Commits

Commits on Feb 17, 2019

Fix issue IntelLabs#157
nzmora
committedFeb 17, 2019
AMC: added configuration option to set the frequency of computing a reward
nzmora
committedFeb 17, 2019
Filter pruning: rank filters by mean value of feature-map channels
nzmora
committedFeb 17, 2019

Commits on Feb 14, 2019

Store config files in logdir/configs directory (IntelLabs#156 )

barrh
authored and
nzmora
committedFeb 14, 2019
Fix automated-compression imports
nzmora
committedFeb 14, 2019
Create directory to hold all YAML schedules for training baseline networks
nzmora
committedFeb 14, 2019

Commits on Feb 13, 2019

Automatic RL compression + Greedy compression (IntelLabs#151 )
nzmora
committedFeb 13, 2019

Commits on Feb 12, 2019

CPU support: fix the case of loading a thinned GPU-model on the CPU
nzmora
committedFeb 12, 2019
Post-train quant bugfix: squeeze scale factor shape when quantizing bias
guyjacob
committedFeb 12, 2019
Fix issue IntelLabs#148 + refactor load_checkpoint.py (IntelLabs#153 )
nzmora
committedFeb 12, 2019

Commits on Feb 11, 2019

Update README.md
nzmora
committedFeb 11, 2019
Update README.md
guyjacob
committedFeb 11, 2019
Post-train quant based on stats + additional modules quantized (IntelLabs#136 )
guyjacob
committedFeb 11, 2019

Commits on Feb 10, 2019

Load different random subset of dataset on each epoch (IntelLabs#149 )
guyjacob
committedFeb 10, 2019

Commits on Feb 6, 2019

Commits on Jan 31, 2019

compress_classifier: add command-line option to "thinnify" a model
nzmora
committedJan 31, 2019
More robust handling of loading non-Distiller checkpoints
nzmora
committedJan 31, 2019

Commits on Jan 27, 2019

DoReFa 1-bit weights (IntelLabs#127 )

JoyFYan
authored and
guyjacob
committedJan 27, 2019

Commits on Jan 24, 2019

Bugfix in test_quantizer
guyjacob
committedJan 24, 2019
Range linear Quant-aware training: Require single GPU, and more
guyjacob
committedJan 24, 2019

Commits on Jan 23, 2019

Quant-aware training: Quantize bias to 32 bits (Hard-coded for now)
guyjacob
committedJan 23, 2019
Fix dtypes in test_post_train_quant.py
guyjacob
committedJan 23, 2019

Commits on Jan 22, 2019

Fix pruner base class bug. (IntelLabs#131 )

innerNULL
authored and
nzmora
committedJan 22, 2019

Commits on Jan 21, 2019

Simplenet: Replace all functional operator calls with explicit layers
guyjacob
committedJan 21, 2019
Updates to quantization parameters calculation:
guyjacob
committedJan 21, 2019

Commits on Jan 16, 2019

compress_classifier.py refactoring (IntelLabs#126 )

barrh
authored and
nzmora
committedJan 16, 2019
Fix for CPU support
nzmora
committedJan 16, 2019

Commits on Jan 15, 2019

CPU support: fix thinning directive tensor migration to CPU/GPU
nzmora
committedJan 15, 2019
Fix for CPU evaluation use-case
nzmora
committedJan 15, 2019