search

bfortuner / pytorch_tiramisu

FC-DenseNet in PyTorch for Semantic Segmentation
MIT License
306 stars 66 forks source link

Performance cannot approach as mention #18

Open tonyli0803 opened 5 years ago

tonyli0803 commented 5 years ago

hi @bfortuner , I' m interested in your pytorch_tiramisu.

I use the train.ipynb as the backbone to train the model. However, I cannot approach the performance as you do.

My FC-DenseNet67 result converge at 0.8357994871794872.

It will be appreciate to help me figure out my questions. Here are my questions:

  1. The origin code doesn't have weighted loss function. The weighted loss function do improve iou, but harm to pixel accuracy. How did you get those weight?

  2. I see some past issue mention that the train.ipynb is slightly different from the original train code. Do you add a lot of things on it? Can you send me the original code for further comparison ?

Here is the modification from the code :

in train.ipynb

  1. N_EPOCHS -> 1000

  2. To use multi-gpu I add this line

image

  3. To implement "exclude the 'background' class" , I add this lines. Try to set 11 class, and 
      ignore void label loss

image

   4.  I train it directly. I didn't not finetune.

image

Please help me figure out these questions.🙏🙏🙏🙏🙇🙇🙇 My email is : tonyli0803@gmail.com