Dr Yifan Liu

About pytorch encoding and the training speed

3

Hi, Thank you for your great work! I am trying to do some distillation work on the small mobile net. I have some questions: The code need the package pytorch...

The network can not be convergent

1

I feel very confused. I use the train_mobile.py to train the model with out pre-trained on the cityscapes dataset. I can only get the mIoU of 30~40. How can we...

Have you tried to train this model on MSCOCO?

1

Have you tried to train your model on MSCOCO? And you write it in the "to do" task.

What is the 'init.pth'

7

Thanks for sharing this amazing work. I wonder what is the 'init.pth' for the dsmil model. Is it trained on some specific dataset? Do I need to load it if...

Hi~Have you used the moblie net which is pretrained on the ImageNet?

5

Hello~ I am working on training the mobileNetV2+aspp these days. I found out if I only train the model on the cityscapes, the mIoU is very low. Do you use...

Batch inference?

12

For example, I have a lot of unpaired images from two domain, A and B. After training, what should I do to transfer A to B with the model?

ATT

Can you share the pre-train weight of IGCV3 on the ImageNet?

3

Hi, Great work! Can you share the pre-train weight of IGCV3 on the ImageNet? I am working on doing some segmentation tasks with the backbone of IGCV3. Can you share...

The inference time

1

Hi, great work! I just want to ask about the inference time. You report the inference time /per frame for baseline DL-18 and AL-18 and DFF. How to calculate this...

What is the meaning of self.v in the MSDN.py

2

Dear author, Thanks for sharing the code. According to Eq 1 in your paper, the attribute, w1, and visual features are timed together. However, in the MSDN.py the code has...