diff --git a/tools/pot/docs/DefaultQuantizationUsage.md b/tools/pot/docs/DefaultQuantizationUsage.md index 5fcec167a44..cd8b8f4170a 100644 --- a/tools/pot/docs/DefaultQuantizationUsage.md +++ b/tools/pot/docs/DefaultQuantizationUsage.md @@ -12,7 +12,7 @@ This guide describes how to apply model quantization with the Default Quantization method without accuracy control, using an unannotated dataset. To use this method, create a Python script using an API of Post-Training Optimization Tool (POT) and implement data preparation logic and quantization pipeline. If you are not familiar with Python, try [command-line interface](@ref pot_compression_cli_README) of POT which is designed to quantize models from OpenVINO [Model Zoo](https://github.com/openvinotoolkit/open_model_zoo). The figure below shows the common workflow of the quantization script implemented with POT API. -![](./images/default_quantization_flow.png) +![](./images/default_quantization_flow.svg) The script should include three basic steps: 1. Prepare data and dataset interface. diff --git a/tools/pot/docs/images/default_quantization_flow.png b/tools/pot/docs/images/default_quantization_flow.png deleted file mode 100644 index b69ee969d8d..00000000000 --- a/tools/pot/docs/images/default_quantization_flow.png +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8c9a0f1cccf731e0740cb91a3a48ff73b9f54cee0a39a03d421316a5599e7955 -size 37302 diff --git a/tools/pot/docs/images/default_quantization_flow.svg b/tools/pot/docs/images/default_quantization_flow.svg new file mode 100644 index 00000000000..12b9da7e804 --- /dev/null +++ b/tools/pot/docs/images/default_quantization_flow.svg @@ -0,0 +1,942 @@ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +