Hsuan
Hsuan
@aliericcantona , when I install r1.6, it recommended cuda 9.0, I don't know whether this is a problem or not. However, I use tf 1.4 instead of tf 1.6, maybe...
You can use the pre-trained model by selecting restore variables. Following is the example: ```python restore_var = [v for v in tf.global_variables() if 'conv6_cls' not in v.name ```
Hey @jacobdang, I think you should contact the original author for details first, I have no idea on such things. But I am glad to know how you can apply...
Hi @jacobdang , since the paper did not specify the architecture of the network, I need to refer to the author's code which implemented in Caffe, and re-implement the code...
@manuel-88, I think the most importance thing is to train on the un-pruned model, and then do model compression based on the l1 norm ( described in the paper). The...
Hi @MikeyLev, you can take a look at the paper. The author said they pruned the filter size to half. So the filter size is double before pruning. Can you...
@david-bernstein , I will update `demo.ipynb` in recent weeks. Thank you for your advice!
Hey @dbredep, You can take a look at the parsing function in `utils.image_reader.py`. What you need to do is to change the decode function!
Hey @chamcham115, sorry for late reply. In this repo, I haven't implemented model pruning, and I'm trying to find some solutions to prune the model (many model pruning approaches are...
Maybe you should update the batch norm variable first. Btw, if you use this code in your github repo, you need to refer to this repository.