Moved quantization templates to openvino/tools/pot (#10814)

tools/pot/docs/BestPractices.md

@@ -18,7 +18,7 @@ Post-training quantization is a basic feature of the POT and it has lots of knob
 quantized model. However, as a starting point we suggest using the `DefaultQuantization` algorithm with default settings.
 In many cases it leads to satisfied accuracy and performance speedup. 
 
-A fragment of the configuration file (`config/default_quantization_template.json` in the POT directory) with default settings is shown below:
+A fragment of the configuration file (`openvino/tools/pot/configs/templates/default_quantization_template.json` in the POT directory) with default settings is shown below:
 ```
 "compression": {
   "target_device": "ANY", // Target device, the specificity of which will be taken into account during optimization.
@@ -78,7 +78,7 @@ For example, the following `range_estimator` for activations can improve the acc
 }
 ```
 
-Find the possible options and their description in the `config/default_quantization_spec.json` file in the POT directory.
+Find the possible options and their description in the `configs/default_quantization_spec.json` file in the POT directory.
 
 4.  The next option is `stat_subset_size`. It controls the size of the calibration dataset used by POT to collect statistics for quantization parameters initialization.
 It is assumed that this dataset should contain a sufficient number of representative samples. Thus, varying this parameter may affect accuracy (higher is better). 
@@ -92,7 +92,7 @@ In case when the steps above do not lead to the accurate quantized model you may
 The whole idea behind that is to revert quantized layers back to floating-point precision based on their contribution to the accuracy drop until the desired accuracy degradation with respect to
 the full-precision model is satisfied.
 
-A fragment of the configuration file with default settings is shown below (`configs/accuracy_aware_quantization_template.json`):
+A fragment of the configuration file with default settings is shown below (`openvino/tools/pot/configs/templates/accuracy_aware_quantization_template.json`):
 ```
 "compression": {
         "target_device": "ANY", // Target device, the specificity of which will be taken into account during optimization.

tools/pot/openvino/tools/pot/algorithms/quantization/accuracy_aware/README.md

@@ -58,7 +58,7 @@ quantization time. Default value is `False`.
 ## Examples
 
 A template and full specification for AccuracyAwareQuantization algorithm:
- * [Template](https://github.com/openvinotoolkit/openvino/blob/master/tools/pot/configs/accuracy_aware_quantization_template.json)
+ * [Template](https://github.com/openvinotoolkit/openvino/blob/master/tools/pot/openvino/tools/pot/configs/templates/accuracy_aware_quantization_template.json)
  * [Full specification](https://github.com/openvinotoolkit/openvino/blob/master/tools/pot/configs/accuracy_aware_quantization_spec.json)
 
 Example of using POT API with Accuracy-aware algorithm:

tools/pot/openvino/tools/pot/algorithms/quantization/default/README.md

@@ -116,7 +116,7 @@ Enabling this option may increase compressed model accuracy, but will result in
 ## Examples
             
  A template and full specification for DefaultQuantization algorithm:
- * [Template](https://github.com/openvinotoolkit/openvino/blob/master/tools/pot/configs/default_quantization_template.json)
+ * [Template](https://github.com/openvinotoolkit/openvino/blob/master/tools/pot/openvino/tools/pot/configs/templates/default_quantization_template.json)
  * [Full specification](https://github.com/openvinotoolkit/openvino/blob/master/tools/pot/configs/default_quantization_spec.json)
 
 Command-line example:

tools/pot/openvino/tools/pot/app/run.py

@@ -76,7 +76,7 @@ def _create_log_path(config):
 
 
 def _update_config_path(args):
-    config_template_folder = os.path.join(Path(__file__).parents[4], 'configs')
+    config_template_folder = os.path.join(Path(__file__).parents[1], 'configs', 'templates')
 
     if args.quantize is not None:
         if args.quantize == 'default':

tools/pot/openvino/tools/pot/configs/templates/__init__.py

@@ -0,0 +1,2 @@
+# Copyright (C) 2022 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0

tools/pot/setup.py

@@ -25,11 +25,6 @@ class InstallCmd(install):
             # install requires
             self.do_egg_install()
 
-        def_quant_path = os.path.join("configs", "default_quantization_template.json")
-        aa_quant_path = os.path.join("configs", "accuracy_aware_quantization_template.json")
-        copyfile(def_quant_path, os.path.join(self.install_purelib, prefix, "pot", def_quant_path))
-        copyfile(aa_quant_path, os.path.join(self.install_purelib, prefix, "pot", aa_quant_path))
-
         version_txt = os.path.join(prefix, "pot", "version.txt")
         if os.path.exists(version_txt):
             copyfile(os.path.join(version_txt),
@@ -143,8 +138,9 @@ setup(
     url='https://software.intel.com/openvino-toolkit',
     packages=find_packages(exclude=["tests", "tests.*",
                                     "tools", "tools.*"]),
-    package_data={"openvino.tools.pot.configs.hardware": ['*.json'],
+    package_data={'openvino.tools.pot.configs.hardware': ['*.json'],
-                  "openvino.tools.pot.api.samples": ['*.md', '*/*.md']},
+                  'openvino.tools.pot.api.samples': ['*.md', '*/*.md'],
+                  'openvino.tools.pot.configs.templates': ['*.json']},
     include_package_data=True,
     cmdclass={
         'install': InstallCmd,