hansbambel / smaat-unet Goto Github PK

Currently, the loss and the other metrics are calculated using separate scripts. They can be combined.

This could be helpful: https://torchmetrics.readthedocs.io/en/stable/pages/implement.html

The persistence model is just manually selecting the last input. It requires a separate function to be calculated. It would be better if it is a proper model that can be called just like the other models in the testing scripts.

After searching for debug methods online,it tells me I should use "from pytorch_lightning.loggers import LearningRateLogger
",cause the LearningRateLogger model is not in pytorch_lightning.callbacks, but after changing the way of import,still give me the ImportError.
My version of pytorch_lightning is :
lightning-utilities 0.10.0
pytorch-lightning 2.1.2
Any wrong? or Are there any other models to replace LearningRateLogger？

I only have this dataset: RAD_NL25_RAC_5min_train_test_2016-2019.h5.
This dataset is placed inside the folder ：SmaAt-UNet-master\data\precipitation.
when I ran a file：train_SmaAtUNet.py, the following error occurs:

Traceback (most recent call last):
File "D:\SmaAt-UNet-master\train_SmaAtUNet.py", line 127, in
voc_dataset_train = dataset_VOC.VOCSegmentation(dataset_folder, image_set='train',
File "D:\SmaAt-UNet-master\utils\dataset_VOC.py", line 91, in init
with open(os.path.join(split_f), "r") as f:
FileNotFoundError: [Errno 2] No such file or directory: 'data/PascalVOC\VOCdevkit\VOC2012\ImageSets/Segmentation\train.txt'

Can someone help me answer this question, thanks!

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

Using the Path module should make loading and saving the models/checkpoints more where you expect them to be.

Otherwise, running the script from another location than the main folder could result in saving/loading from the run-location instead of the root-location of the repository.

After training UNet and other models separately, I changed the model_floder in the test_precip_lightning.py file as a way to change the different models, but I ended up with the same evaluation results. Is there a detail I'm not paying attention to?

Hello, can you provide the direct download links?

Hello!
Sorry if this question sounds a bit dumb in advance
I was trying to use my custom dataset compiled using a number of Tiff files, I made an utility class to help me convert it to hdf5 and then tried to use create_dataset.py to make it as close as possible to the original dataset as mine was pre-processed.
However, I find myself with numerous issues, one being the lost function requiring the target tensor to be 3D while I found mine to be 4D instead. Even after i got the error away, the results after training wasn't much better with most of the stats showing to be NaN or not fluctuating at all.
I was wondering if the problem is from my dataset or I'm just not applying it correctly.
demo.zip
I put in most of the files that i find myself modifying
(P.S I used copilot to help me understand the code so it probably contributed to me failing so hard ;-;)

Hello @HansBambel @SMehrkanoon @tstanczyk95 ,

I hope you're doing well. I'm interested in your project and would greatly appreciate your guidance on two key aspects:

Dataset Access:
I'm looking for information on how to access the project dataset. Having access to the dataset is crucial for me to work with your project effectively.

Project Setup:
I'd like to run the project on my local system. Could you provide detailed steps or a guide on how to set up the project environment? Information on software dependencies, configuration settings, and installation instructions would be incredibly helpful.

Your assistance in these matters would be invaluable, and I'm eager to explore and learn from your project.

Thank you for your time and consideration. I'm posting this as a GitHub issue to ensure clear communication and documentation for others who might have similar questions.

I'm sorry to border that I have a question about the test_precip_lightning.py
i've run the train_precip_lightning.py and it has 60 epochs in the following folder

and I run the test file it has such error, I'd like to ask how to solve it.

I'sorry to border again! I've successfully get the pictures through the ipynb file which similar to the article,
but the result of the test_precip_lightning.py is like this:

There're only one model's scores
I'd like to ask how to get all the scores in the article

I don't know if the result is set to be like the first one or is there any problem to the code.
Thank you so much!