Skip to content

Commit

Permalink
Update the text in tutorials for generic timm and torochvision model (#…
Browse files Browse the repository at this point in the history 
…1200)
  • Loading branch information
@Idan-BenAmi
Idan-BenAmi authored Sep 4, 2024
1 parent 3a3cc2c commit a514d5f
Show file tree
Hide file tree
Showing 3 changed files with 20 additions and 21 deletions.
31 changes: 13 additions & 18 deletions tutorials/notebooks/imx500_notebooks/README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -25,7 +25,7 @@ deployment performance.
<tr>
<td rowspan="9">Classification</td>
<td>MobilenetV2</td>
<td> <a href="keras/example_keras_mobilenetv2_for_imx500.ipynb">Keras</a></td>
<td> <a href="keras/example_keras_mobilenetv2_for_imx500.ipynb">ipynb (Keras)</a></td>
<td><a href="https://keras.io/api/applications/mobilenet/">Keras Applications</a></td>
<td></td>
<td>ImageNet</td>
Expand All @@ -34,7 +34,7 @@ deployment performance.
</tr>
<tr>
<td>MobileVit</td>
<td> <a href="pytorch/pytorch_mobilevit_xs_for_imx500.ipynb">PyTorch</a></td>
<td> <a href="pytorch/pytorch_mobilevit_xs_for_imx500.ipynb">ipynb (PyTorch)</a></td>
<td><a href="https://github.com/huggingface/pytorch-image-models">Timm</a></td>
<td><a href="https://huggingface.co/SSI-DNN/pytorch_mobilevit_xs">mct-model-garden</a></td>
<td>ImageNet</td>
Expand All @@ -43,7 +43,7 @@ deployment performance.
</tr>
<tr>
<td>regnety_002.pycls_in1k</td>
<td> <a href="pytorch/pytorch_timm_classification_model_for_imx500.ipynb">PyTorch</a></td>
<td rowspan="3"> <a href="pytorch/pytorch_timm_classification_model_for_imx500.ipynb">ipynb (PyTorch)</a></td>
<td><a href="https://github.com/huggingface/pytorch-image-models">Timm</a></td>
<td></td>
<td>ImageNet</td>
Expand All @@ -52,7 +52,6 @@ deployment performance.
</tr>
<tr>
<td>regnetx_002.pycls_in1k</td>
<td> <a href="pytorch/pytorch_timm_classification_model_for_imx500.ipynb">PyTorch</a></td>
<td><a href="https://github.com/huggingface/pytorch-image-models">Timm</a></td>
<td></td>
<td>ImageNet</td>
Expand All @@ -61,7 +60,6 @@ deployment performance.
</tr>
<tr>
<td>regnety_004.pycls_in1k</td>
<td> <a href="pytorch/pytorch_timm_classification_model_for_imx500.ipynb">PyTorch</a></td>
<td><a href="https://github.com/huggingface/pytorch-image-models">Timm</a></td>
<td></td>
<td>ImageNet</td>
Expand All @@ -70,7 +68,7 @@ deployment performance.
</tr>
<tr>
<td>mnasnet1_0</td>
<td> <a href="pytorch/pytorch_torchvision_classification_model_for_imx500.ipynb">PyTorch</a></td>
<td rowspan="4"> <a href="pytorch/pytorch_torchvision_classification_model_for_imx500.ipynb">ipynb (PyTorch)</a></td>
<td><a href="https://pytorch.org/vision/stable/models/generated/torchvision.models.mnasnet1_0.html#torchvision.models.MNASNet1_0_Weights">torchvision</a></td>
<td></td>
<td>ImageNet</td>
Expand All @@ -79,7 +77,6 @@ deployment performance.
</tr>
<tr>
<td>mobilenet_v2</td>
<td> <a href="pytorch/pytorch_torchvision_classification_model_for_imx500.ipynb">PyTorch</a></td>
<td><a href="https://pytorch.org/vision/stable/models/generated/torchvision.models.mobilenet_v2.html#torchvision.models.MobileNet_V2_Weights">torchvision</a></td>
<td></td>
<td>ImageNet</td>
Expand All @@ -88,7 +85,6 @@ deployment performance.
</tr>
<tr>
<td>regnet_y_400mf</td>
<td> <a href="pytorch/pytorch_torchvision_classification_model_for_imx500.ipynb">PyTorch</a></td>
<td><a href="https://pytorch.org/vision/stable/models/generated/torchvision.models.regnet_y_400mf.html#torchvision.models.RegNet_Y_400MF_Weights">torchvision</a></td>
<td></td>
<td>ImageNet</td>
Expand All @@ -97,7 +93,6 @@ deployment performance.
</tr>
<tr>
<td>shufflenet_v2_x1_5</td>
<td> <a href="pytorch/pytorch_torchvision_classification_model_for_imx500.ipynb">PyTorch</a></td>
<td><a href="https://pytorch.org/vision/stable/models/generated/torchvision.models.shufflenet_v2_x1_5.html#torchvision.models.ShuffleNet_V2_X1_5_Weights">torchvision</a></td>
<td></td>
<td>ImageNet</td>
Expand All @@ -108,7 +103,7 @@ deployment performance.
<tr>
<td rowspan="4">Object Detection</td>
<td>YOLOv8n</td>
<td> <a href="keras/keras_yolov8n_for_imx500.ipynb">Keras</a></td>
<td> <a href="keras/keras_yolov8n_for_imx500.ipynb">ipynb (Keras)</a></td>
<td><a href="https://github.com/ultralytics">Ultralytics</a></td>
<td><a href="https://huggingface.co/SSI-DNN/keras_yolov8n_640x640_pp">mct-model-garden</a></td>
<td>COCO</td>
Expand All @@ -117,7 +112,7 @@ deployment performance.
</tr>
<tr>
<td>YOLOv8n</td>
<td> <a href="pytorch/pytorch_yolov8n_for_imx500.ipynb">PyTorch</a></td>
<td> <a href="pytorch/pytorch_yolov8n_for_imx500.ipynb">ipynb (PyTorch)</a></td>
<td><a href="https://github.com/ultralytics">Ultralytics</a></td>
<td><a href="https://huggingface.co/SSI-DNN/pytorch_yolov8n_640x640_bb_decoding">mct-model-garden</a></td>
<td>COCO</td>
Expand All @@ -126,7 +121,7 @@ deployment performance.
</tr>
<tr>
<td>NanoDet-Plus-m-416</td>
<td> <a href="keras/example_keras_nanodet_plus_for_imx500.ipynb">Keras</a></td>
<td> <a href="keras/example_keras_nanodet_plus_for_imx500.ipynb">ipynb (Keras)</a></td>
<td><a href="https://github.com/RangiLyu/nanodet">Nanodet</a></td>
<td><a href="https://huggingface.co/SSI-DNN/keras_nanodet_plus_x1.5_416x416">mct-model-garden</a></td>
<td>COCO</td>
Expand All @@ -135,7 +130,7 @@ deployment performance.
</tr>
<tr>
<td>EfficientDet-lite0</td>
<td> <a href="keras/example_keras_effdet_lite0_for_imx500.ipynb">Keras</a></td>
<td> <a href="keras/example_keras_effdet_lite0_for_imx500.ipynb">ipynb (Keras)</a></td>
<td> <a href="https://github.com/rwightman/efficientdet-pytorch">efficientdet-pytorch</a></td>
<td><a href="https://github.com/sony/model_optimization/blob/main/tutorials/mct_model_garden/models_keras/efficientdet/effdet_keras.py">mct-model-garden</a></td>
<td>COCO</td>
Expand All @@ -145,7 +140,7 @@ deployment performance.
<tr>
<td>Semantic Segmentation</td>
<td>Deeplabv3plus</td>
<td> <a href="keras/keras_deeplabv3plus_for_imx500.ipynb">Keras</a></td>
<td> <a href="keras/keras_deeplabv3plus_for_imx500.ipynb">ipynb (Keras)</a></td>
<td> <a href="https://github.com/bonlime/keras-deeplab-v3-plus">bonlime</a></td>
<td><a href="https://huggingface.co/SSI-DNN/keras_deeplabv3_plus_320">mct-model-garden</a></td>
<td>PASCAL VOC</td>
Expand All @@ -155,7 +150,7 @@ deployment performance.
<tr>
<td >Instance Segmentation</td>
<td>YOLOv8n-seg</td>
<td> <a href="pytorch/pytorch_yolov8n_seg_for_imx500.ipynb">PyTorch</a></td>
<td> <a href="pytorch/pytorch_yolov8n_seg_for_imx500.ipynb">ipynb (PyTorch)</a></td>
<td><a href="https://github.com/ultralytics">Ultralytics</a></td>
<td><a href="https://huggingface.co/SSI-DNN/pytorch_yolov8n_inst_seg_640x640">mct-model-garden</a></td>
<td>COCO</td>
Expand All @@ -165,7 +160,7 @@ deployment performance.
<tr>
<td>Pose Estimation</td>
<td>YOLOv8n-pose</td>
<td> <a href="pytorch/pytorch_yolov8n_pose_for_imx500.ipynb">PyTorch</a></td>
<td> <a href="pytorch/pytorch_yolov8n_pose_for_imx500.ipynb">ipynb (PyTorch)</a></td>
<td><a href="https://github.com/ultralytics">Ultralytics</a></td>
<td><a href="https://huggingface.co/SSI-DNN/pytorch_yolov8n_640x640">mct-model-garden</a></td>
<td>COCO</td>
Expand All @@ -175,8 +170,8 @@ deployment performance.
<tr>
<td>Anomaly Detection</td>
<td>Efficient AD</td>
<td> <a href="pytorch/pytorch_efficient_anomaly_detection.ipynb">PyTorch</a></td>
<td><a href="https://arxiv.org/pdf/2303.08730v3">Ultralytics</a></td>
<td> <a href="pytorch/pytorch_efficient_anomaly_detection.ipynb">ipynb (PyTorch)</a></td>
<td><a href="https://arxiv.org/pdf/2303.08730v3">*EfficientAD paper</a></td>
<td><a href="https://huggingface.co/SSI-DNN/Efficient_Anomaly_Detection">mct-model-garden</a></td>
<td>MvTech</td>
<td>98.56</td>
Expand Down
3 changes: 2 additions & 1 deletion ...als/notebooks/imx500_notebooks/pytorch/pytorch_timm_classification_model_for_imx500.ipynb
View file
Original file line number Diff line number Diff line change
Expand Up @@ -111,7 +111,8 @@
"source": [
"## Model Quantization\n",
"\n",
"### Download a Pre-Trained Model \n"
"### Download a Pre-Trained Model - Please select a Timm model\n",
"The tutorial is pre-configured to download `mobilenet_v2` model. In case you wish to use a different model - please change the model & weights below, based on [Timm](https://github.com/huggingface/pytorch-image-models)"
]
},
{
Expand Down
7 changes: 5 additions & 2 deletions ...ebooks/imx500_notebooks/pytorch/pytorch_torchvision_classification_model_for_imx500.ipynb
View file
Original file line number Diff line number Diff line change
Expand Up @@ -39,7 +39,9 @@
{
"metadata": {},
"cell_type": "markdown",
"source": "Install MCT (if it’s not already installed). Additionally, in order to use all the necessary utility functions for this tutorial, we also copy [MCT tutorials folder](https://github.com/sony/model_optimization/tree/main/tutorials) and add it to the system path.",
"source": [
"Install MCT (if it’s not already installed). Additionally, in order to use all the necessary utility functions for this tutorial, we also copy [MCT tutorials folder](https://github.com/sony/model_optimization/tree/main/tutorials) and add it to the system path."
],
"id": "b1a05efedd4dbc77"
},
{
Expand Down Expand Up @@ -93,7 +95,8 @@
"source": [
"## Model Quantization\n",
"\n",
"### Download a Pre-Trained Model"
"### Download a pre-trained model - Please select a Torchvision model\n",
"The tutorial is pre-configured to download `mobilenet_v2` model. In case you wish to use a different model - please change the model & weights below, based on [torchvision](https://pytorch.org/vision/stable/models.html)"
],
"id": "7059e58ac6efff74"
},
Expand Down

0 comments on commit a514d5f

Please sign in to comment.