-
I've noticed that you have tested slimmable techs on SR tasks. How can I reproduce the US-WDSR result you mentioned in [Universally_Slimmable_Networks_and_Improved_Training_Techniques]? Or is there an…
-
It is not easy to deploy gate operator with some other backends, like TensorRT.
So my question is can we futher improve autoalim without the dynamic gate when inference?Any ongoing work are doing t…
twmht updated
2 years ago
-
Hi @liuzhuang13,
There is another issue I met in my experiments: inference time reduced when **channel number to be pruned was a pow of 2** , otherwise it increased and longer than baselin…
-
Hi Jiahui,
Thanks for the great work. I'm trying to reproduce AutoSlim for CIFAR-10 (Table 2).
Could you please provide a detailed hyperparameter you used for it?
I'm able to train the baseline…
-
微博内容精选
-
In my specific use case I would like to use slices (i.e. `StridedSlice`) on Tensors with `begin:end` being defined by values provided at runtime. Looks like this is currently not supported in TFLM 🥺
…
-
The below model (when converted to TF-Lite) maps a constant (read only) tensor as input to a `IF` op which causes some problems with the new implementation of the `IF` operation as discussed in #853 …
-
Hi,
I found out the original implementation recalibrate BN stats after training.
https://github.com/JiahuiYu/slimmable_networks/blob/5dc14d0357ccfc596d706281acdc8a5b0b66c6d6/models/slimmable_ops…
twmht updated
2 years ago
-
Hi,
I have tried to prune object detection model with autoslim.
But the accuracy is always low, even the training loss is normal.
I have checked the original repo (https://github.com/JiahuiYu…
twmht updated
2 years ago
-
Thank you for your very nice work,I want to know that the effect of gumble softmax，because I think the network can be trained without gumble softmax.
Is the gumbel softmax just aimed to increase the…