omg777

class_list.txt

11

Hi, thanks for your nice codes. I modified `class_list.txt` and then run main.py, It shows aeroplane AP, bicycle AP, bird AP ,... regardless of the class_list file. How can I...

Hi, Can you explain why multiply 0.5 from gradient ? ``` # Average the gradients for p in D_shared.parameters(): p.grad.data = 0.5 * p.grad.data ```

Distributions and Histograms

1

Hi, thanks for the nice work. Can you explain about distributions and histograms in tensorboard ? How can I understand?

Help me from tutorial code error.

12

Thanks for your tutorial from scratch. It helps me a lot. In article part 3, there are some codes you wrote. I copy that codes but some error for me....

In adapt.py ,81 line. Why optimizer_critic.zero_grad() is needed ? ``` # zero gradients for optimizer optimizer_critic.zero_grad() optimizer_tgt.zero_grad() ```

About the classifiers.

1

Thanks for the code ! I have a question about the classifier. What is the differences between F1 and F2 classifier? I can't find any differences in the code of...

"For CIFAR-10 [34], 4,000 labeled examples are kept as labeled data while 41,000 examples are used as unlabeled data" Why did you use whole data for the unlabeled data? ```...

What is "contrast_mode" in loss?

2

Thanks for your code. What is "contrast_mode" in loss?

Questions about t-SNE plot.

2

Hi, Thanks for you code sharing. I have a question about t-SNE plot. Which features are used to draw t-SNE plot? features after encoder or after classifier ?

I can`t understand about PS-ROI-Pooling. In paper figure 3,4, why there is all different 9 score maps ?