[Feat] Add quantized exports for Resnet18 and LeNet

Context

This MR aims to enable the quantized (int8) export of LeNet and Resnet18.

It brings a lot of changes to clean and update the CPP Aidge export with the latest features.

Refactor

• Operators register files are splitted from one global operators.py file to several ones, each matching the actual implementation of one given kernel.
• Some utils functions are added to simplify the export file :
  • cpp_fuse_to_metaops() : Fuse the model nodes into the metaoperators corresponding to the operators supported by the export.
  • set_nodes_names() : Rename the newly created nodes based on their type, position in the graph and number of iterations of this specific layer, to ensure readability once exported.
  • set_nodes_datatypes() : Datatypes have to be set manually after the quantization step.
  • read_log_file() : May be useful to compare intermediate results...

New features

General improvements

• Now possible to ignore some producers during export phase, as some of them are not relevant (for instance, the producers holding scaling coefficients, which are handled separately).
• Change the kernels_to_copy export node system as the old system had some limitations. Indeed, it was not possible to chose a destination path different from the source path. Also the copied kernel would be automatically included in the fwd file, sometime leading to unused includes. The new system is based on a list of dict holding, for each kernel file, the source and destination paths as well as a boolean to indicate if the kernel needs to be included in the forward file.
• Add an aidge_cmp feature which allows to compare the aidge_export_cpp and aidge_backend_cpu intermediate tensor results for the exported model.

New models examples

• Add LeNet generation file (create_lenet.py) which create, train and export a LeNet model.
• Add a lenet.py file describing the different steps to perform an export with the aidge_export_cpp module.
• Add a resnet18.py file describing the different steps to perform an export with the aidge_export_cpp module.
• Add these example scripts to the CI

Quantized Export

• Handle the scaling nodes added by the PTQ and transmit the shift and coef parameters to the generated export.
• Add scaling functions (SingleShift and FixedPoint) to the export.
• Add a aidge_cmp option which exports the log_outputs (intermediate tensors generated by the aidge_backend_cpu) as reference and allow data comparison at inference time.
• Now possible to export quantized LeNet using --dtype int8 while running the lenet.py file.
• Now possible to export quantized Resnet18 using --dtype int8 while running the resnet18.py file.

Related Issues

[ResNet18] Export Cpp INT8 not supported

TODO

• LeNet Export fp32
• LeNet Export int8
• Resnet18 Export fp32
• Resnet18 Export int8

changed milestone to %aidge v0.6.0

added Enhancement ⭐ Feature 🚀 Refactoring🎨 StatusWork in Progress TopicExport User Request 🖐 labels

requested review from @cmoineau

assigned to @axelfarr

changed the description

mentioned in merge request aidge_core!369 (merged)

added 5 commits

• 53ced27c - [Feat] Add int8 support
• c786170b - [Feat] Add the aidge_cmp option and refactor save_outputs function
• c2553503 - [Fix] Remove memory info from arg list for save_outputs function
• a7c84c70 - [Fix] Temporarily generate only layer params for the first input as all inputs...
• 7efbd360 - [Feat] Change the kernels_to_copy system

Compare with previous version

marked the checklist item LeNet Export int8 as completed

changed the description

assigned to @mick94

mentioned in merge request !24 (closed)

added 19 commits

• 7efbd360...45d12a67 - 2 commits from branch dev
• 45d12a67...8773a8a3 - 7 earlier commits
• 6c40bd8e - [Chore] Remove unecessary tabulation
• 5eb416c8 - [Chore] Include export_utils file
• 2bf25de0 - [Chore] Temporarily comment the save_outputs generated code
• a6389992 - [Refactor] Reformat transpose jinja
• 73636be9 - [Feat] Add new lenet export and generation files
• b9066014 - [Feat] Add int8 support
• 1ca57eea - [Feat] Add the aidge_cmp option and refactor save_outputs function
• 5e9baa6d - [Fix] Remove memory info from arg list for save_outputs function
• cf68a996 - [Fix] Temporarily generate only layer params for the first input as all inputs...
• ba2b04cc - [Feat] Change the kernels_to_copy system

Compare with previous version

added 1 commit

• aad912c8 - [Fix] MetaOp Recipe Pooling

Compare with previous version

marked the checklist item Resnet18 Export int8 as completed

marked the checklist item Resnet18 Export fp32 as completed

added 20 commits

• aad912c8...e4317fd7 - 10 earlier commits
• 7801dc6f - [Feat] Adaptation of the Aidge_cmp function for an integer datatype
• b063bd16 - [Refactor] Change formatting
• 56ecbd41 - [Feat] Add jinja file for forward build
• cea36246 - [Fix] Add #define to enable or disable the OpenMP option for compilation
• cf3509dd - [Feature] Add Python script for ResNet18 export (cpp)
• 1057b2e0 - [Git] Add gitignores
• 55b4d096 - [Fix] Suppress constexpr and simplified the choice of rounding
• 3f58cc01 - [Refactor] Simplify aidge_cmp function to one function for float or interger
• c2cfa8d1 - Fix Add ignore attribute to False by default
• 00890524 - Merge branch 'feat_add_quantized_export' into 'feat_add_quantized_export'

Compare with previous version

added 1 commit

• b7e28ee1 - [Chore] Class typo in foc comment

Compare with previous version