HansBambel/SmaAt-UNet

python train_SmaAtUNet.py error :

wy22vv opened this issue · 6 comments

No such file or directory: 'data/PascalVOC/VOCdevkit/VOC2012/ImageSets/Segmentation/train.txt
Are the documents provided incomplete?
in your file,i can not see the folder /data

Hey, thanks for opening the issue.

As I've written in the README.md the train_SmaAtUNet.py is just an example script on how to train the model using standard PyTorch. For this, I used a pretty common dataset called PascalVOC. When you have downloaded this dataset and placed it in the appropriate folder the script should work.

The training that was reported in the Paper used the train_precip_lightning.py script which, in addition to pytorch, used pytorch-lightning as well. I hope this helps!

Hey, thanks for opening the issue.

As I've written in the README.md the train_SmaAtUNet.py is just an example script on how to train the model using standard PyTorch. For this, I used a pretty common dataset called PascalVOC. When you have downloaded this dataset and placed it in the appropriate folder the script should work.

The training that was reported in the Paper used the train_precip_lightning.py script which, in addition to pytorch, used pytorch-lightning as well. I hope this helps!

Hey, thanks for opening the issue.

As I've written in the README.md the train_SmaAtUNet.py is just an example script on how to train the model using standard PyTorch. For this, I used a pretty common dataset called PascalVOC. When you have downloaded this dataset and placed it in the appropriate folder the script should work.

The training that was reported in the Paper used the train_precip_lightning.py script which, in addition to pytorch, used pytorch-lightning as well. I hope this helps!

My training process is like this,
1 create a new folder dataprecipitation and then put the downloaded dataset in this folder, then put this folder in the file you uploaded,
2 use the command
pip install -r requirements.txt
python create_datasets.py,
python train_precip_lightning.py
but there were errors in the test and train_SmaAtUNet.py running,

That sounds like the right process. What are the errors that you encountered then?

At 2022-10-01 17:27:52, "Kevin Trebing" @.> wrote: That sounds like the right process. What are the errors that you encountered then? — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.>

Because it's too long,I have explained it to you in detail in the email,

Closing, because errors were because of missing folder structure.