Generate report for troubleshooting the quantized model#1090

Merged

reuvenperetz merged 50 commits intosony:mainfrom reuvenperetz:xquant

Jun 17, 2024

+3,089-2

Commits on May 27, 2024

Init commit of xquant
reuvenp
committed
remove commented out code
reuvenp
committed
add dummy methods if fw packages were not found
reuvenp
committed

Commits on May 29, 2024

add histograms collection
reuvenp
committed
Fix comments in pytorch facade
reuvenp
committed

Commits on Jun 3, 2024

Remove core config comments
reuvenp
committed
simplify xquantconfig
reuvenp
committed
rename core report generator
reuvenp
committed
add similarity metrics
reuvenp
committed
Add dataset utils
reuvenp
committed
Add model folding module
reuvenp
committed

Commits on Jun 4, 2024

add tensorboard utils
reuvenp
committed
remove logger and use python logging
reuvenp
committed
Move default similarity metrics names to SimilarityMetrics
reuvenp
committed

Commits on Jun 5, 2024

split similarity into module with similairty functions and similarity calculator
reuvenp
committed
add tests for more networks
reuvenp
committed
Add tests to workflows
reuvenp
committed
add temp logger
reuvenp
committed

Commits on Jun 6, 2024

finalize logger
reuvenp
committed
add comments
reuvenp
committed
rename wrapped_dataset to prepare_dataset
reuvenp
committed

Commits on Jun 9, 2024

rename metrics to similarity metrics
reuvenp
committed
add comments
reuvenp
committed
move package into model_compression_toolkit
reuvenp
committed
add keras notebook
reuvenp
committed
add pytorch notebook
reuvenp
committed
fix imports in keras notebook
reuvenp
committed

Commits on Jun 10, 2024

replace inference manner to solve issue with tf2.13
reuvenp
committed

Commits on Jun 13, 2024

use mct logger
reuvenp
committed
remove duplicated license
reuvenp
committed
rename arguments in similairty functions to be more general
reuvenp
committed
avoid double access to dict in similarity calculator
reuvenp
committed
rename ModelAnalyzerUtils to ModelAnalyzer
reuvenp
committed
add no cover for abstract methods error
reuvenp
committed
use tag for tensorboard artifacts
reuvenp
committed
take stats collection out of tensorboard class
reuvenp
committed
store model outputs in activation dict during activations extraction
reuvenp
committed
add keras model analyzer unit tests
reuvenp
committed

Commits on Jun 15, 2024

add similarity functions unit tests
reuvenp
committed

Commits on Jun 16, 2024

add tests to tensorboard
reuvenp
committed
add colab links to notebooks
reuvenp
committed
remove outputs from keras notebook
reuvenp
committed
fix install sections in notebooks
reuvenp
committed
remove sys part from notebooks
reuvenp
committed
fix bad import in pytorch notebook
reuvenp
committed
fix PR comments
reuvenp
committed
use different images in notebooks with red marks for clarity
reuvenp
committed
limit numpy to version < 2.0 that was just released
reuvenp
committed

Commits on Jun 17, 2024

Revert "limit numpy to version < 2.0 that was just released"
reuvenp
committed
Merge remote-tracking branch 'origin/main' into xquant
reuvenp
committed