Project: Fake Quantization

[ranbir-sharma]

Project : Fake Quantization

Project Details are as follow

• Fake Quantization is added to the LLTFI to ensure fault injection within the targeted layers specified in the yaml files
• Lib code is added to facilitate the process for maintaining and Quantizing / Dequantizing the values as provided
• CMake Files are updated to build the code for Fake Quantization while setting up LLTFI

[abrahamchan]

Do you have any tests for this feature? For instance, it would be great to have an example under sample_programs/ml_sample_programs/ to demonstrate the use of fake quantization.

Also, the input_masterlist_ml.yaml should be updated with the fakeQuant option.

[ranbir-sharma]

Do you have any tests for this feature? For instance, it would be great to have an example under sample_programs/ml_sample_programs/ to demonstrate the use of fake quantization.

Also, the input_masterlist_ml.yaml should be updated with the fakeQuant option.

Added Documentation in the input_masterlist_ml.yaml with examples and added error checks for fake Quantization in instrument.py

[abrahamchan]

Could you please make a second round of changes? I didn't catch the other typos in the previous review.

[ManiSadati]

Does this PR still have the unexpected large value issue for all the models?

[abrahamchan]

LGTM!