remove imx500 tutorials

sony · Oct 21, 2024 · 28a9cb5 · 28a9cb5
1 parent 7b4a2ad
commit 28a9cb5
Show file tree

Hide file tree

Showing 86 changed files with 115 additions and 17,195 deletions.
diff --git a/tutorials/README.md b/tutorials/README.md
@@ -9,18 +9,121 @@ Learn how to quickly quantize pre-trained models using MCT's post-training quant
 - [Post training quantization with Keras](notebooks/imx500_notebooks/keras/example_keras_mobilenetv2_for_imx500.ipynb)
 - [Post training quantization with PyTorch](notebooks/mct_features_notebooks/pytorch/example_pytorch_post_training_quantization.ipynb)
 
-## MCT Features
-This set of tutorials covers all the quantization tools provided by MCT. 
-The notebooks in this section demonstrate how to configure and run simple and advanced post-training quantization methods.
-This includes fine-tuning PTQ (Post-Training Quantization) configurations, exporting models,
-and exploring advanced compression techniques. 
+## MCT features
+This tutorial set introduces the various quantization tools offered by MCT. 
+The notebooks included here illustrate the setup and usage of both basic and advanced post-training quantization methods. 
+You'll learn how to refine PTQ (Post-Training Quantization) settings, export models, and explore advanced compression 
+techniques such as GPTQ (Gradient-Based Post-Training Quantization), Mixed precision quantization and more.
 These techniques are essential for further optimizing models and achieving superior performance in deployment scenarios.
-- [MCT Features notebooks](notebooks/mct_features_notebooks/README.md)
 
-## Quantization for Sony-IMX500 deployment
+### Keras Tutorials
 
-This section provides several guides on quantizing pre-trained models to meet specific constraints for deployment on the
-[Sony-IMX500](https://developer.sony.com/imx500/) processing platform. 
-We will cover various tasks and demonstrate the necessary steps to achieve efficient quantization for optimal
-deployment performance.
-- [MCT IMX500 notebooks](notebooks/imx500_notebooks/README.md)
+<details id="keras-ptq">
+  <summary>Post-Training Quantization (PTQ)</summary>
+
+  | Tutorial                                                                                   | Included Features                                                                                   |
+  |--------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------|
+  | [Basic Post-Training Quantization (PTQ)](keras/example_keras_post-training_quantization.ipynb)          | &#x2705; PTQ                                                                                        |
+  | [MobileNetV2](../imx500_notebooks/keras/example_keras_mobilenetv2_for_imx500.ipynb)        | &#x2705; PTQ                                                                                        |
+  | [Mixed-Precision MobileNetV2](keras/example_keras_mobilenet_mixed_precision.ipynb)         | &#x2705; PTQ <br/> &#x2705; Mixed-Precision                                                         |
+  | [Nanodet-Plus](../imx500_notebooks/keras/example_keras_nanodet_plus_for_imx500.ipynb)      | &#x2705; PTQ                                                                                        |
+  | [EfficientDetLite0](../imx500_notebooks/keras/example_keras_effdet_lite0_for_imx500.ipynb) | &#x2705; PTQ <br/> &#x2705; [sony-custom-layers](https://github.com/sony/custom_layers) integration |
+
+</details>
+
+<details id="keras-gptq">
+  <summary>Gradient-Based Post-Training Quantization (GPTQ)</summary>
+
+  | Tutorial                     | Included Features       |
+  |------------------------------|---------------|
+  | [MobileNetV2](keras/example_keras_mobilenet_gptq.ipynb)           | &#x2705; GPTQ |
+
+</details>
+
+<details id="keras-qat">
+  <summary>Quantization-Aware Training (QAT)</summary>
+
+  | Tutorial                                          | Included Features      |
+  |---------------------------------------------------|--------------|
+  | [QAT on MNIST](keras/example_keras_qat.ipynb) | &#x2705; QAT |
+
+</details>
+
+
+<details id="keras-pruning">
+  <summary>Structured Pruning</summary>
+
+  | Tutorial                                                            | Included Features          |
+  |---------------------------------------------------------------------|------------------|
+  | [Fully-Connected Model Pruning](keras/example_keras_pruning_mnist.ipynb) | &#x2705; Pruning |
+
+</details>
+
+<details id="keras-export">
+  <summary>Export Quantized Models</summary>
+
+  | Tutorial                                                                              | Included Features         |
+  |---------------------------------------------------------------------------------------|-----------------|
+  | [Exporter Usage](keras/example_keras_export.ipynb) | &#x2705; Export |
+
+</details>
+
+<details id="keras-debug">
+  <summary>Debug Tools</summary>
+
+  | Tutorial                                                                            | Included Features       |
+  |-------------------------------------------------------------------------------------|-------------------------|
+  | [Network Editor Usage](keras/example_keras_network_editor.ipynb) | &#x2705; Network Editor |
+
+</details>
+
+### Pytorch Tutorials
+
+
+<details id="pytorch-ptq">
+  <summary>Post-Training Quantization (PTQ)</summary>
+
+  | Tutorial                                                                                                  | Included Features                           |
+  |-----------------------------------------------------------------------------------------------------------|---------------------------------------------|
+  | [Basic Post-Training Quantization (PTQ)](pytorch/example_pytorch_post_training_quantization.ipynb)        | &#x2705; PTQ                                |
+  | [Mixed-Precision Post-Training Quantization](pytorch/example_pytorch_mixed_precision_ptq.ipynb)           | &#x2705; PTQ <br/> &#x2705; Mixed-Precision |
+  | [Advanced Gradient-Based Post-Training Quantization (GPTQ)](pytorch/example_pytorch_mobilenet_gptq.ipynb) | &#x2705; GPTQ                               |
+
+</details>
+
+<details id="pytorch-pruning">
+  <summary>Structured Pruning</summary>
+
+  | Tutorial                                                                             | Included Features          |
+  |--------------------------------------------------------------------------------------|------------------|
+  | [Fully-Connected Model Pruning](pytorch/example_pytorch_pruning_mnist.ipynb) | &#x2705; Pruning |
+
+
+</details>
+
+<details id="pytorch-data-generation">
+  <summary>Data Generation</summary>
+
+  | Tutorial                                                                                            | Included Features                                                                                    |
+  |-----------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------|
+  | [Zero-Shot Quantization (ZSQ) using Data Generation](pytorch/example_pytorch_data_generation.ipynb) | &#x2705; PTQ <br/> &#x2705; ZSQ <br/> &#x2705; Data-Free Quantization <br/> &#x2705; Data Generation |
+
+</details>
+
+
+<details id="pytorch-export">
+  <summary>Export Quantized Models</summary>
+
+  | Tutorial                                                                              | Included Features         |
+  |---------------------------------------------------------------------------------------|-----------------|
+  | [Exporter Usage](pytorch/example_pytorch_export.ipynb) | &#x2705; Export |
+
+</details>
+<details id="pytorch-xquant">
+  <summary>Quantization Troubleshooting</summary>
+
+  | Tutorial                                                                                       | Included Features |
+  |------------------------------------------------------------------------------------------------|-------------------|
+  | [Quantization Troubleshooting using the Xquant Feature](pytorch/example_pytorch_xquant.ipynb) | &#x2705; Debug    |
+
+</details>
diff --git a/...mct_model_garden/models_keras/__init__.py → tutorials/keras/__init__.py b/...mct_model_garden/models_keras/__init__.py → tutorials/keras/__init__.py
diff --git a/...e_keras_activation_threshold_search.ipynb → ...e_keras_activation_threshold_search.ipynb b/...e_keras_activation_threshold_search.ipynb → ...e_keras_activation_threshold_search.ipynb
diff --git a/..._keras_activation_z_score_threshold.ipynb → ..._keras_activation_z_score_threshold.ipynb b/..._keras_activation_z_score_threshold.ipynb → ..._keras_activation_z_score_threshold.ipynb
diff --git a/...otebooks/keras/example_keras_export.ipynb → tutorials/keras/example_keras_export.ipynb b/...otebooks/keras/example_keras_export.ipynb → tutorials/keras/example_keras_export.ipynb
diff --git a/.../keras/example_keras_mobilenet_gptq.ipynb → .../keras/example_keras_mobilenet_gptq.ipynb b/.../keras/example_keras_mobilenet_gptq.ipynb → .../keras/example_keras_mobilenet_gptq.ipynb
diff --git a/...ple_keras_mobilenet_mixed_precision.ipynb → ...ple_keras_mobilenet_mixed_precision.ipynb b/...ple_keras_mobilenet_mixed_precision.ipynb → ...ple_keras_mobilenet_mixed_precision.ipynb
diff --git a/.../keras/example_keras_network_editor.ipynb → .../keras/example_keras_network_editor.ipynb b/.../keras/example_keras_network_editor.ipynb → .../keras/example_keras_network_editor.ipynb
diff --git a/...le_keras_post-training_quantization.ipynb → ...le_keras_post-training_quantization.ipynb b/...le_keras_post-training_quantization.ipynb → ...le_keras_post-training_quantization.ipynb
diff --git a/...s/keras/example_keras_pruning_mnist.ipynb → ...s/keras/example_keras_pruning_mnist.ipynb b/...s/keras/example_keras_pruning_mnist.ipynb → ...s/keras/example_keras_pruning_mnist.ipynb
diff --git a/...s_notebooks/keras/example_keras_qat.ipynb → tutorials/keras/example_keras_qat.ipynb b/...s_notebooks/keras/example_keras_qat.ipynb → tutorials/keras/example_keras_qat.ipynb
diff --git a/...otebooks/keras/example_keras_xquant.ipynb → tutorials/keras/example_keras_xquant.ipynb b/...otebooks/keras/example_keras_xquant.ipynb → tutorials/keras/example_keras_xquant.ipynb
diff --git a/tutorials/mct_model_garden/README.md b/tutorials/mct_model_garden/README.md
diff --git a/tutorials/mct_model_garden/__init__.py b/tutorials/mct_model_garden/__init__.py
diff --git a/tutorials/mct_model_garden/evaluation_metrics/__init__.py b/tutorials/mct_model_garden/evaluation_metrics/__init__.py