pytorch-semseg

pytorch-semseg copied to clipboard

INFO:ptsemseg:Iter [260/300000] Loss: 2.5116 Time/Image: 0.2828 lr=0.000020 [ 0 4 5 7 9 10 12] [ 0 4 5 7 9 10 12] [ 0 5 7 10 12] [...

18022443868

This is PR test.

OhJoon-Kwon

Can you specify me the optimizer required for unet model on Pascal Voc segmentation using focal loss. Should I have to use any learning rate schedulers?

saiphanish7

Hi Thanks for efforts Could you plz post your implementations results?

ali-masoudi

Problem solved, question was stupid.

schmitt-muc

Dataloader Type Error

2

Thank you for your code. I met a problem: /content/pytorch-semseg RUNDIR: runs/fcn8s_cityscapes/19903 Found 2975 train images Found 500 val images INFO:ptsemseg:Using cross_entropy with {'size_average': False} params INFO:ptsemseg:Using loss functools.partial(, size_average=False)...

ReinaFeng

Are pre-trained SegNet model/weights made available?

1

Great work on the repo! Thank you! I was planning to use SegNet as an intermediate step for another learning task. I have two options available: 1. Find a pre-trained...

ambareeshsrja16

Hello, with which network are the best results achieved, it would be nice to have also the configuration file, i want to test the cityscapes network!!! Best regards.

MaximumProgrammer

ValueError: Expected input batch_size to match target batch_size when running fcn8 on Pascal

4

Hi! I'm trying `python train.py` with default settings on the Pascal dataset with fcn8. After 1000 iterations I get the following error: `Traceback (most recent call last): File "train.py", line...

mariaka

RuntimeError: invalid argument 0: Sizes of tensors must match except in dimension 1. Got 46 and 47 in dimension 3 at /opt/conda/conda-bld/pytorch_1550780889552/work/aten/src/THC/generic/THCTensorMath.cu:83

5

$ python3 train.py RUNDIR: runs/unet_pascal/82417 Traceback (most recent call last): File "train.py", line 231, in train(cfg, writer, logger) File "train.py", line 129, in train outputs = model(images) File "/home/lxc/anaconda3/lib/python3.6/site-packages/torch/nn/modules/module.py", line...

li0128