PyPi install guides improvements

[slyubimt]

Details:

• Human rights notice
• Components descriptions update & console scripts
• Extras requirements definition
• Change verification step to Model Optimizer call

Tickets:

• CVS-49252

[slyubimt]

@helena-intel , @ryanloney You do not mind if we add a link to OpenVINO™ Notebooks in Additional Resources? This will provide a quick start for OpenVINO Python API users.

[helena-intel]

@helena-intel , @ryanloney You do not mind if we add a link to OpenVINO™ Notebooks in Additional Resources? This will provide a quick start for OpenVINO Python API users.

I think that is useful indeed, thanks! A minor note: our README also covers installing openvino-dev, so it has some overlap with this README. In a feature branch our README is more aligned, we have an "if you already installed openvino-dev, skip to step 3" message. That will be merged soon.

[slyubimt]

Improved document codestyle according to @andrew-zaytsev recommendations.

[SDxKeeper]

looks nice! especially section with openvino-dev[extras]
@slyubimt Am I right that MO dependencies will be installed in this scenario?

[slyubimt]

looks nice! especially section with openvino-dev[extras]
@slyubimt Am I right that MO dependencies will be installed in this scenario?

yes. You can use this table to check :

Extra	Component	Requirements
caffe	MO	https://github.com/openvinotoolkit/openvino/blob/master/model-optimizer/requirements_caffe.txt
kaldi	MO	https://github.com/openvinotoolkit/openvino/blob/master/model-optimizer/requirements_kaldi.txt
mxnet	MO	https://github.com/openvinotoolkit/openvino/blob/master/model-optimizer/requirements_mxnet.txt
onnx	MO	https://github.com/openvinotoolkit/openvino/blob/master/model-optimizer/requirements_onnx.txt
tensorflow	MO	https://github.com/openvinotoolkit/openvino/blob/master/model-optimizer/requirements_tf.txt
tensorflow2	MO, Model Downloader	https://github.com/openvinotoolkit/openvino/blob/master/model-optimizer/requirements_tf2.txt https://github.com/openvinotoolkit/open_model_zoo/blob/develop/tools/downloader/requirements-tensorflow.in
pytorch	Model Downloader	https://github.com/openvinotoolkit/open_model_zoo/blob/develop/tools/downloader/requirements-pytorch.in
caffe2	Model Downloader	https://github.com/openvinotoolkit/open_model_zoo/blob/develop/tools/downloader/requirements-caffe2.in

[baychub]

Here are language edits for Andrey to incorporate (just the Component table):

Component	Console Script	Description
Model Optimizer	mo	Model Optimizer imports, converts, and optimizes models that were trained in popular frameworks to a format usable by Intel tools, especially the Inference Engine. Supported frameworks include Caffe*, TensorFlow*, MXNet*, and ONNX*.
Benchmark Tool	benchmark_app	Benchmark Application allows you to estimate deep learning inference performance on supported devices in synchronous and asynchronous modes.
Accuracy Checker and Annotation Converter	accuracy_check convert_annotation	Accuracy Checker is a deep learning accuracy validation tool that allows you collect accuracy metrics against popular datasets. The main advantages of the tool are the flexibility of configuration and an impressive set of supported datasets, preprocessing, postprocessing, and metrics. Annotation Converter is a utility for offline conversion dataset to suitable for metric evaluation format used in Accuracy Checker.
Post-Training Optimization Tool	pot	Post-Training Optimization Tool allows you to optimize trained models with advanced capabilities, such as quantization and low-precision optimizations, without the need to re-train or fine-tune models.
Model Downloader and other OMZ tools	omz_downloader omz_converter omz_quantizer omz_info_dumper	Model Downloader is a tool for getting access to collection of high quality and extremely fast pre-trained deep learning public and intel-trained models. These free pre-trained models can be used instead of training your own models, to speed up the development and production deployment process. The purpose of the tool is as follows: it downloads model files from online sources and, if necessary, patches them to make them more usable with Model Optimizer. A number of additional tools are also provided to automate the process of working with downloaded models: Model Converter is a tool for conversion the models that are stored not in the Inference Engine IR format into that format using Model Optimizer. Model Quantizer is a tool for automatic quantization full-precision models in the IR format into low-precision versions using the Post-Training Optimization Tool. Model Information Dumper is a helper utility for dumping information about the models to a stable, machine-readable format.

[helena-intel]

A few suggestions about the edited table:

"Model Quantizer is a tool for automatic quantization full-precision models in the IR format into low-precision versions using the Post-Training Optimization Tool." suggest changing to "quantization of full-precision models"

Model Converter is a tool for conversion the models suggest changing to "conversion of the models" or "converting the models"

that are stored not in suggest changing to "that are not stored in"

Model Downloader is a tool for getting access to collection suggest changing to "getting access to a collection"

Accuracy Checker is a deep learning accuracy validation tool that allows you collect accuracy metrics suggest changing to "allows you to collect"

Annotation Converter is a utility for offline conversion dataset to suitable for metric evaluation format There are a few words missing in this sentence

an impressive set of supported datasets, preprocessing, postprocessing, and metrics. An impressive set of preprocessing? Suggest adding a noun.

Would it be helpful to modify the description for model converter and quantizer to say "OMZ models" instead of models, because these tools only work for OMZ models? omz is in the name of the command, but not everyone may make the connection to "only for OMZ models".

[ryanloney]

Thank @helena-intel for catching these typos.

My suggestion is to change Annotation Converter is a utility for offline conversion dataset to suitable for metric evaluation format to "Annotation Converter is a utility that converts datasets to a format suitable for accuracy measurement"

Using the word impressive does not seem appropriate here. I suggest removing it.

I agree with Helena that we should be specific about Model Converter being for Open Model Zoo models. It causes confusion about Model Optimizer with such a similar name.