Defa Zhu

icon location deep learning & computer vision

Results 8 comments of Defa Zhu

training miou 47~48

![image](https://user-images.githubusercontent.com/23050243/77218314-f5070100-6b64-11ea-8cce-ef8682417899.png) I just migrated it to torch 1.4. It looks not something wrong.

synthia_mapped_to_cityscapes(labels)

+1

Inception Score calculation

So,which version is correct？

train_val.prototxt and solver.prototxt for Resnet-152 layer

@tyrosathish can you share me your resnet-152.caffemodel?i can't parse the caffemodel which i download from the link. Check failed: ReadProtoFromBinaryFile(param_file, param) Failed to parse NetParameter file: ResNet-152-model.caffemodel

the normalize function in utils.py

Hi, I want to know which version is better in the classification.

256*256

> Hello, have you tried pytorch to implement 256*256 network in SNGAN? Sorry for the late reply. I don't often check the status of GitHub because our company uses GitLab....

Questions about the formula (2) in the paper

thx for your reply！ ![image](https://user-images.githubusercontent.com/23050243/43898509-a3d8e646-9c11-11e8-9edb-35a3d277793a.png) And I am not sure why the last item in the (5) is a constant as described in the paper. ![image](https://user-images.githubusercontent.com/23050243/43898700-3f78b964-9c12-11e8-9bec-8d62cbbf034f.png)

Are the Parameter of the chainer version of inception model the same as the TF's ?

I think the code is correct. I also found that the inception score and FID by the pytorch code is different from TF code. I think it is because the...