dolphintear / tensorflow_model_pruning Public

Notifications You must be signed in to change notification settings
Fork 0
Star 2

Train a pruned CNN on CIFAR-10 dataset with TensorFlow model_pruning module

2 stars 0 forks Branches Tags Activity

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 1 Commit
CIFAR-10		CIFAR-10
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
TFRecords_encoder.py		TFRecords_encoder.py
frozen_graph.py		frozen_graph.py
model_quantization.py		model_quantization.py
networks.py		networks.py
test_pb.py		test_pb.py
test_tflite.py		test_tflite.py
train.py		train.py

Repository files navigation

tensorflow_model_pruning

Train a pruned CNN on CIFAR-10 dataset with TensorFlow model_pruning module

Environment

python==3.6.5
tensorflow==1.14.0

Getting Started

Download and extract the CIFAR-10 dataset in the "CIFAR-10" folder.
Run "TFRecords_encoder.py" to create TFRecords of the dataset.
Run "train.py" to train the network with and without pruning.
Run "frozen_graph.py" and "model_quantization.py" to convert model format.
Run "test_pb.py" and "test_tflite.py" to test models on CIFAR-10 test set.

About

Train a pruned CNN on CIFAR-10 dataset with TensorFlow model_pruning module

Report repository

Releases

No releases published

Packages

No packages published

Languages

Python 100.0%