GitHub - Aryabb29/Exploring-Compression-Techniques-on-Deep-Neural-Network: Deep Neural Networks (DNNs) can be compressed in a 3-stage pipeline: Pruning, Quantization and Huffman COding. This 3-stage pipeline can significantly reduce the memory usage of DNNs without inducing much performance degradation.

Aryabb29 / Exploring-Compression-Techniques-on-Deep-Neural-Network Public

Notifications You must be signed in to change notification settings
Fork 0
Star 1

Deep Neural Networks (DNNs) can be compressed in a 3-stage pipeline: Pruning, Quantization and Huffman COding. This 3-stage pipeline can significantly reduce the memory usage of DNNs without inducing much performance degradation.

1 star 0 forks Branches Tags Activity

Star

Notifications

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 4 Commits
Checkpoints		Checkpoints
aspotti_DeepCompression.ipynb		aspotti_DeepCompression.ipynb
huffman_coding.py		huffman_coding.py
prune.py		prune.py
pruned_layers.py		pruned_layers.py
quantize.py		quantize.py
summary.py		summary.py
train_util.py		train_util.py
vgg16.py		vgg16.py