Eoin Kenny

Bug in Encoding and Decoding Parameters

3

Hi I've been messing around with this code and found an error. If I use these params in the AttGAN class ``` def __init__(self, enc_dim=64, enc_layers=6, enc_norm_fn='batchnorm', enc_acti_fn='lrelu', dec_dim=64, dec_layers=6,...

Could you please explain what --ema is in the readme? Also, could you provide a requirements.txt, that's really important when dealing with tensorflow and pytorch combined. Thank you!

Get Contributions Scores for Middle Layer

5

Is it possible to use DeepExplain to get contribution scores for a layer in the middle of the network? I presume you have to change this line `target_tensor = fModel(input_tensor)...

_If the influence function is calculated for multiple test images, the helpfulness is ordered by average helpfulness to the prediction outcome of the processed test samples._ I don't understand this....

Input size in pre-trained and paper

1

Hi, thanks a lot for the excellent repo! A small question. The input size on the pre-trained models ( https://github.com/Alibaba-MIIL/ImageNet21K/blob/main/MODEL_ZOO.md) is 224x224, but your paper seems to use 384x384? Am...

Segmentation Fault in KMeans

17

### Describe the bug Hi when I run this code ``` import numpy as np from sklearn.cluster import KMeans X_train = np.random.random((10, 2)) kmeans = KMeans(n_clusters=3, random_state=0).fit(X_train) ``` I get...

Saving Multi-GPU

1

Thanks for this repo! It seems like the code doesn't save the models correctly when using multiple GPUs right? Like if I train on multiple GPUs but load on a...