CURL

CURL copied to clipboard

Seems like there needs to be some lines to be changed

1

``` File "/*/CURL/ImageNet/released_model/pruned_model.py", line 76, in __init__ index_list = list(np.load('index.npy')) File "/opt/conda/lib/python3.8/site-packages/numpy/lib/npyio.py", line 439, in load return format.read_array(fid, allow_pickle=allow_pickle, File "/opt/conda/lib/python3.8/site-packages/numpy/lib/format.py", line 727, in read_array raise ValueError("Object arrays cannot be...

ooodragon94

Question about how to use it

1

Dear Luo, I have read this paper about how select filter to prune, I have a question that should I calculate the scores of every layer's every channel ? If...

Wang-zipeng

First, thanks very much of your code, It is realy great work, however could you release the **code of organize the CUB200 dataset of train val** thanks very much

wml666

Dear @Roll920 , Despite the computational cost, KL-divergence is a very interesting criterion for assigning filter importance. However, I'm a bit confused about the KL loss implemented in your code...

arturjordao