4174334\tbest: 0.4185272 (710)\ttotal: 2.6s\tremaining: 947ms\n",
      "733:\tlearn: 0.7380748\ttest: 0.4136897\tbest: 0.4185272 (710)\ttotal: 2.6s\tremaining: 944ms\n",
      "734:\tlearn: 0.7382332\ttest: 0.4136897\tbest: 0.4185272 (710)\ttotal: 2.61s\tremaining: 940ms\n",
      "735:\tlearn: 0.7382332\ttest: 0.4155632\tbest: 0.4185272 (710)\ttotal: 2.61s\tremaining: 936ms\n",
      "736:\tlearn: 0.7401892\ttest: 0.4155632\tbest: 0.4185272 (710)\ttotal: 2.61s\tremaining: 933ms\n",
      "737:\tlearn: 0.7415197\ttest: 0.4152012\tbest: 0.4185272 (710)\ttotal: 2.62s\tremaining: 929ms\n",
      "738:\tlearn: 0.7412883\ttest: 0.4140510\tbest: 0.4185272 (710)\ttotal: 2.62s\tremaining: 925ms\n",
      "739:\tlearn: 0.7405068\ttest: 0.4140510\tbest: 0.4185272 (710)\ttotal: 2.62s\tremaining: 922ms\n",
      "740:\tlearn: 0.7408976\ttest: 0.4140510\tbest: 0.4185272 (710)\ttotal: 2.63s\tremaining: 918ms\n",
      "741:\tlearn: 0.7402747\ttest: 0.4140510\tbest: 0.4185272 (710)\ttotal: 2.63s\tremaining: 914ms\n",
      "742:\tlearn: 0.7395660\ttest: 0.4140510\tbest: 0.4185272 (710)\ttotal: 2.63s\tremaining: 911ms\n",
      "743:\tlearn: 0.7406657\ttest: 0.4140510\tbest: 0.4185272 (710)\ttotal: 2.64s\tremaining: 907ms\n",
      "744:\tlearn: 0.7423153\ttest: 0.4133290\tbest: 0.4185272 (710)\ttotal: 2.64s\tremaining: 903ms\n",
      "745:\tlearn: 0.7444962\ttest: 0.4136897\tbest: 0.4185272 (710)\ttotal: 2.64s\tremaining: 900ms\n",
      "746:\tlearn: 0.7444962\ttest: 0.4136897\tbest: 0.4185272 (710)\ttotal: 2.65s\tremaining: 896ms\n",
      "747:\tlearn: 0.7443367\ttest: 0.4133290\tbest: 0.4185272 (710)\ttotal: 2.65s\tremaining: 893ms\n",
      "748:\tlearn: 0.7443367\ttest: 0.4133290\tbest: 0.4185272 (710)\ttotal: 2.65s\tremaining: 889ms\n",
      "749:\tlearn: 0.7448857\ttest: 0.4133290\tbest: 0.4185272 (710)\ttotal: 2.65s\tremaining: 885ms\n",
      "750:\tlearn: 0.7460530\ttest: 0.4133290\tbest: 0.4185272 (710)\ttotal: 2.66s\tremaining: 882ms\n",
      "751:\tlearn: 0.7456640\ttest: 0.4129689\tbest: 0.4185272 (710)\ttotal: 2.66s\tremaining: 878ms\n",
      "752:\tlearn: 0.7464418\ttest: 0.4129689\tbest: 0.4185272 (710)\ttotal: 2.67s\tremaining: 874ms\n",
      "753:\tlearn: 0.7464418\ttest: 0.4129689\tbest: 0.4185272 (710)\ttotal: 2.67s\tremaining: 871ms\n",
      "754:\tlearn: 0.7460530\ttest: 0.4148398\tbest: 0.4185272 (710)\ttotal: 2.67s\tremaining: 867ms\n",
      "755:\tlearn: 0.7469902\ttest: 0.4148398\tbest: 0.4185272 (710)\ttotal: 2.67s\tremaining: 863ms\n",
      "756:\tlearn: 0.7473787\ttest: 0.4129689\tbest: 0.4185272 (710)\ttotal: 2.68s\tremaining: 860ms\n",
      "757:\tlearn: 0.7472188\ttest: 0.4129689\tbest: 0.4185272 (710)\ttotal: 2.68s\tremaining: 856ms\n",
      "758:\tlearn: 0.7494789\ttest: 0.4133290\tbest: 0.4185272 (710)\ttotal: 2.68s\tremaining: 853ms\n",
      "759:\tlearn: 0.7490912\ttest: 0.4152012\tbest: 0.4185272 (710)\ttotal: 2.69s\tremaining: 849ms\n",
      "760:\tlearn: 0.7504142\ttest: 0.4152012\tbest: 0.4185272 (710)\ttotal: 2.69s\tremaining: 845ms\n",
      "761:\tlearn: 0.7510280\ttest: 0.4152012\tbest: 0.4185272 (710)\ttotal: 2.69s\tremaining: 842ms\n",
      "762:\tlearn: 0.7515754\ttest: 0.4174334\tbest: 0.4185272 (710)\ttotal: 2.7s\tremaining: 838ms\n",
      "763:\tlearn: 0.7515754\ttest: 0.4174334\tbest: 0.4185272 (710)\ttotal: 2.7s\tremaining: 835ms\n",
      "764:\tlearn: 0.7519622\ttest: 0.4174334\tbest: 0.4185272 (710)\ttotal: 2.7s\tremaining: 831ms\n",
      "765:\tlearn: 0.7512545\ttest: 0.4170701\tbest: 0.4185272 (710)\ttotal: 2.71s\tremaining: 827ms\n",
      "766:\tlearn: 0.7517068\ttest: 0.4226575\tbest: 0.4226575 (766)\ttotal: 2.71s\tremaining: 824ms\n",
      "767:\tlearn: 0.7517068\ttest: 0.4226575\tbest: 0.4226575 (766)\ttotal: 2.71s\tremaining: 820ms\n",
      "768:\tlearn: 0.7513204\ttest: 0.4189358\tbest: 0.4226575 (766)\ttotal: 2.72s\tremaining: 816ms\n",
      "769:\tlearn: 0.7509338\ttest: 0.4193004\tbest: 0.4226575 (766)\ttotal: 2.72s\tremaining: 813ms\n",
      "770:\tlearn: 0.7518672\ttest: 0.4193004\tbest: 0.4226575 (766)\ttotal: 2.73s\tremaining: 809ms\n",
      "771:\tlearn: 0.7514807\ttest: 0.4211642\tbest: 0.4226575 (766)\ttotal: 2.73s\tremaining: 806ms\n",
      "772:\tlearn: 0.7522535\ttest: 0.4193004\tbest: 0.4226575 (766)\ttotal: 2.73s\tremaining: 803ms\n",
      "773:\tlearn: 0.7510941\ttest: 0.4177974\tbest: 0.4226575 (766)\ttotal: 2.73s\tremaining: 799ms\n",
      "774:\tlearn: 0.7510941\ttest: 0.4177974\tbest: 0.4226575 (766)\ttotal: 2.74s\tremaining: 796ms\n",
      "775:\tlearn: 0.7510941\ttest: 0.4196657\tbest: 0.4226575 (766)\ttotal: 2.74s\tremaining: 792ms\n",
      "776:\tlearn: 0.7510941\ttest: 0.4196657\tbest: 0.4226575 (766)\ttotal: 2.75s\tremaining: 788ms\n",
      "777:\tlearn: 0.7507073\ttest: 0.4196657\tbest: 0.4226575 (766)\ttotal: 2.75s\tremaining: 785ms\n",
      "778:\tlearn: 0.7508676\ttest: 0.4196657\tbest: 0.4226575 (766)\ttotal: 2.75s\tremaining: 781ms\n",
      "779:\tlearn: 0.7520276\ttest: 0.4177974\tbest: 0.4226575 (766)\ttotal: 2.76s\tremaining: 778ms\n",
      "780:\tlearn: 0.7535720\ttest: 0.4181620\tbest: 0.4226575 (766)\ttotal: 2.76s\tremaining: 774ms\n",
      "781:\tlearn: 0.7543432\ttest: 0.4181620\tbest: 0.4226575 (766)\ttotal: 2.76s\tremaining: 770ms\n",
      "782:\tlearn: 0.7551770\ttest: 0.4181620\tbest: 0.4226575 (766)\ttotal: 2.77s\tremaining: 767ms\n",
      "783:\tlearn: 0.7544070\ttest: 0.4177974\tbest: 0.4226575 (766)\ttotal: 2.77s\tremaining: 763ms\n",
      "784:\tlearn: 0.7544070\ttest: 0.4177974\tbest: 0.4226575 (766)\ttotal: 2.77s\tremaining: 760ms\n",
      "785:\tlearn: 0.7536363\ttest: 0.4159258\tbest: 0.4226575 (766)\ttotal: 2.78s\tremaining: 756ms\n",
      "786:\tlearn: 0.7538611\ttest: 0.4159258\tbest: 0.4226575 (766)\ttotal: 2.78s\tremaining: 752ms\n",
      "787:\tlearn: 0.7555618\ttest: 0.4174334\tbest: 0.4226575 (766)\ttotal: 2.78s\tremaining: 749ms\n",
      "788:\tlearn: 0.7551770\ttest: 0.4174334\tbest: 0.4226575 (766)\ttotal: 2.79s\tremaining: 745ms\n",
      "789:\tlearn: 0.7574831\ttest: 0.4174334\tbest: 0.4226575 (766)\ttotal: 2.79s\tremaining: 741ms\n",
      "790:\tlearn: 0.7591786\ttest: 0.4144129\tbest: 0.4226575 (766)\ttotal: 2.79s\tremaining: 738ms\n",
      "791:\tlearn: 0.7582505\ttest: 0.4155632\tbest: 0.4226575 (766)\ttotal: 2.79s\tremaining: 734ms\n",
      "792:\tlearn: 0.7601062\ttest: 0.4177974\tbest: 0.4226575 (766)\ttotal: 2.8s\tremaining: 731ms\n",
      "793:\tlearn: 0.7604891\ttest: 0.4211642\tbest: 0.4226575 (766)\ttotal: 2.8s\tremaining: 727ms\n",
      "794:\tlearn: 0.7616369\ttest: 0.4230247\tbest: 0.4230247 (794)\ttotal: 2.8s\tremaining: 723ms\n",
      "795:\tlearn: 0.7612544\ttest: 0.4230247\tbest: 0.4230247 (794)\ttotal: 2.81s\tremaining: 720ms\n",
      "796:\tlearn: 0.7611952\ttest: 0.4245135\tbest: 0.4245135 (796)\ttotal: 2.81s\tremaining: 716ms\n",
      "797:\tlearn: 0.7617986\ttest: 0.4241456\tbest: 0.4245135 (796)\ttotal: 2.81s\tremaining: 712ms\n",
      "798:\tlearn: 0.7617986\ttest: 0.4207983\tbest: 0.4245135 (796)\ttotal: 2.82s\tremaining: 709ms\n",
      "799:\tlearn: 0.7623427\ttest: 0.4211642\tbest: 0.4245135 (796)\ttotal: 2.82s\tremaining: 705ms\n",
      "800:\tlearn: 0.7631069\ttest: 0.4230247\tbest: 0.4245135 (796)\ttotal: 2.82s\tremaining: 701ms\n",
      "801:\tlearn: 0.7627249\ttest: 0.4233926\tbest: 0.4245135 (796)\ttotal: 2.83s\tremaining: 698ms\n",
      "802:\tlearn: 0.7627831\ttest: 0.4233926\tbest: 0.4245135 (796)\ttotal: 2.83s\tremaining: 694ms\n",
      "803:\tlearn: 0.7631649\ttest: 0.4233926\tbest: 0.4245135 (796)\ttotal: 2.83s\tremaining: 691ms\n",
      "804:\tlearn: 0.7641472\ttest: 0.4230247\tbest: 0.4245135 (796)\ttotal: 2.83s\tremaining: 687ms\n",
      "805:\tlearn: 0.7633846\ttest: 0.4230247\tbest: 0.4245135 (796)\ttotal: 2.84s\tremaining: 683ms\n",
      "806:\tlearn: 0.7636041\ttest: 0.4230247\tbest: 0.4245135 (796)\ttotal: 2.84s\tremaining: 680ms\n",
      "807:\tlearn: 0.7645852\ttest: 0.4211642\tbest: 0.4245135 (796)\ttotal: 2.85s\tremaining: 676ms\n",
      "808:\tlearn: 0.7649659\ttest: 0.4211642\tbest: 0.4245135 (796)\ttotal: 2.85s\tremaining: 672ms\n",
      "809:\tlearn: 0.7653464\ttest: 0.4211642\tbest: 0.4245135 (796)\ttotal: 2.85s\tremaining: 669ms\n",
      "810:\tlearn: 0.7672464\ttest: 0.4230247\tbest: 0.4245135 (796)\ttotal: 2.85s\tremaining: 665ms\n",
      "811:\tlearn: 0.7672464\ttest: 0.4230247\tbest: 0.4245135 (796)\ttotal: 2.86s\tremaining: 662ms\n",
      "812:\tlearn: 0.7682222\ttest: 0.4222909\tbest: 0.4245135 (796)\ttotal: 2.86s\tremaining: 658ms\n",
      "813:\tlearn: 0.7682222\ttest: 0.4222909\tbest: 0.4245135 (796)\ttotal: 2.86s\tremaining: 654ms\n",
      "814:\tlearn: 0.7674637\ttest: 0.4222909\tbest: 0.4245135 (796)\ttotal: 2.87s\tremaining: 651ms\n",
      "815:\tlearn: 0.7686012\ttest: 0.4207983\tbest: 0.4245135 (796)\ttotal: 2.87s\tremaining: 647ms\n",
      "816:\tlearn: 0.7691424\ttest: 0.4204330\tbest: 0.4245135 (796)\ttotal: 2.87s\tremaining: 644ms\n",
      "817:\tlearn: 0.7683845\ttest: 0.4200683\tbest: 0.4245135 (796)\ttotal: 2.88s\tremaining: 640ms\n",
      "818:\tlearn: 0.7695211\ttest: 0.4182085\tbest: 0.4245135 (796)\ttotal: 2.88s\tremaining: 636ms\n",
      "819:\tlearn: 0.7697372\ttest: 0.4182085\tbest: 0.4245135 (796)\ttotal: 2.88s\tremaining: 633ms\n",
      "820:\tlearn: 0.7702250\ttest: 0.4182085\tbest: 0.4245135 (796)\ttotal: 2.89s\tremaining: 629ms\n",
      "821:\tlearn: 0.7702250\ttest: 0.4178457\tbest: 0.4245135 (796)\ttotal: 2.89s\tremaining: 626ms\n",
      "822:\tlearn: 0.7703877\ttest: 0.4178457\tbest: 0.4245135 (796)\ttotal: 2.89s\tremaining: 622ms\n",
      "823:\tlearn: 0.7706035\ttest: 0.4200683\tbest: 0.4245135 (796)\ttotal: 2.9s\tremaining: 618ms\n",
      "824:\tlearn: 0.7704407\ttest: 0.4197043\tbest: 0.4245135 (796)\ttotal: 2.9s\tremaining: 615ms\n",
      "825:\tlearn: 0.7708190\ttest: 0.4200683\tbest: 0.4245135 (796)\ttotal: 2.9s\tremaining: 611ms\n",
      "826:\tlearn: 0.7715751\ttest: 0.4197043\tbest: 0.4245135 (796)\ttotal: 2.9s\tremaining: 608ms\n",
      "827:\tlearn: 0.7715751\ttest: 0.4197043\tbest: 0.4245135 (796)\ttotal: 2.91s\tremaining: 604ms\n",
      "828:\tlearn: 0.7711972\ttest: 0.4178457\tbest: 0.4245135 (796)\ttotal: 2.91s\tremaining: 600ms\n",
      "829:\tlearn: 0.7745932\ttest: 0.4156231\tbest: 0.4245135 (796)\ttotal: 2.91s\tremaining: 597ms\n",
      "830:\tlearn: 0.7745932\ttest: 0.4156231\tbest: 0.4245135 (796)\ttotal: 2.92s\tremaining: 593ms\n",
      "831:\tlearn: 0.7742165\ttest: 0.4178457\tbest: 0.4245135 (796)\ttotal: 2.92s\tremaining: 590ms\n",
      "832:\tlearn: 0.7745932\ttest: 0.4159839\tbest: 0.4245135 (796)\ttotal: 2.92s\tremaining: 586ms\n",
      "833:\tlearn: 0.7757223\ttest: 0.4152630\tbest: 0.4245135 (796)\ttotal: 2.93s\tremaining: 583ms\n",
      "834:\tlearn: 0.7764743\ttest: 0.4152630\tbest: 0.4245135 (796)\ttotal: 2.93s\tremaining: 579ms\n",
      "835:\tlearn: 0.7770622\ttest: 0.4189781\tbest: 0.4245135 (796)\ttotal: 2.93s\tremaining: 576ms\n",
      "836:\tlearn: 0.7770622\ttest: 0.4189781\tbest: 0.4245135 (796)\ttotal: 2.94s\tremaining: 572ms\n",
      "837:\tlearn: 0.7770622\ttest: 0.4189781\tbest: 0.4245135 (796)\ttotal: 2.94s\tremaining: 568ms\n",
      "838:\tlearn: 0.7774376\ttest: 0.4189781\tbest: 0.4245135 (796)\ttotal: 2.94s\tremaining: 565ms\n",
      "839:\tlearn: 0.7774376\ttest: 0.4189781\tbest: 0.4245135 (796)\ttotal: 2.95s\tremaining: 561ms\n",
      "840:\tlearn: 0.7772256\ttest: 0.4171221\tbest: 0.4245135 (796)\ttotal: 2.95s\tremaining: 558ms\n",
      "841:\tlearn: 0.7772256\ttest: 0.4171221\tbest: 0.4245135 (796)\ttotal: 2.95s\tremaining: 554ms\n",
      "842:\tlearn: 0.7776011\ttest: 0.4171221\tbest: 0.4245135 (796)\ttotal: 2.96s\tremaining: 551ms\n",
      "843:\tlearn: 0.7787264\ttest: 0.4171221\tbest: 0.4245135 (796)\ttotal: 2.96s\tremaining: 547ms\n",
      "844:\tlearn: 0.7800142\ttest: 0.4208309\tbest: 0.4245135 (796)\ttotal: 2.96s\tremaining: 543ms\n",
      "845:\tlearn: 0.7805525\ttest: 0.4186160\tbest: 0.4245135 (796)\ttotal: 2.96s\tremaining: 540ms\n",
      "846:\tlearn: 0.7786788\ttest: 0.4145445\tbest: 0.4245135 (796)\ttotal: 2.97s\tremaining: 536ms\n",
      "847:\tlearn: 0.7779283\ttest: 0.4145445\tbest: 0.4245135 (796)\ttotal: 2.97s\tremaining: 533ms\n",
      "848:\tlearn: 0.7790539\ttest: 0.4182544\tbest: 0.4245135 (796)\ttotal: 2.97s\tremaining: 529ms\n",
      "849:\tlearn: 0.7794288\ttest: 0.4182544\tbest: 0.4245135 (796)\ttotal: 2.98s\tremaining: 526ms\n",
      "850:\tlearn: 0.7798035\ttest: 0.4182544\tbest: 0.4245135 (796)\ttotal: 2.98s\tremaining: 522ms\n",
      "851:\tlearn: 0.7811368\ttest: 0.4186160\tbest: 0.4245135 (796)\ttotal: 2.98s\tremaining: 518ms\n",
      "852:\tlearn: 0.7815107\ttest: 0.4186160\tbest: 0.4245135 (796)\ttotal: 2.99s\tremaining: 515ms\n",
      "853:\tlearn: 0.7822128\ttest: 0.4186160\tbest: 0.4245135 (796)\ttotal: 2.99s\tremaining: 511ms\n",
      "854:\tlearn: 0.7829151\ttest: 0.4164011\tbest: 0.4245135 (796)\ttotal: 2.99s\tremaining: 508ms\n",
      "855:\tlearn: 0.7836622\ttest: 0.4164011\tbest: 0.4245135 (796)\ttotal: 3s\tremaining: 504ms\n",
      "856:\tlearn: 0.7832887\ttest: 0.4164011\tbest: 0.4245135 (796)\ttotal: 3s\tremaining: 501ms\n",
      "857:\tlearn: 0.7840355\ttest: 0.4164011\tbest: 0.4245135 (796)\ttotal: 3s\tremaining: 497ms\n",
      "858:\tlearn: 0.7840796\ttest: 0.4145445\tbest: 0.4245135 (796)\ttotal: 3.01s\tremaining: 494ms\n",
      "859:\tlearn: 0.7842441\ttest: 0.4145445\tbest: 0.4245135 (796)\ttotal: 3.01s\tremaining: 490ms\n",
      "860:\tlearn: 0.7834977\ttest: 0.4145445\tbest: 0.4245135 (796)\ttotal: 3.01s\tremaining: 486ms\n",
      "861:\tlearn: 0.7836177\ttest: 0.4126847\tbest: 0.4245135 (796)\ttotal: 3.02s\tremaining: 483ms\n",
      "862:\tlearn: 0.7832441\ttest: 0.4126847\tbest: 0.4245135 (796)\ttotal: 3.02s\tremaining: 479ms\n",
      "863:\tlearn: 0.7832441\ttest: 0.4130423\tbest: 0.4245135 (796)\ttotal: 3.02s\tremaining: 476ms\n",
      "864:\tlearn: 0.7836177\ttest: 0.4149034\tbest: 0.4245135 (796)\ttotal: 3.02s\tremaining: 472ms\n",
      "865:\tlearn: 0.7843647\ttest: 0.4111780\tbest: 0.4245135 (796)\ttotal: 3.03s\tremaining: 469ms\n",
      "866:\tlearn: 0.7854839\ttest: 0.4093104\tbest: 0.4245135 (796)\ttotal: 3.04s\tremaining: 466ms\n",
      "867:\tlearn: 0.7858566\ttest: 0.4093104\tbest: 0.4245135 (796)\ttotal: 3.04s\tremaining: 462ms\n",
      "868:\tlearn: 0.7862292\ttest: 0.4111780\tbest: 0.4245135 (796)\ttotal: 3.04s\tremaining: 459ms\n",
      "869:\tlearn: 0.7852758\ttest: 0.4134006\tbest: 0.4245135 (796)\ttotal: 3.04s\tremaining: 455ms\n",
      "870:\tlearn: 0.7852758\ttest: 0.4134006\tbest: 0.4245135 (796)\ttotal: 3.05s\tremaining: 452ms\n",
      "871:\tlearn: 0.7871390\ttest: 0.4130423\tbest: 0.4245135 (796)\ttotal: 3.05s\tremaining: 448ms\n",
      "872:\tlearn: 0.7867666\ttest: 0.4130423\tbest: 0.4245135 (796)\ttotal: 3.05s\tremaining: 444ms\n",
      "873:\tlearn: 0.7867666\ttest: 0.4134006\tbest: 0.4245135 (796)\ttotal: 3.06s\tremaining: 441ms\n",
      "874:\tlearn: 0.7861865\ttest: 0.4134006\tbest: 0.4245135 (796)\ttotal: 3.06s\tremaining: 437ms\n",
      "875:\tlearn: 0.7875111\ttest: 0.4115349\tbest: 0.4245135 (796)\ttotal: 3.06s\tremaining: 434ms\n",
      "876:\tlearn: 0.7878832\ttest: 0.4115349\tbest: 0.4245135 (796)\ttotal: 3.07s\tremaining: 430ms\n",
      "877:\tlearn: 0.7871390\ttest: 0.4137594\tbest: 0.4245135 (796)\ttotal: 3.07s\tremaining: 427ms\n",
      "878:\tlearn: 0.7880899\ttest: 0.4152630\tbest: 0.4245135 (796)\ttotal: 3.07s\tremaining: 423ms\n",
      "879:\tlearn: 0.7888330\ttest: 0.4152630\tbest: 0.4245135 (796)\ttotal: 3.08s\tremaining: 420ms\n",
      "880:\tlearn: 0.7884615\ttest: 0.4171221\tbest: 0.4245135 (796)\ttotal: 3.08s\tremaining: 416ms\n",
      "881:\tlearn: 0.7884615\ttest: 0.4171221\tbest: 0.4245135 (796)\ttotal: 3.08s\tremaining: 412ms\n",
      "882:\tlearn: 0.7892044\ttest: 0.4152630\tbest: 0.4245135 (796)\ttotal: 3.09s\tremaining: 409ms\n",
      "883:\tlearn: 0.7897813\ttest: 0.4171221\tbest: 0.4245135 (796)\ttotal: 3.09s\tremaining: 405ms\n",
      "884:\tlearn: 0.7901522\ttest: 0.4171221\tbest: 0.4245135 (796)\ttotal: 3.09s\tremaining: 402ms\n",
      "885:\tlearn: 0.7894103\ttest: 0.4171221\tbest: 0.4245135 (796)\ttotal: 3.1s\tremaining: 398ms\n",
      "886:\tlearn: 0.7899466\ttest: 0.4167613\tbest: 0.4245135 (796)\ttotal: 3.1s\tremaining: 395ms\n",
      "887:\tlearn: 0.7899466\ttest: 0.4167613\tbest: 0.4245135 (796)\ttotal: 3.1s\tremaining: 391ms\n",
      "888:\tlearn: 0.7912243\ttest: 0.4152630\tbest: 0.4245135 (796)\ttotal: 3.1s\tremaining: 388ms\n",
      "889:\tlearn: 0.7908537\ttest: 0.4152630\tbest: 0.4245135 (796)\ttotal: 3.11s\tremaining: 384ms\n",
      "890:\tlearn: 0.7904829\ttest: 0.4145445\tbest: 0.4245135 (796)\ttotal: 3.11s\tremaining: 381ms\n",
      "891:\tlearn: 0.7915948\ttest: 0.4145445\tbest: 0.4245135 (796)\ttotal: 3.12s\tremaining: 377ms\n",
      "892:\tlearn: 0.7919651\ttest: 0.4149034\tbest: 0.4245135 (796)\ttotal: 3.12s\tremaining: 374ms\n",
      "893:\tlearn: 0.7927052\ttest: 0.4149034\tbest: 0.4245135 (796)\ttotal: 3.12s\tremaining: 370ms\n",
      "894:\tlearn: 0.7932408\ttest: 0.4118925\tbest: 0.4245135 (796)\ttotal: 3.13s\tremaining: 367ms\n",
      "895:\tlearn: 0.7932408\ttest: 0.4137594\tbest: 0.4245135 (796)\ttotal: 3.13s\tremaining: 363ms\n",
      "896:\tlearn: 0.7941837\ttest: 0.4118925\tbest: 0.4245135 (796)\ttotal: 3.13s\tremaining: 360ms\n",
      "897:\tlearn: 0.7949220\ttest: 0.4141189\tbest: 0.4245135 (796)\ttotal: 3.13s\tremaining: 356ms\n",
      "898:\tlearn: 0.7945529\ttest: 0.4144790\tbest: 0.4245135 (796)\ttotal: 3.14s\tremaining: 353ms\n",
      "899:\tlearn: 0.7941837\ttest: 0.4126095\tbest: 0.4245135 (796)\ttotal: 3.14s\tremaining: 349ms\n",
      "900:\tlearn: 0.7941837\ttest: 0.4122507\tbest: 0.4245135 (796)\ttotal: 3.15s\tremaining: 346ms\n",
      "901:\tlearn: 0.7949220\ttest: 0.4100223\tbest: 0.4245135 (796)\ttotal: 3.15s\tremaining: 342ms\n",
      "902:\tlearn: 0.7945529\ttest: 0.4156231\tbest: 0.4245135 (796)\ttotal: 3.15s\tremaining: 339ms\n",
      "903:\tlearn: 0.7961944\ttest: 0.4156231\tbest: 0.4245135 (796)\ttotal: 3.15s\tremaining: 335ms\n",
      "904:\tlearn: 0.7965629\ttest: 0.4156231\tbest: 0.4245135 (796)\ttotal: 3.16s\tremaining: 332ms\n",
      "905:\tlearn: 0.7969312\ttest: 0.4156231\tbest: 0.4245135 (796)\ttotal: 3.16s\tremaining: 328ms\n",
      "906:\tlearn: 0.7969670\ttest: 0.4137594\tbest: 0.4245135 (796)\ttotal: 3.16s\tremaining: 324ms\n",
      "907:\tlearn: 0.7977028\ttest: 0.4156231\tbest: 0.4245135 (796)\ttotal: 3.17s\tremaining: 321ms\n",
      "908:\tlearn: 0.7980353\ttest: 0.4137594\tbest: 0.4245135 (796)\ttotal: 3.17s\tremaining: 317ms\n",
      "909:\tlearn: 0.7980353\ttest: 0.4137594\tbest: 0.4245135 (796)\ttotal: 3.17s\tremaining: 314ms\n",
      "910:\tlearn: 0.7980353\ttest: 0.4137594\tbest: 0.4245135 (796)\ttotal: 3.18s\tremaining: 310ms\n",
      "911:\tlearn: 0.7995053\ttest: 0.4137594\tbest: 0.4245135 (796)\ttotal: 3.18s\tremaining: 307ms\n",
      "912:\tlearn: 0.8011727\ttest: 0.4134006\tbest: 0.4245135 (796)\ttotal: 3.18s\tremaining: 303ms\n",
      "913:\tlearn: 0.8015390\ttest: 0.4152630\tbest: 0.4245135 (796)\ttotal: 3.19s\tremaining: 300ms\n",
      "914:\tlearn: 0.8019052\ttest: 0.4141189\tbest: 0.4245135 (796)\ttotal: 3.19s\tremaining: 296ms\n",
      "915:\tlearn: 0.8031697\ttest: 0.4137594\tbest: 0.4245135 (796)\ttotal: 3.19s\tremaining: 293ms\n",
      "916:\tlearn: 0.8031697\ttest: 0.4137594\tbest: 0.4245135 (796)\ttotal: 3.19s\tremaining: 289ms\n",
      "917:\tlearn: 0.8031697\ttest: 0.4137594\tbest: 0.4245135 (796)\ttotal: 3.2s\tremaining: 286ms\n",
      "918:\tlearn: 0.8031697\ttest: 0.4137594\tbest: 0.4245135 (796)\ttotal: 3.2s\tremaining: 282ms\n",
      "919:\tlearn: 0.8035352\ttest: 0.4152630\tbest: 0.4245135 (796)\ttotal: 3.2s\tremaining: 279ms\n",
      "920:\tlearn: 0.8042660\ttest: 0.4152630\tbest: 0.4245135 (796)\ttotal: 3.21s\tremaining: 275ms\n",
      "921:\tlearn: 0.8040677\ttest: 0.4174836\tbest: 0.4245135 (796)\ttotal: 3.21s\tremaining: 272ms\n",
      "922:\tlearn: 0.8053305\ttest: 0.4149034\tbest: 0.4245135 (796)\ttotal: 3.21s\tremaining: 268ms\n",
      "923:\tlearn: 0.8053305\ttest: 0.4149034\tbest: 0.4245135 (796)\ttotal: 3.22s\tremaining: 265ms\n",
      "924:\tlearn: 0.8053305\ttest: 0.4145445\tbest: 0.4245135 (796)\ttotal: 3.22s\tremaining: 261ms\n",
      "925:\tlearn: 0.8054978\ttest: 0.4149034\tbest: 0.4245135 (796)\ttotal: 3.22s\tremaining: 258ms\n",
      "926:\tlearn: 0.8056954\ttest: 0.4171221\tbest: 0.4245135 (796)\ttotal: 3.23s\tremaining: 254ms\n",
      "927:\tlearn: 0.8054978\ttest: 0.4130423\tbest: 0.4245135 (796)\ttotal: 3.23s\tremaining: 251ms\n",
      "928:\tlearn: 0.8053001\ttest: 0.4130423\tbest: 0.4245135 (796)\ttotal: 3.23s\tremaining: 247ms\n",
      "929:\tlearn: 0.8056652\ttest: 0.4152630\tbest: 0.4245135 (796)\ttotal: 3.24s\tremaining: 244ms\n",
      "930:\tlearn: 0.8049655\ttest: 0.4130423\tbest: 0.4245135 (796)\ttotal: 3.24s\tremaining: 240ms\n",
      "931:\tlearn: 0.8062276\ttest: 0.4149034\tbest: 0.4245135 (796)\ttotal: 3.24s\tremaining: 237ms\n",
      "932:\tlearn: 0.8062276\ttest: 0.4149034\tbest: 0.4245135 (796)\ttotal: 3.25s\tremaining: 233ms\n",
      "933:\tlearn: 0.8062276\ttest: 0.4152630\tbest: 0.4245135 (796)\ttotal: 3.25s\tremaining: 230ms\n",
      "934:\tlearn: 0.8065923\ttest: 0.4152630\tbest: 0.4245135 (796)\ttotal: 3.25s\tremaining: 226ms\n",
      "935:\tlearn: 0.8069568\ttest: 0.4149034\tbest: 0.4245135 (796)\ttotal: 3.25s\tremaining: 223ms\n",
      "936:\tlearn: 0.8069568\ttest: 0.4149034\tbest: 0.4245135 (796)\ttotal: 3.26s\tremaining: 219ms\n",
      "937:\tlearn: 0.8073211\ttest: 0.4149034\tbest: 0.4245135 (796)\ttotal: 3.26s\tremaining: 216ms\n",
      "938:\tlearn: 0.8074887\ttest: 0.4167613\tbest: 0.4245135 (796)\ttotal: 3.26s\tremaining: 212ms\n",
      "939:\tlearn: 0.8087488\ttest: 0.4167613\tbest: 0.4245135 (796)\ttotal: 3.27s\tremaining: 209ms\n",
      "940:\tlearn: 0.8092805\ttest: 0.4182544\tbest: 0.4245135 (796)\ttotal: 3.27s\tremaining: 205ms\n",
      "941:\tlearn: 0.8105393\ttest: 0.4201046\tbest: 0.4245135 (796)\ttotal: 3.27s\tremaining: 202ms\n",
      "942:\tlearn: 0.8109026\ttest: 0.4197424\tbest: 0.4245135 (796)\ttotal: 3.28s\tremaining: 198ms\n",
      "943:\tlearn: 0.8114340\ttest: 0.4197424\tbest: 0.4245135 (796)\ttotal: 3.28s\tremaining: 195ms\n",
      "944:\tlearn: 0.8121599\ttest: 0.4215882\tbest: 0.4245135 (796)\ttotal: 3.28s\tremaining: 191ms\n",
      "945:\tlearn: 0.8116023\ttest: 0.4249047\tbest: 0.4249047 (945)\ttotal: 3.29s\tremaining: 188ms\n",
      "946:\tlearn: 0.8106804\ttest: 0.4252701\tbest: 0.4252701 (946)\ttotal: 3.29s\tremaining: 184ms\n",
      "947:\tlearn: 0.8105121\ttest: 0.4252701\tbest: 0.4252701 (946)\ttotal: 3.29s\tremaining: 181ms\n",
      "948:\tlearn: 0.8119654\ttest: 0.4252701\tbest: 0.4252701 (946)\ttotal: 3.29s\tremaining: 177ms\n",
      "949:\tlearn: 0.8112390\ttest: 0.4252701\tbest: 0.4252701 (946)\ttotal: 3.3s\tremaining: 174ms\n",
      "950:\tlearn: 0.8116023\ttest: 0.4256361\tbest: 0.4256361 (950)\ttotal: 3.3s\tremaining: 170ms\n",
      "951:\tlearn: 0.8126911\ttest: 0.4256361\tbest: 0.4256361 (950)\ttotal: 3.3s\tremaining: 167ms\n",
      "952:\tlearn: 0.8139721\ttest: 0.4256361\tbest: 0.4256361 (950)\ttotal: 3.31s\tremaining: 163ms\n",
      "953:\tlearn: 0.8136100\ttest: 0.4256361\tbest: 0.4256361 (950)\ttotal: 3.31s\tremaining: 160ms\n",
      "954:\tlearn: 0.8136100\ttest: 0.4256361\tbest: 0.4256361 (950)\ttotal: 3.31s\tremaining: 156ms\n",
      "955:\tlearn: 0.8143094\ttest: 0.4256361\tbest: 0.4256361 (950)\ttotal: 3.32s\tremaining: 153ms\n",
      "956:\tlearn: 0.8146715\ttest: 0.4219516\tbest: 0.4256361 (950)\ttotal: 3.32s\tremaining: 149ms\n",
      "957:\tlearn: 0.8153952\ttest: 0.4256361\tbest: 0.4256361 (950)\ttotal: 3.32s\tremaining: 146ms\n",
      "958:\tlearn: 0.8157568\ttest: 0.4215882\tbest: 0.4256361 (950)\ttotal: 3.33s\tremaining: 142ms\n",
      "959:\tlearn: 0.8162872\ttest: 0.4197424\tbest: 0.4256361 (950)\ttotal: 3.33s\tremaining: 139ms\n",
      "960:\tlearn: 0.8170098\ttest: 0.4178935\tbest: 0.4256361 (950)\ttotal: 3.33s\tremaining: 135ms\n",
      "961:\tlearn: 0.8173708\ttest: 0.4178935\tbest: 0.4256361 (950)\ttotal: 3.34s\tremaining: 132ms\n",
      "962:\tlearn: 0.8175400\ttest: 0.4171736\tbest: 0.4256361 (950)\ttotal: 3.34s\tremaining: 128ms\n",
      "963:\tlearn: 0.8175400\ttest: 0.4134714\tbest: 0.4256361 (950)\ttotal: 3.34s\tremaining: 125ms\n",
      "964:\tlearn: 0.8171789\ttest: 0.4153241\tbest: 0.4256361 (950)\ttotal: 3.35s\tremaining: 121ms\n",
      "965:\tlearn: 0.8168176\ttest: 0.4153241\tbest: 0.4256361 (950)\ttotal: 3.35s\tremaining: 118ms\n",
      "966:\tlearn: 0.8171789\ttest: 0.4153241\tbest: 0.4256361 (950)\ttotal: 3.35s\tremaining: 114ms\n",
      "967:\tlearn: 0.8182617\ttest: 0.4171736\tbest: 0.4256361 (950)\ttotal: 3.35s\tremaining: 111ms\n",
      "968:\tlearn: 0.8175400\ttest: 0.4212253\tbest: 0.4256361 (950)\ttotal: 3.36s\tremaining: 107ms\n",
      "969:\tlearn: 0.8179009\ttest: 0.4212253\tbest: 0.4256361 (950)\ttotal: 3.36s\tremaining: 104ms\n",
      "970:\tlearn: 0.8196607\ttest: 0.4230666\tbest: 0.4256361 (950)\ttotal: 3.37s\tremaining: 101ms\n",
      "971:\tlearn: 0.8198303\ttest: 0.4230666\tbest: 0.4256361 (950)\ttotal: 3.37s\tremaining: 97ms\n",
      "972:\tlearn: 0.8200212\ttest: 0.4230666\tbest: 0.4256361 (950)\ttotal: 3.37s\tremaining: 93.5ms\n",
      "973:\tlearn: 0.8203606\ttest: 0.4234307\tbest: 0.4256361 (950)\ttotal: 3.37s\tremaining: 90.1ms\n",
      "974:\tlearn: 0.8207210\ttest: 0.4215882\tbest: 0.4256361 (950)\ttotal: 3.38s\tremaining: 86.6ms\n",
      "975:\tlearn: 0.8207210\ttest: 0.4212253\tbest: 0.4256361 (950)\ttotal: 3.38s\tremaining: 83.1ms\n",
      "976:\tlearn: 0.8205513\ttest: 0.4193809\tbest: 0.4256361 (950)\ttotal: 3.38s\tremaining: 79.7ms\n",
      "977:\tlearn: 0.8209115\ttest: 0.4193809\tbest: 0.4256361 (950)\ttotal: 3.39s\tremaining: 76.2ms\n",
      "978:\tlearn: 0.8212716\ttest: 0.4193809\tbest: 0.4256361 (950)\ttotal: 3.39s\tremaining: 72.7ms\n",
      "979:\tlearn: 0.8212716\ttest: 0.4193809\tbest: 0.4256361 (950)\ttotal: 3.39s\tremaining: 69.3ms\n",
      "980:\tlearn: 0.8212716\ttest: 0.4193809\tbest: 0.4256361 (950)\ttotal: 3.4s\tremaining: 65.8ms\n",
      "981:\tlearn: 0.8216316\ttest: 0.4212253\tbest: 0.4256361 (950)\ttotal: 3.4s\tremaining: 62.3ms\n",
      "982:\tlearn: 0.8219914\ttest: 0.4197424\tbest: 0.4256361 (950)\ttotal: 3.4s\tremaining: 58.9ms\n",
      "983:\tlearn: 0.8219914\ttest: 0.4252701\tbest: 0.4256361 (950)\ttotal: 3.41s\tremaining: 55.4ms\n",
      "984:\tlearn: 0.8227106\ttest: 0.4252701\tbest: 0.4256361 (950)\ttotal: 3.41s\tremaining: 51.9ms\n",
      "985:\tlearn: 0.8239584\ttest: 0.4267396\tbest: 0.4267396 (985)\ttotal: 3.41s\tremaining: 48.5ms\n",
      "986:\tlearn: 0.8239584\ttest: 0.4249047\tbest: 0.4267396 (985)\ttotal: 3.42s\tremaining: 45ms\n",
      "987:\tlearn: 0.8253933\ttest: 0.4252701\tbest: 0.4267396 (985)\ttotal: 3.42s\tremaining: 41.5ms\n",
      "988:\tlearn: 0.8253933\ttest: 0.4271063\tbest: 0.4271063 (988)\ttotal: 3.42s\tremaining: 38.1ms\n",
      "989:\tlearn: 0.8257517\ttest: 0.4252701\tbest: 0.4271063 (988)\ttotal: 3.42s\tremaining: 34.6ms\n",
      "990:\tlearn: 0.8257517\ttest: 0.4252701\tbest: 0.4271063 (988)\ttotal: 3.43s\tremaining: 31.1ms\n",
      "991:\tlearn: 0.8257517\ttest: 0.4252701\tbest: 0.4271063 (988)\ttotal: 3.43s\tremaining: 27.7ms\n",
      "992:\tlearn: 0.8257517\ttest: 0.4252701\tbest: 0.4271063 (988)\ttotal: 3.43s\tremaining: 24.2ms\n",
      "993:\tlearn: 0.8262804\ttest: 0.4234307\tbest: 0.4271063 (988)\ttotal: 3.44s\tremaining: 20.8ms\n",
      "994:\tlearn: 0.8262804\ttest: 0.4234307\tbest: 0.4271063 (988)\ttotal: 3.44s\tremaining: 17.3ms\n",
      "995:\tlearn: 0.8262804\ttest: 0.4252701\tbest: 0.4271063 (988)\ttotal: 3.44s\tremaining: 13.8ms\n",
      "996:\tlearn: 0.8262804\ttest: 0.4237954\tbest: 0.4271063 (988)\ttotal: 3.45s\tremaining: 10.4ms\n",
      "997:\tlearn: 0.8262804\ttest: 0.4237954\tbest: 0.4271063 (988)\ttotal: 3.45s\tremaining: 6.92ms\n",
      "998:\tlearn: 0.8266385\ttest: 0.4230666\tbest: 0.4271063 (988)\ttotal: 3.45s\tremaining: 3.46ms\n",
      "999:\tlearn: 0.8266385\ttest: 0.4230666\tbest: 0.4271063 (988)\ttotal: 3.46s\tremaining: 0us\n",
      "\n",
      "bestTest = 0.4271062664\n",
      "bestIteration = 988\n",
      "\n",
      "Shrink model to first 989 iterations.\n",
      "F_1.25 score - 0.42710626644471894\n"
     ]
    }
   ],
   "source": [
    "X_train, X_valid, y_train, y_valid = train_test_split(join_df, data['target'].astype(int), test_size=.2, random_state=42)\n",
    "train_dataset = Pool(X_train, y_train)\n",
    "val_dataset = Pool(X_valid, y_valid)\n",
    "\n",
    "cat_default = CatBoostClassifier(eval_metric='F:beta=1.25')\n",
    "cat_default.fit(train_dataset, use_best_model=True, eval_set=val_dataset)\n",
    "\n",
    "print(f'F_1.25 score - {fbeta_score(y_valid, cat_default.predict(val_dataset), beta=1.25)}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## С параметрами"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0:\tloss: 0.0409672\tbest: 0.0409672 (0)\ttotal: 197ms\tremaining: 47.1s\n",
      "1:\tloss: 0.0686564\tbest: 0.0686564 (1)\ttotal: 351ms\tremaining: 41.8s\n",
      "2:\tloss: 0.1200869\tbest: 0.1200869 (2)\ttotal: 503ms\tremaining: 39.7s\n",
      "3:\tloss: 0.2021626\tbest: 0.2021626 (3)\ttotal: 655ms\tremaining: 38.6s\n",
      "4:\tloss: 0.1621597\tbest: 0.2021626 (3)\ttotal: 1.11s\tremaining: 52.1s\n",
      "5:\tloss: 0.1975298\tbest: 0.2021626 (3)\ttotal: 1.57s\tremaining: 1m 1s\n",
      "6:\tloss: 0.2491560\tbest: 0.2491560 (6)\ttotal: 2.04s\tremaining: 1m 7s\n",
      "7:\tloss: 0.3205169\tbest: 0.3205169 (7)\ttotal: 2.5s\tremaining: 1m 12s\n",
      "8:\tloss: 0.2350274\tbest: 0.3205169 (7)\ttotal: 3.24s\tremaining: 1m 23s\n",
      "9:\tloss: 0.2754645\tbest: 0.3205169 (7)\ttotal: 3.98s\tremaining: 1m 31s\n",
      "10:\tloss: 0.2837948\tbest: 0.3205169 (7)\ttotal: 4.78s\tremaining: 1m 39s\n",
      "11:\tloss: 0.3631392\tbest: 0.3631392 (11)\ttotal: 5.55s\tremaining: 1m 45s\n",
      "12:\tloss: 0.2897765\tbest: 0.3631392 (11)\ttotal: 6.74s\tremaining: 1m 57s\n",
      "13:\tloss: 0.3000976\tbest: 0.3631392 (11)\ttotal: 8.01s\tremaining: 2m 9s\n",
      "14:\tloss: 0.3258106\tbest: 0.3631392 (11)\ttotal: 9.23s\tremaining: 2m 18s\n",
      "15:\tloss: 0.3631890\tbest: 0.3631890 (15)\ttotal: 10.5s\tremaining: 2m 26s\n",
      "16:\tloss: 0.2938111\tbest: 0.3631890 (15)\ttotal: 12.1s\tremaining: 2m 39s\n",
      "17:\tloss: 0.3299076\tbest: 0.3631890 (15)\ttotal: 13.8s\tremaining: 2m 50s\n",
      "18:\tloss: 0.3338868\tbest: 0.3631890 (15)\ttotal: 15.4s\tremaining: 2m 59s\n",
      "19:\tloss: 0.3689024\tbest: 0.3689024 (19)\ttotal: 17.1s\tremaining: 3m 8s\n",
      "20:\tloss: 0.0834950\tbest: 0.3689024 (19)\ttotal: 17.3s\tremaining: 3m\n",
      "21:\tloss: 0.1349699\tbest: 0.3689024 (19)\ttotal: 17.5s\tremaining: 2m 53s\n",
      "22:\tloss: 0.1687089\tbest: 0.3689024 (19)\ttotal: 17.7s\tremaining: 2m 46s\n",
      "23:\tloss: 0.2390974\tbest: 0.3689024 (19)\ttotal: 17.9s\tremaining: 2m 40s\n",
      "24:\tloss: 0.1989238\tbest: 0.3689024 (19)\ttotal: 18.4s\tremaining: 2m 38s\n",
      "25:\tloss: 0.2235855\tbest: 0.3689024 (19)\ttotal: 19s\tremaining: 2m 35s\n",
      "26:\tloss: 0.2841584\tbest: 0.3689024 (19)\ttotal: 19.5s\tremaining: 2m 33s\n",
      "27:\tloss: 0.3492173\tbest: 0.3689024 (19)\ttotal: 20s\tremaining: 2m 31s\n",
      "28:\tloss: 0.2456955\tbest: 0.3689024 (19)\ttotal: 21s\tremaining: 2m 32s\n",
      "29:\tloss: 0.2938111\tbest: 0.3689024 (19)\ttotal: 22.3s\tremaining: 2m 36s\n",
      "30:\tloss: 0.3319048\tbest: 0.3689024 (19)\ttotal: 23.6s\tremaining: 2m 39s\n",
      "31:\tloss: 0.3592720\tbest: 0.3689024 (19)\ttotal: 25.1s\tremaining: 2m 43s\n",
      "32:\tloss: 0.3055736\tbest: 0.3689024 (19)\ttotal: 27.6s\tremaining: 2m 53s\n",
      "33:\tloss: 0.3269231\tbest: 0.3689024 (19)\ttotal: 29.7s\tremaining: 2m 59s\n",
      "34:\tloss: 0.3352804\tbest: 0.3689024 (19)\ttotal: 31.5s\tremaining: 3m 4s\n",
      "35:\tloss: 0.3853383\tbest: 0.3853383 (35)\ttotal: 33.4s\tremaining: 3m 9s\n",
      "36:\tloss: 0.3233748\tbest: 0.3853383 (35)\ttotal: 36.1s\tremaining: 3m 17s\n",
      "37:\tloss: 0.3408173\tbest: 0.3853383 (35)\ttotal: 38.8s\tremaining: 3m 26s\n",
      "38:\tloss: 0.3423006\tbest: 0.3853383 (35)\ttotal: 41.2s\tremaining: 3m 32s\n",
      "39:\tloss: 0.3924151\tbest: 0.3924151 (39)\ttotal: 43.9s\tremaining: 3m 39s\n",
      "40:\tloss: 0.1138889\tbest: 0.3924151 (39)\ttotal: 44.3s\tremaining: 3m 34s\n",
      "41:\tloss: 0.1470751\tbest: 0.3924151 (39)\ttotal: 44.7s\tremaining: 3m 30s\n",
      "42:\tloss: 0.1692391\tbest: 0.3924151 (39)\ttotal: 45.1s\tremaining: 3m 26s\n",
      "43:\tloss: 0.2799800\tbest: 0.3924151 (39)\ttotal: 45.6s\tremaining: 3m 22s\n",
      "44:\tloss: 0.2257398\tbest: 0.3924151 (39)\ttotal: 46.5s\tremaining: 3m 21s\n",
      "45:\tloss: 0.2564209\tbest: 0.3924151 (39)\ttotal: 47.2s\tremaining: 3m 18s\n",
      "46:\tloss: 0.2904723\tbest: 0.3924151 (39)\ttotal: 47.9s\tremaining: 3m 16s\n",
      "47:\tloss: 0.3654134\tbest: 0.3924151 (39)\ttotal: 48.9s\tremaining: 3m 15s\n",
      "48:\tloss: 0.2693779\tbest: 0.3924151 (39)\ttotal: 50.2s\tremaining: 3m 15s\n",
      "49:\tloss: 0.3265773\tbest: 0.3924151 (39)\ttotal: 51.6s\tremaining: 3m 16s\n",
      "50:\tloss: 0.3320872\tbest: 0.3924151 (39)\ttotal: 53.3s\tremaining: 3m 17s\n",
      "51:\tloss: 0.3865141\tbest: 0.3924151 (39)\ttotal: 55s\tremaining: 3m 19s\n",
      "52:\tloss: 0.3217489\tbest: 0.3924151 (39)\ttotal: 57.6s\tremaining: 3m 23s\n",
      "53:\tloss: 0.3523634\tbest: 0.3924151 (39)\ttotal: 59.9s\tremaining: 3m 26s\n",
      "54:\tloss: 0.3460712\tbest: 0.3924151 (39)\ttotal: 1m 2s\tremaining: 3m 30s\n",
      "55:\tloss: 0.3942308\tbest: 0.3942308 (55)\ttotal: 1m 4s\tremaining: 3m 33s\n",
      "56:\tloss: 0.3380523\tbest: 0.3942308 (55)\ttotal: 1m 7s\tremaining: 3m 36s\n",
      "57:\tloss: 0.3600848\tbest: 0.3942308 (55)\ttotal: 1m 10s\tremaining: 3m 40s\n",
      "58:\tloss: 0.3548314\tbest: 0.3942308 (55)\ttotal: 1m 12s\tremaining: 3m 43s\n",
      "59:\tloss: 0.4147075\tbest: 0.4147075 (59)\ttotal: 1m 15s\tremaining: 3m 46s\n",
      "60:\tloss: 0.1490492\tbest: 0.4147075 (59)\ttotal: 1m 15s\tremaining: 3m 42s\n",
      "61:\tloss: 0.1925053\tbest: 0.4147075 (59)\ttotal: 1m 15s\tremaining: 3m 38s\n",
      "62:\tloss: 0.2214286\tbest: 0.4147075 (59)\ttotal: 1m 16s\tremaining: 3m 34s\n",
      "63:\tloss: 0.2815934\tbest: 0.4147075 (59)\ttotal: 1m 16s\tremaining: 3m 30s\n",
      "64:\tloss: 0.2387745\tbest: 0.4147075 (59)\ttotal: 1m 17s\tremaining: 3m 28s\n",
      "65:\tloss: 0.2897173\tbest: 0.4147075 (59)\ttotal: 1m 18s\tremaining: 3m 26s\n",
      "66:\tloss: 0.3262253\tbest: 0.4147075 (59)\ttotal: 1m 19s\tremaining: 3m 25s\n",
      "67:\tloss: 0.3679947\tbest: 0.4147075 (59)\ttotal: 1m 20s\tremaining: 3m 23s\n",
      "68:\tloss: 0.2940845\tbest: 0.4147075 (59)\ttotal: 1m 21s\tremaining: 3m 23s\n",
      "69:\tloss: 0.3209142\tbest: 0.4147075 (59)\ttotal: 1m 23s\tremaining: 3m 23s\n",
      "70:\tloss: 0.3463267\tbest: 0.4147075 (59)\ttotal: 1m 25s\tremaining: 3m 23s\n",
      "71:\tloss: 0.3821144\tbest: 0.4147075 (59)\ttotal: 1m 27s\tremaining: 3m 23s\n",
      "72:\tloss: 0.3288941\tbest: 0.4147075 (59)\ttotal: 1m 32s\tremaining: 3m 31s\n",
      "73:\tloss: 0.3387659\tbest: 0.4147075 (59)\ttotal: 1m 36s\tremaining: 3m 36s\n",
      "74:\tloss: 0.3680106\tbest: 0.4147075 (59)\ttotal: 1m 40s\tremaining: 3m 41s\n",
      "75:\tloss: 0.3983898\tbest: 0.4147075 (59)\ttotal: 1m 43s\tremaining: 3m 42s\n",
      "76:\tloss: 0.3449818\tbest: 0.4147075 (59)\ttotal: 1m 46s\tremaining: 3m 46s\n",
      "77:\tloss: 0.3469892\tbest: 0.4147075 (59)\ttotal: 1m 49s\tremaining: 3m 47s\n",
      "78:\tloss: 0.3820617\tbest: 0.4147075 (59)\ttotal: 1m 52s\tremaining: 3m 49s\n",
      "79:\tloss: 0.4020159\tbest: 0.4147075 (59)\ttotal: 1m 55s\tremaining: 3m 51s\n",
      "80:\tloss: 0.0409672\tbest: 0.4147075 (59)\ttotal: 1m 55s\tremaining: 3m 47s\n",
      "81:\tloss: 0.0686564\tbest: 0.4147075 (59)\ttotal: 1m 55s\tremaining: 3m 43s\n",
      "82:\tloss: 0.1200869\tbest: 0.4147075 (59)\ttotal: 1m 56s\tremaining: 3m 39s\n",
      "83:\tloss: 0.2021626\tbest: 0.4147075 (59)\ttotal: 1m 56s\tremaining: 3m 35s\n",
      "84:\tloss: 0.1621597\tbest: 0.4147075 (59)\ttotal: 1m 56s\tremaining: 3m 32s\n",
      "85:\tloss: 0.1975298\tbest: 0.4147075 (59)\ttotal: 1m 57s\tremaining: 3m 29s\n",
      "86:\tloss: 0.2491560\tbest: 0.4147075 (59)\ttotal: 1m 57s\tremaining: 3m 26s\n",
      "87:\tloss: 0.3205169\tbest: 0.4147075 (59)\ttotal: 1m 58s\tremaining: 3m 24s\n",
      "88:\tloss: 0.2350274\tbest: 0.4147075 (59)\ttotal: 1m 58s\tremaining: 3m 21s\n",
      "89:\tloss: 0.2754645\tbest: 0.4147075 (59)\ttotal: 1m 59s\tremaining: 3m 19s\n",
      "90:\tloss: 0.2837948\tbest: 0.4147075 (59)\ttotal: 2m\tremaining: 3m 17s\n",
      "91:\tloss: 0.3631392\tbest: 0.4147075 (59)\ttotal: 2m 1s\tremaining: 3m 15s\n",
      "92:\tloss: 0.2897765\tbest: 0.4147075 (59)\ttotal: 2m 3s\tremaining: 3m 15s\n",
      "93:\tloss: 0.3000976\tbest: 0.4147075 (59)\ttotal: 2m 5s\tremaining: 3m 15s\n",
      "94:\tloss: 0.3258106\tbest: 0.4147075 (59)\ttotal: 2m 7s\tremaining: 3m 14s\n",
      "95:\tloss: 0.3631890\tbest: 0.4147075 (59)\ttotal: 2m 9s\tremaining: 3m 14s\n",
      "96:\tloss: 0.2938111\tbest: 0.4147075 (59)\ttotal: 2m 11s\tremaining: 3m 13s\n",
      "97:\tloss: 0.3299076\tbest: 0.4147075 (59)\ttotal: 2m 13s\tremaining: 3m 13s\n",
      "98:\tloss: 0.3338868\tbest: 0.4147075 (59)\ttotal: 2m 15s\tremaining: 3m 12s\n",
      "99:\tloss: 0.3689024\tbest: 0.4147075 (59)\ttotal: 2m 17s\tremaining: 3m 12s\n",
      "100:\tloss: 0.0834950\tbest: 0.4147075 (59)\ttotal: 2m 17s\tremaining: 3m 9s\n",
      "101:\tloss: 0.1349699\tbest: 0.4147075 (59)\ttotal: 2m 17s\tremaining: 3m 6s\n",
      "102:\tloss: 0.1687089\tbest: 0.4147075 (59)\ttotal: 2m 17s\tremaining: 3m 3s\n",
      "103:\tloss: 0.2390974\tbest: 0.4147075 (59)\ttotal: 2m 18s\tremaining: 3m\n",
      "104:\tloss: 0.1989238\tbest: 0.4147075 (59)\ttotal: 2m 18s\tremaining: 2m 58s\n",
      "105:\tloss: 0.2235855\tbest: 0.4147075 (59)\ttotal: 2m 19s\tremaining: 2m 56s\n",
      "106:\tloss: 0.2841584\tbest: 0.4147075 (59)\ttotal: 2m 19s\tremaining: 2m 53s\n",
      "107:\tloss: 0.3492173\tbest: 0.4147075 (59)\ttotal: 2m 20s\tremaining: 2m 51s\n",
      "108:\tloss: 0.2456955\tbest: 0.4147075 (59)\ttotal: 2m 21s\tremaining: 2m 50s\n",
      "109:\tloss: 0.2938111\tbest: 0.4147075 (59)\ttotal: 2m 22s\tremaining: 2m 48s\n",
      "110:\tloss: 0.3319048\tbest: 0.4147075 (59)\ttotal: 2m 23s\tremaining: 2m 46s\n",
      "111:\tloss: 0.3592720\tbest: 0.4147075 (59)\ttotal: 2m 24s\tremaining: 2m 45s\n",
      "112:\tloss: 0.3055736\tbest: 0.4147075 (59)\ttotal: 2m 26s\tremaining: 2m 44s\n",
      "113:\tloss: 0.3269231\tbest: 0.4147075 (59)\ttotal: 2m 28s\tremaining: 2m 43s\n",
      "114:\tloss: 0.3352804\tbest: 0.4147075 (59)\ttotal: 2m 29s\tremaining: 2m 42s\n",
      "115:\tloss: 0.3853383\tbest: 0.4147075 (59)\ttotal: 2m 31s\tremaining: 2m 42s\n",
      "116:\tloss: 0.3233748\tbest: 0.4147075 (59)\ttotal: 2m 34s\tremaining: 2m 42s\n",
      "117:\tloss: 0.3408173\tbest: 0.4147075 (59)\ttotal: 2m 36s\tremaining: 2m 41s\n",
      "118:\tloss: 0.3423006\tbest: 0.4147075 (59)\ttotal: 2m 38s\tremaining: 2m 41s\n",
      "119:\tloss: 0.3924151\tbest: 0.4147075 (59)\ttotal: 2m 40s\tremaining: 2m 40s\n",
      "120:\tloss: 0.1138889\tbest: 0.4147075 (59)\ttotal: 2m 40s\tremaining: 2m 38s\n",
      "121:\tloss: 0.1470751\tbest: 0.4147075 (59)\ttotal: 2m 41s\tremaining: 2m 35s\n",
      "122:\tloss: 0.1692391\tbest: 0.4147075 (59)\ttotal: 2m 41s\tremaining: 2m 33s\n",
      "123:\tloss: 0.2799800\tbest: 0.4147075 (59)\ttotal: 2m 41s\tremaining: 2m 31s\n",
      "124:\tloss: 0.2257398\tbest: 0.4147075 (59)\ttotal: 2m 42s\tremaining: 2m 29s\n",
      "125:\tloss: 0.2564209\tbest: 0.4147075 (59)\ttotal: 2m 43s\tremaining: 2m 27s\n",
      "126:\tloss: 0.2904723\tbest: 0.4147075 (59)\ttotal: 2m 43s\tremaining: 2m 25s\n",
      "127:\tloss: 0.3654134\tbest: 0.4147075 (59)\ttotal: 2m 44s\tremaining: 2m 24s\n",
      "128:\tloss: 0.2693779\tbest: 0.4147075 (59)\ttotal: 2m 45s\tremaining: 2m 22s\n",
      "129:\tloss: 0.3265773\tbest: 0.4147075 (59)\ttotal: 2m 47s\tremaining: 2m 21s\n",
      "130:\tloss: 0.3320872\tbest: 0.4147075 (59)\ttotal: 2m 49s\tremaining: 2m 20s\n",
      "131:\tloss: 0.3865141\tbest: 0.4147075 (59)\ttotal: 2m 50s\tremaining: 2m 19s\n",
      "132:\tloss: 0.3217489\tbest: 0.4147075 (59)\ttotal: 2m 52s\tremaining: 2m 18s\n",
      "133:\tloss: 0.3523634\tbest: 0.4147075 (59)\ttotal: 2m 55s\tremaining: 2m 18s\n",
      "134:\tloss: 0.3460712\tbest: 0.4147075 (59)\ttotal: 2m 57s\tremaining: 2m 17s\n",
      "135:\tloss: 0.3942308\tbest: 0.4147075 (59)\ttotal: 2m 59s\tremaining: 2m 17s\n",
      "136:\tloss: 0.3380523\tbest: 0.4147075 (59)\ttotal: 3m 3s\tremaining: 2m 17s\n",
      "137:\tloss: 0.3600848\tbest: 0.4147075 (59)\ttotal: 3m 6s\tremaining: 2m 18s\n",
      "138:\tloss: 0.3548314\tbest: 0.4147075 (59)\ttotal: 3m 9s\tremaining: 2m 17s\n",
      "139:\tloss: 0.4147075\tbest: 0.4147075 (59)\ttotal: 3m 12s\tremaining: 2m 17s\n",
      "140:\tloss: 0.1490492\tbest: 0.4147075 (59)\ttotal: 3m 12s\tremaining: 2m 15s\n",
      "141:\tloss: 0.1925053\tbest: 0.4147075 (59)\ttotal: 3m 12s\tremaining: 2m 13s\n",
      "142:\tloss: 0.2214286\tbest: 0.4147075 (59)\ttotal: 3m 13s\tremaining: 2m 11s\n",
      "143:\tloss: 0.2815934\tbest: 0.4147075 (59)\ttotal: 3m 13s\tremaining: 2m 8s\n",
      "144:\tloss: 0.2387745\tbest: 0.4147075 (59)\ttotal: 3m 14s\tremaining: 2m 7s\n",
      "145:\tloss: 0.2897173\tbest: 0.4147075 (59)\ttotal: 3m 15s\tremaining: 2m 5s\n",
      "146:\tloss: 0.3262253\tbest: 0.4147075 (59)\ttotal: 3m 16s\tremaining: 2m 4s\n",
      "147:\tloss: 0.3679947\tbest: 0.4147075 (59)\ttotal: 3m 17s\tremaining: 2m 2s\n",
      "148:\tloss: 0.2940845\tbest: 0.4147075 (59)\ttotal: 3m 18s\tremaining: 2m 1s\n",
      "149:\tloss: 0.3209142\tbest: 0.4147075 (59)\ttotal: 3m 20s\tremaining: 2m\n",
      "150:\tloss: 0.3463267\tbest: 0.4147075 (59)\ttotal: 3m 21s\tremaining: 1m 59s\n",
      "151:\tloss: 0.3821144\tbest: 0.4147075 (59)\ttotal: 3m 23s\tremaining: 1m 57s\n",
      "152:\tloss: 0.3288941\tbest: 0.4147075 (59)\ttotal: 3m 26s\tremaining: 1m 57s\n",
      "153:\tloss: 0.3387659\tbest: 0.4147075 (59)\ttotal: 3m 28s\tremaining: 1m 56s\n",
      "154:\tloss: 0.3680106\tbest: 0.4147075 (59)\ttotal: 3m 31s\tremaining: 1m 55s\n",
      "155:\tloss: 0.3983898\tbest: 0.4147075 (59)\ttotal: 3m 33s\tremaining: 1m 55s\n",
      "156:\tloss: 0.3449818\tbest: 0.4147075 (59)\ttotal: 3m 36s\tremaining: 1m 54s\n",
      "157:\tloss: 0.3469892\tbest: 0.4147075 (59)\ttotal: 3m 39s\tremaining: 1m 53s\n",
      "158:\tloss: 0.3820617\tbest: 0.4147075 (59)\ttotal: 3m 42s\tremaining: 1m 53s\n",
      "159:\tloss: 0.4020159\tbest: 0.4147075 (59)\ttotal: 3m 45s\tremaining: 1m 52s\n",
      "160:\tloss: 0.0409672\tbest: 0.4147075 (59)\ttotal: 3m 45s\tremaining: 1m 50s\n",
      "161:\tloss: 0.0686564\tbest: 0.4147075 (59)\ttotal: 3m 45s\tremaining: 1m 48s\n",
      "162:\tloss: 0.1200869\tbest: 0.4147075 (59)\ttotal: 3m 45s\tremaining: 1m 46s\n",
      "163:\tloss: 0.2021626\tbest: 0.4147075 (59)\ttotal: 3m 46s\tremaining: 1m 44s\n",
      "164:\tloss: 0.1621597\tbest: 0.4147075 (59)\ttotal: 3m 46s\tremaining: 1m 43s\n",
      "165:\tloss: 0.1975298\tbest: 0.4147075 (59)\ttotal: 3m 47s\tremaining: 1m 41s\n",
      "166:\tloss: 0.2491560\tbest: 0.4147075 (59)\ttotal: 3m 47s\tremaining: 1m 39s\n",
      "167:\tloss: 0.3205169\tbest: 0.4147075 (59)\ttotal: 3m 48s\tremaining: 1m 38s\n",
      "168:\tloss: 0.2350274\tbest: 0.4147075 (59)\ttotal: 3m 49s\tremaining: 1m 36s\n",
      "169:\tloss: 0.2754645\tbest: 0.4147075 (59)\ttotal: 3m 50s\tremaining: 1m 34s\n",
      "170:\tloss: 0.2837948\tbest: 0.4147075 (59)\ttotal: 3m 51s\tremaining: 1m 33s\n",
      "171:\tloss: 0.3631392\tbest: 0.4147075 (59)\ttotal: 3m 52s\tremaining: 1m 31s\n",
      "172:\tloss: 0.2897765\tbest: 0.4147075 (59)\ttotal: 3m 53s\tremaining: 1m 30s\n",
      "173:\tloss: 0.3000976\tbest: 0.4147075 (59)\ttotal: 3m 55s\tremaining: 1m 29s\n",
      "174:\tloss: 0.3258106\tbest: 0.4147075 (59)\ttotal: 3m 56s\tremaining: 1m 27s\n",
      "175:\tloss: 0.3631890\tbest: 0.4147075 (59)\ttotal: 3m 58s\tremaining: 1m 26s\n",
      "176:\tloss: 0.2938111\tbest: 0.4147075 (59)\ttotal: 4m\tremaining: 1m 25s\n",
      "177:\tloss: 0.3299076\tbest: 0.4147075 (59)\ttotal: 4m 2s\tremaining: 1m 24s\n",
      "178:\tloss: 0.3338868\tbest: 0.4147075 (59)\ttotal: 4m 4s\tremaining: 1m 23s\n",
      "179:\tloss: 0.3689024\tbest: 0.4147075 (59)\ttotal: 4m 7s\tremaining: 1m 22s\n",
      "180:\tloss: 0.0834950\tbest: 0.4147075 (59)\ttotal: 4m 7s\tremaining: 1m 20s\n",
      "181:\tloss: 0.1349699\tbest: 0.4147075 (59)\ttotal: 4m 7s\tremaining: 1m 18s\n",
      "182:\tloss: 0.1687089\tbest: 0.4147075 (59)\ttotal: 4m 7s\tremaining: 1m 17s\n",
      "183:\tloss: 0.2390974\tbest: 0.4147075 (59)\ttotal: 4m 7s\tremaining: 1m 15s\n",
      "184:\tloss: 0.1989238\tbest: 0.4147075 (59)\ttotal: 4m 8s\tremaining: 1m 13s\n",
      "185:\tloss: 0.2235855\tbest: 0.4147075 (59)\ttotal: 4m 9s\tremaining: 1m 12s\n",
      "186:\tloss: 0.2841584\tbest: 0.4147075 (59)\ttotal: 4m 10s\tremaining: 1m 10s\n",
      "187:\tloss: 0.3492173\tbest: 0.4147075 (59)\ttotal: 4m 10s\tremaining: 1m 9s\n",
      "188:\tloss: 0.2456955\tbest: 0.4147075 (59)\ttotal: 4m 11s\tremaining: 1m 7s\n",
      "189:\tloss: 0.2938111\tbest: 0.4147075 (59)\ttotal: 4m 13s\tremaining: 1m 6s\n",
      "190:\tloss: 0.3319048\tbest: 0.4147075 (59)\ttotal: 4m 14s\tremaining: 1m 5s\n",
      "191:\tloss: 0.3592720\tbest: 0.4147075 (59)\ttotal: 4m 15s\tremaining: 1m 3s\n",
      "192:\tloss: 0.3055736\tbest: 0.4147075 (59)\ttotal: 4m 17s\tremaining: 1m 2s\n",
      "193:\tloss: 0.3269231\tbest: 0.4147075 (59)\ttotal: 4m 19s\tremaining: 1m 1s\n",
      "194:\tloss: 0.3352804\tbest: 0.4147075 (59)\ttotal: 4m 21s\tremaining: 1m\n",
      "195:\tloss: 0.3853383\tbest: 0.4147075 (59)\ttotal: 4m 23s\tremaining: 59.2s\n",
      "196:\tloss: 0.3233748\tbest: 0.4147075 (59)\ttotal: 4m 26s\tremaining: 58.2s\n",
      "197:\tloss: 0.3408173\tbest: 0.4147075 (59)\ttotal: 4m 28s\tremaining: 57s\n",
      "198:\tloss: 0.3423006\tbest: 0.4147075 (59)\ttotal: 4m 30s\tremaining: 55.8s\n",
      "199:\tloss: 0.3924151\tbest: 0.4147075 (59)\ttotal: 4m 33s\tremaining: 54.7s\n",
      "200:\tloss: 0.1138889\tbest: 0.4147075 (59)\ttotal: 4m 33s\tremaining: 53.1s\n",
      "201:\tloss: 0.1470751\tbest: 0.4147075 (59)\ttotal: 4m 33s\tremaining: 51.5s\n",
      "202:\tloss: 0.1692391\tbest: 0.4147075 (59)\ttotal: 4m 34s\tremaining: 50s\n",
      "203:\tloss: 0.2799800\tbest: 0.4147075 (59)\ttotal: 4m 34s\tremaining: 48.4s\n",
      "204:\tloss: 0.2257398\tbest: 0.4147075 (59)\ttotal: 4m 35s\tremaining: 47s\n",
      "205:\tloss: 0.2564209\tbest: 0.4147075 (59)\ttotal: 4m 36s\tremaining: 45.6s\n",
      "206:\tloss: 0.2904723\tbest: 0.4147075 (59)\ttotal: 4m 36s\tremaining: 44.1s\n",
      "207:\tloss: 0.3654134\tbest: 0.4147075 (59)\ttotal: 4m 37s\tremaining: 42.7s\n",
      "208:\tloss: 0.2693779\tbest: 0.4147075 (59)\ttotal: 4m 38s\tremaining: 41.4s\n",
      "209:\tloss: 0.3265773\tbest: 0.4147075 (59)\ttotal: 4m 40s\tremaining: 40s\n",
      "210:\tloss: 0.3320872\tbest: 0.4147075 (59)\ttotal: 4m 41s\tremaining: 38.7s\n",
      "211:\tloss: 0.3865141\tbest: 0.4147075 (59)\ttotal: 4m 42s\tremaining: 37.3s\n",
      "212:\tloss: 0.3217489\tbest: 0.4147075 (59)\ttotal: 4m 44s\tremaining: 36.1s\n",
      "213:\tloss: 0.3523634\tbest: 0.4147075 (59)\ttotal: 4m 46s\tremaining: 34.8s\n",
      "214:\tloss: 0.3460712\tbest: 0.4147075 (59)\ttotal: 4m 48s\tremaining: 33.6s\n",
      "215:\tloss: 0.3942308\tbest: 0.4147075 (59)\ttotal: 4m 51s\tremaining: 32.4s\n",
      "216:\tloss: 0.3380523\tbest: 0.4147075 (59)\ttotal: 4m 54s\tremaining: 31.2s\n",
      "217:\tloss: 0.3600848\tbest: 0.4147075 (59)\ttotal: 4m 57s\tremaining: 30.1s\n",
      "218:\tloss: 0.3548314\tbest: 0.4147075 (59)\ttotal: 5m\tremaining: 28.8s\n",
      "219:\tloss: 0.4147075\tbest: 0.4147075 (59)\ttotal: 5m 3s\tremaining: 27.6s\n",
      "220:\tloss: 0.1490492\tbest: 0.4147075 (59)\ttotal: 5m 3s\tremaining: 26.1s\n",
      "221:\tloss: 0.1925053\tbest: 0.4147075 (59)\ttotal: 5m 3s\tremaining: 24.6s\n",
      "222:\tloss: 0.2214286\tbest: 0.4147075 (59)\ttotal: 5m 4s\tremaining: 23.2s\n",
      "223:\tloss: 0.2815934\tbest: 0.4147075 (59)\ttotal: 5m 4s\tremaining: 21.7s\n",
      "224:\tloss: 0.2387745\tbest: 0.4147075 (59)\ttotal: 5m 5s\tremaining: 20.3s\n",
      "225:\tloss: 0.2897173\tbest: 0.4147075 (59)\ttotal: 5m 6s\tremaining: 19s\n",
      "226:\tloss: 0.3262253\tbest: 0.4147075 (59)\ttotal: 5m 6s\tremaining: 17.6s\n",
      "227:\tloss: 0.3679947\tbest: 0.4147075 (59)\ttotal: 5m 7s\tremaining: 16.2s\n",
      "228:\tloss: 0.2940845\tbest: 0.4147075 (59)\ttotal: 5m 9s\tremaining: 14.9s\n",
      "229:\tloss: 0.3209142\tbest: 0.4147075 (59)\ttotal: 5m 10s\tremaining: 13.5s\n",
      "230:\tloss: 0.3463267\tbest: 0.4147075 (59)\ttotal: 5m 12s\tremaining: 12.2s\n",
      "231:\tloss: 0.3821144\tbest: 0.4147075 (59)\ttotal: 5m 13s\tremaining: 10.8s\n",
      "232:\tloss: 0.3288941\tbest: 0.4147075 (59)\ttotal: 5m 16s\tremaining: 9.5s\n",
      "233:\tloss: 0.3387659\tbest: 0.4147075 (59)\ttotal: 5m 18s\tremaining: 8.17s\n",
      "234:\tloss: 0.3680106\tbest: 0.4147075 (59)\ttotal: 5m 20s\tremaining: 6.83s\n",
      "235:\tloss: 0.3983898\tbest: 0.4147075 (59)\ttotal: 5m 23s\tremaining: 5.48s\n",
      "236:\tloss: 0.3449818\tbest: 0.4147075 (59)\ttotal: 5m 25s\tremaining: 4.13s\n",
      "237:\tloss: 0.3469892\tbest: 0.4147075 (59)\ttotal: 5m 29s\tremaining: 2.76s\n",
      "238:\tloss: 0.3820617\tbest: 0.4147075 (59)\ttotal: 5m 32s\tremaining: 1.39s\n",
      "239:\tloss: 0.4020159\tbest: 0.4147075 (59)\ttotal: 5m 35s\tremaining: 0us\n",
      "Estimating final quality...\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<catboost.core.CatBoostClassifier at 0x1f567de5310>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "param_grid = {\n",
    "    'depth': [2, 3, 4, 5],\n",
    "    'learning_rate': [0.02, 0.03, 0.04, 0.1],\n",
    "    'iterations' : [100, 300, 500, 800, 1000],\n",
    "    'bagging_temperature' : [1, 1.2, 1.5]\n",
    "    }\n",
    "base_params = {\n",
    "    'logging_level':'Silent',\n",
    "    'loss_function':'Logloss',\n",
    "    'l2_leaf_reg': 3,\n",
    "    'random_strength': 1.2,\n",
    "    'eval_metric': 'F:beta=1.25', \n",
    "    'random_state': 42,\n",
    "}\n",
    "cat_param = CatBoostClassifier(**base_params)\n",
    "search = cat_param.grid_search(param_grid, train_dataset, cv=5)\n",
    "cat_param.fit(train_dataset, use_best_model=True, eval_set=val_dataset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'bagging_temperature': 1, 'depth': 4, 'learning_rate': 0.1, 'iterations': 1000}\n",
      "F_1.25 score - 0.42050153036567683\n"
     ]
    }
   ],
   "source": [
    "print(search['params'])\n",
    "print(f'F_1.25 score - {fbeta_score(y_valid, cat_param.predict(val_dataset), beta=1.25)}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "def get_scores(y_true, y_pred) -> pd.Series:\n",
    "    return pd.Series(np.array([accuracy_score(y_true, y_pred), \n",
    "                     recall_score(y_true, y_pred), \n",
    "                     f1_score(y_true, y_pred),\n",
    "                     precision_score(y_true, y_pred),\n",
    "                     fbeta_score(y_true, y_pred, beta=1.25),\n",
    "                     roc_auc_score(y_true, y_pred)]), \n",
    "                     index=['Accuracy', 'Recall', 'F1', 'Precision', 'F_beta', 'ROC_AUC'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Accuracy</th>\n",
       "      <th>Recall</th>\n",
       "      <th>F1</th>\n",
       "      <th>Precision</th>\n",
       "      <th>F_beta</th>\n",
       "      <th>ROC_AUC</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>catboost_default</th>\n",
       "      <td>0.643661</td>\n",
       "      <td>0.362617</td>\n",
       "      <td>0.449594</td>\n",
       "      <td>0.591463</td>\n",
       "      <td>0.427106</td>\n",
       "      <td>0.597349</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>catboost_grid</th>\n",
       "      <td>0.639160</td>\n",
       "      <td>0.357009</td>\n",
       "      <td>0.442642</td>\n",
       "      <td>0.582317</td>\n",
       "      <td>0.420502</td>\n",
       "      <td>0.592665</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  Accuracy    Recall        F1  Precision    F_beta   ROC_AUC\n",
       "catboost_default  0.643661  0.362617  0.449594   0.591463  0.427106  0.597349\n",
       "catboost_grid     0.639160  0.357009  0.442642   0.582317  0.420502  0.592665"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "catboost_results = pd.DataFrame({\n",
    "    'catboost_default': util.get_scores(y_valid, cat_default.predict(val_dataset)),\n",
    "    'catboost_grid': util.get_scores(y_valid, cat_param.predict(val_dataset))\n",
    "}).T\n",
    "catboost_results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "disp = ConfusionMatrixDisplay(confusion_matrix(y_valid, cat_default.predict(val_dataset)),display_labels=[0, 1])\n",
    "disp.plot()\n",
    "plt.title(\"Confusion Matrix for catboost without parameters\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "disp = ConfusionMatrixDisplay(confusion_matrix(y_valid, cat_param.predict(val_dataset)),display_labels=[0, 1])\n",
    "disp.plot()\n",
    "plt.title(\"Confusion Matrix for catboost with parameters\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "# Cross-Validation 2.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 288 candidates, totalling 1440 fits\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[16], line 69\u001b[0m\n\u001b[0;32m     66\u001b[0m     \u001b[38;5;28mprint\u001b[39m(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mF_beta score on test set: \u001b[39m\u001b[38;5;124m\"\u001b[39m, test_score)\n\u001b[0;32m     68\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m best_model, util\u001b[38;5;241m.\u001b[39mget_scores(y_test, y_pred)\n\u001b[1;32m---> 69\u001b[0m model, scores \u001b[38;5;241m=\u001b[39m \u001b[43moptimize_catboost\u001b[49m\u001b[43m(\u001b[49m\u001b[43mnew_features\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mdrop\u001b[49m\u001b[43m(\u001b[49m\u001b[43mcolumns\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m[\u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mtarget\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mmodel_response_length\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mmodel_response_unique\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m]\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mtfidf_matrix\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdata\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mtarget\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m]\u001b[49m\u001b[43m)\u001b[49m\n",
      "Cell \u001b[1;32mIn[16], line 59\u001b[0m, in \u001b[0;36moptimize_catboost\u001b[1;34m(df_features, tfidf_matrix, target, beta)\u001b[0m\n\u001b[0;32m     56\u001b[0m grid_search \u001b[38;5;241m=\u001b[39m GridSearchCV(pipeline, param_grid, cv\u001b[38;5;241m=\u001b[39mcv, scoring\u001b[38;5;241m=\u001b[39mscorer, n_jobs\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m-\u001b[39m\u001b[38;5;241m1\u001b[39m, verbose\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m10\u001b[39m)\n\u001b[0;32m     58\u001b[0m \u001b[38;5;66;03m# Обучаем GridSearchCV\u001b[39;00m\n\u001b[1;32m---> 59\u001b[0m \u001b[43mgrid_search\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfit\u001b[49m\u001b[43m(\u001b[49m\u001b[43mX_train\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43my_train\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m     61\u001b[0m \u001b[38;5;66;03m# Проверяем лучшую модель на тестовой выборке\u001b[39;00m\n\u001b[0;32m     62\u001b[0m best_model \u001b[38;5;241m=\u001b[39m grid_search\u001b[38;5;241m.\u001b[39mbest_estimator_\n",
      "File \u001b[1;32mc:\\Users\\light\\anaconda3\\lib\\site-packages\\sklearn\\base.py:1151\u001b[0m, in \u001b[0;36m_fit_context.<locals>.decorator.<locals>.wrapper\u001b[1;34m(estimator, *args, **kwargs)\u001b[0m\n\u001b[0;32m   1144\u001b[0m     estimator\u001b[38;5;241m.\u001b[39m_validate_params()\n\u001b[0;32m   1146\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m config_context(\n\u001b[0;32m   1147\u001b[0m     skip_parameter_validation\u001b[38;5;241m=\u001b[39m(\n\u001b[0;32m   1148\u001b[0m         prefer_skip_nested_validation \u001b[38;5;129;01mor\u001b[39;00m global_skip_validation\n\u001b[0;32m   1149\u001b[0m     )\n\u001b[0;32m   1150\u001b[0m ):\n\u001b[1;32m-> 1151\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m fit_method(estimator, \u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n",
      "File \u001b[1;32mc:\\Users\\light\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_search.py:898\u001b[0m, in \u001b[0;36mBaseSearchCV.fit\u001b[1;34m(self, X, y, groups, **fit_params)\u001b[0m\n\u001b[0;32m    892\u001b[0m     results \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_format_results(\n\u001b[0;32m    893\u001b[0m         all_candidate_params, n_splits, all_out, all_more_results\n\u001b[0;32m    894\u001b[0m     )\n\u001b[0;32m    896\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m results\n\u001b[1;32m--> 898\u001b[0m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_run_search\u001b[49m\u001b[43m(\u001b[49m\u001b[43mevaluate_candidates\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    900\u001b[0m \u001b[38;5;66;03m# multimetric is determined here because in the case of a callable\u001b[39;00m\n\u001b[0;32m    901\u001b[0m \u001b[38;5;66;03m# self.scoring the return type is only known after calling\u001b[39;00m\n\u001b[0;32m    902\u001b[0m first_test_score \u001b[38;5;241m=\u001b[39m all_out[\u001b[38;5;241m0\u001b[39m][\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mtest_scores\u001b[39m\u001b[38;5;124m\"\u001b[39m]\n",
      "File \u001b[1;32mc:\\Users\\light\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_search.py:1419\u001b[0m, in \u001b[0;36mGridSearchCV._run_search\u001b[1;34m(self, evaluate_candidates)\u001b[0m\n\u001b[0;32m   1417\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m_run_search\u001b[39m(\u001b[38;5;28mself\u001b[39m, evaluate_candidates):\n\u001b[0;32m   1418\u001b[0m \u001b[38;5;250m    \u001b[39m\u001b[38;5;124;03m\"\"\"Search all candidates in param_grid\"\"\"\u001b[39;00m\n\u001b[1;32m-> 1419\u001b[0m     \u001b[43mevaluate_candidates\u001b[49m\u001b[43m(\u001b[49m\u001b[43mParameterGrid\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mparam_grid\u001b[49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[1;32mc:\\Users\\light\\anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_search.py:845\u001b[0m, in \u001b[0;36mBaseSearchCV.fit.<locals>.evaluate_candidates\u001b[1;34m(candidate_params, cv, more_results)\u001b[0m\n\u001b[0;32m    837\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mverbose \u001b[38;5;241m>\u001b[39m \u001b[38;5;241m0\u001b[39m:\n\u001b[0;32m    838\u001b[0m     \u001b[38;5;28mprint\u001b[39m(\n\u001b[0;32m    839\u001b[0m         \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mFitting \u001b[39m\u001b[38;5;132;01m{0}\u001b[39;00m\u001b[38;5;124m folds for each of \u001b[39m\u001b[38;5;132;01m{1}\u001b[39;00m\u001b[38;5;124m candidates,\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m    840\u001b[0m         \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m totalling \u001b[39m\u001b[38;5;132;01m{2}\u001b[39;00m\u001b[38;5;124m fits\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;241m.\u001b[39mformat(\n\u001b[0;32m    841\u001b[0m             n_splits, n_candidates, n_candidates \u001b[38;5;241m*\u001b[39m n_splits\n\u001b[0;32m    842\u001b[0m         )\n\u001b[0;32m    843\u001b[0m     )\n\u001b[1;32m--> 845\u001b[0m out \u001b[38;5;241m=\u001b[39m \u001b[43mparallel\u001b[49m\u001b[43m(\u001b[49m\n\u001b[0;32m    846\u001b[0m \u001b[43m    \u001b[49m\u001b[43mdelayed\u001b[49m\u001b[43m(\u001b[49m\u001b[43m_fit_and_score\u001b[49m\u001b[43m)\u001b[49m\u001b[43m(\u001b[49m\n\u001b[0;32m    847\u001b[0m \u001b[43m        \u001b[49m\u001b[43mclone\u001b[49m\u001b[43m(\u001b[49m\u001b[43mbase_estimator\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m    848\u001b[0m \u001b[43m        \u001b[49m\u001b[43mX\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m    849\u001b[0m \u001b[43m        \u001b[49m\u001b[43my\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m    850\u001b[0m \u001b[43m        \u001b[49m\u001b[43mtrain\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtrain\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m    851\u001b[0m \u001b[43m        \u001b[49m\u001b[43mtest\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtest\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m    852\u001b[0m \u001b[43m        \u001b[49m\u001b[43mparameters\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mparameters\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m    853\u001b[0m \u001b[43m        \u001b[49m\u001b[43msplit_progress\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43msplit_idx\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mn_splits\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m    854\u001b[0m \u001b[43m        \u001b[49m\u001b[43mcandidate_progress\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43mcand_idx\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mn_candidates\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m    855\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mfit_and_score_kwargs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m    856\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    857\u001b[0m \u001b[43m    \u001b[49m\u001b[38;5;28;43;01mfor\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43m(\u001b[49m\u001b[43mcand_idx\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mparameters\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43m(\u001b[49m\u001b[43msplit_idx\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43m(\u001b[49m\u001b[43mtrain\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mtest\u001b[49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;129;43;01min\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43mproduct\u001b[49m\u001b[43m(\u001b[49m\n\u001b[0;32m    858\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;28;43menumerate\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43mcandidate_params\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43menumerate\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43mcv\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43msplit\u001b[49m\u001b[43m(\u001b[49m\u001b[43mX\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43my\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mgroups\u001b[49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    859\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    860\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    862\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mlen\u001b[39m(out) \u001b[38;5;241m<\u001b[39m \u001b[38;5;241m1\u001b[39m:\n\u001b[0;32m    863\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\n\u001b[0;32m    864\u001b[0m         \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mNo fits were performed. \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m    865\u001b[0m         \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mWas the CV iterator empty? \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m    866\u001b[0m         \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mWere there no candidates?\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m    867\u001b[0m     )\n",
      "File \u001b[1;32mc:\\Users\\light\\anaconda3\\lib\\site-packages\\sklearn\\utils\\parallel.py:65\u001b[0m, in \u001b[0;36mParallel.__call__\u001b[1;34m(self, iterable)\u001b[0m\n\u001b[0;32m     60\u001b[0m config \u001b[38;5;241m=\u001b[39m get_config()\n\u001b[0;32m     61\u001b[0m iterable_with_config \u001b[38;5;241m=\u001b[39m (\n\u001b[0;32m     62\u001b[0m     (_with_config(delayed_func, config), args, kwargs)\n\u001b[0;32m     63\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m delayed_func, args, kwargs \u001b[38;5;129;01min\u001b[39;00m iterable\n\u001b[0;32m     64\u001b[0m )\n\u001b[1;32m---> 65\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43msuper\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[38;5;21;43m__call__\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43miterable_with_config\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[1;32mc:\\Users\\light\\anaconda3\\lib\\site-packages\\joblib\\parallel.py:1098\u001b[0m, in \u001b[0;36mParallel.__call__\u001b[1;34m(self, iterable)\u001b[0m\n\u001b[0;32m   1095\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_iterating \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mFalse\u001b[39;00m\n\u001b[0;32m   1097\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backend\u001b[38;5;241m.\u001b[39mretrieval_context():\n\u001b[1;32m-> 1098\u001b[0m     \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mretrieve\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m   1099\u001b[0m \u001b[38;5;66;03m# Make sure that we get a last message telling us we are done\u001b[39;00m\n\u001b[0;32m   1100\u001b[0m elapsed_time \u001b[38;5;241m=\u001b[39m time\u001b[38;5;241m.\u001b[39mtime() \u001b[38;5;241m-\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_start_time\n",
      "File \u001b[1;32mc:\\Users\\light\\anaconda3\\lib\\site-packages\\joblib\\parallel.py:975\u001b[0m, in \u001b[0;36mParallel.retrieve\u001b[1;34m(self)\u001b[0m\n\u001b[0;32m    973\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m    974\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mgetattr\u001b[39m(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backend, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124msupports_timeout\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;28;01mFalse\u001b[39;00m):\n\u001b[1;32m--> 975\u001b[0m         \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_output\u001b[38;5;241m.\u001b[39mextend(\u001b[43mjob\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtimeout\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mtimeout\u001b[49m\u001b[43m)\u001b[49m)\n\u001b[0;32m    976\u001b[0m     \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m    977\u001b[0m         \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_output\u001b[38;5;241m.\u001b[39mextend(job\u001b[38;5;241m.\u001b[39mget())\n",
      "File \u001b[1;32mc:\\Users\\light\\anaconda3\\lib\\site-packages\\joblib\\_parallel_backends.py:567\u001b[0m, in \u001b[0;36mLokyBackend.wrap_future_result\u001b[1;34m(future, timeout)\u001b[0m\n\u001b[0;32m    564\u001b[0m \u001b[38;5;250m\u001b[39m\u001b[38;5;124;03m\"\"\"Wrapper for Future.result to implement the same behaviour as\u001b[39;00m\n\u001b[0;32m    565\u001b[0m \u001b[38;5;124;03mAsyncResults.get from multiprocessing.\"\"\"\u001b[39;00m\n\u001b[0;32m    566\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m--> 567\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mfuture\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mresult\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtimeout\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtimeout\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    568\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m CfTimeoutError \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[0;32m    569\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mTimeoutError\u001b[39;00m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01me\u001b[39;00m\n",
      "File \u001b[1;32mc:\\Users\\light\\anaconda3\\lib\\concurrent\\futures\\_base.py:441\u001b[0m, in \u001b[0;36mFuture.result\u001b[1;34m(self, timeout)\u001b[0m\n\u001b[0;32m    438\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_state \u001b[38;5;241m==\u001b[39m FINISHED:\n\u001b[0;32m    439\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m__get_result()\n\u001b[1;32m--> 441\u001b[0m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_condition\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mwait\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtimeout\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    443\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_state \u001b[38;5;129;01min\u001b[39;00m [CANCELLED, CANCELLED_AND_NOTIFIED]:\n\u001b[0;32m    444\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m CancelledError()\n",
      "File \u001b[1;32mc:\\Users\\light\\anaconda3\\lib\\threading.py:312\u001b[0m, in \u001b[0;36mCondition.wait\u001b[1;34m(self, timeout)\u001b[0m\n\u001b[0;32m    310\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:    \u001b[38;5;66;03m# restore state no matter what (e.g., KeyboardInterrupt)\u001b[39;00m\n\u001b[0;32m    311\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m timeout \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m--> 312\u001b[0m         \u001b[43mwaiter\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43macquire\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    313\u001b[0m         gotit \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mTrue\u001b[39;00m\n\u001b[0;32m    314\u001b[0m     \u001b[38;5;28;01melse\u001b[39;00m:\n",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "def optimize_catboost(df_features, tfidf_matrix, target, beta=1.25):\n",
    "    # Преобразуем tfidf_matrix в DataFrame\n",
    "    tfidf_df = pd.DataFrame.sparse.from_spmatrix(tfidf_matrix)\n",
    "    tfidf_df.columns = tfidf_df.columns.astype(str)\n",
    "    \n",
    "    # Объединяем df_features и tfidf_df\n",
    "    combined_features = pd.concat([df_features.reset_index(drop=True), tfidf_df.reset_index(drop=True)], axis=1)\n",
    "\n",
    "    \n",
    "    # Разделяем данные на обучающую и тестовую выборки\n",
    "    X_train, X_test, y_train, y_test = train_test_split(combined_features, \n",
    "                                                        target, \n",
    "                                                        test_size=0.2, \n",
    "                                                        random_state=42, \n",
    "                                                        stratify=target)\n",
    "    # Стратифицированное разделение данных на обучающую и тестовую выборки гарантирует, \n",
    "    # что в каждой из выборок будет представлено одинаковое соотношение классов.\n",
    "    \n",
    "    \n",
    "    # Определяем параметры для GridSearchCV\n",
    "    param_grid = {\n",
    "        'preprocessor__tfidf__n_components': [3, 50, 100],\n",
    "        'catboost__iterations': [100, 200, 400, 800],\n",
    "        'catboost__learning_rate': [0.03, 0.1],\n",
    "        'catboost__depth': [3, 4, 6, 8],\n",
    "        'catboost__bagging_temperature' : [1, 1.2, 1.5]\n",
    "    }\n",
    "    base_params = {\n",
    "        'logging_level':'Info',\n",
    "        'loss_function':'Logloss',\n",
    "        'l2_leaf_reg': 3,\n",
    "        'random_strength': 1.2,\n",
    "        'eval_metric': f'F:beta={beta}',\n",
    "        'random_state': 42,\n",
    "    }\n",
    "    \n",
    "    # Создаем пайплайн\n",
    "    pipeline = Pipeline([\n",
    "        ('preprocessor', ColumnTransformer(\n",
    "        transformers=[\n",
    "            ('tfidf', TruncatedSVD(random_state=42), tfidf_df.columns), #svm \n",
    "            ('num', StandardScaler(), df_features.columns)\n",
    "        ]\n",
    "        )),\n",
    "        ('catboost', CatBoostClassifier(**base_params))  # CatBoostClassifier\n",
    "    ])\n",
    "\n",
    "    # Определяем метрику f_beta\n",
    "    scorer = make_scorer(fbeta_score, beta=beta)\n",
    "\n",
    "    # Создаем кросс-валидатор\n",
    "    cv = StratifiedKFold(n_splits=5, shuffle=True, random_state=42)\n",
    "    \n",
    "    # Создаем GridSearchCV\n",
    "    grid_search = GridSearchCV(pipeline, param_grid, cv=cv, scoring=scorer, n_jobs=-1, verbose=10)\n",
    "\n",
    "    # Обучаем GridSearchCV\n",
    "    grid_search.fit(X_train, y_train)\n",
    "\n",
    "    # Проверяем лучшую модель на тестовой выборке\n",
    "    best_model = grid_search.best_estimator_\n",
    "    y_pred = best_model.predict(X_test)\n",
    "    test_score = fbeta_score(y_test, y_pred, beta=beta)\n",
    "    \n",
    "    print(\"F_beta score on test set: \", test_score)\n",
    "\n",
    "    return best_model, util.get_scores(y_test, y_pred)\n",
    "model, scores = optimize_catboost(new_features.drop(columns=['target', 'model_response_length', 'model_response_unique']), tfidf_matrix, data['target'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "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": "3.9.17"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}