Sparse Training via Boosting Pruning Plasticity with Neuroregeneration
Open paper
Benchmark
Model
Rank
Results
sparse-learning-on-imagenet
Resnet-50: 80% Sparse 100 epochs
#3
Top-1 Accuracy: 76
sparse-learning-on-imagenet
Resnet-50: 90% Sparse 100 epochs
#5
Top-1 Accuracy: 74.5
Rank counts only results with a code link.