train_function error while running train.py

[Pranjal-bisht]

Hi @ManishSahu53 Sir, I am trying to setup the project, and followed all the necessary steps while setting up mentioned, integrated tensorflow GPU. but when I am trying to run

python train.py --pretrained model.h5 --weight False

I am getting the following error.

Can you please help me resolve this?

[Pranjal-bisht]

when I am trying to run these 2 commands , they are generating empty folders in the path_output

1. python generateMultiRes.py

If tiling is already done, then skip this step.

2. python generateDataset.py

As this listResolution is empty , so the for loop inside the main for loop is not performing any operation.

I tried printing the current resolution of every image and found that the getRes function is giving the resolution of every image to be 1.0 , thats why the listResolution which was supposed to take some range but it remained empty as listResolution = range(1,1)

I am not able to understand what is wrong, either the data I have downloaded is of poor quality or there is some other issue maybe ?

The data I am testing on is the Massachusetts building dataset on kaggle Can you suggest possible solution for this?

[ManishSahu53]

@Pranjal-bisht Do you want to Train the model or Test it? If only Testing then you can use the test.py directly.

python test.py [-h] [--data] [--skipGridding] [--pretrained] [--weight] [--output]

[Pranjal-bisht]

I can do testing, But for that, I would require the requirements.txt file and the python version that you used for this project.

[ManishSahu53]

I used python 3.6, Tensorflow 2. I will try to add requirements.txt later

[dante3112]

requirements.txt

U can use this, but I am facing the same issue wherein the output folder is generating no file