PYTORCH Prediction output

[Giogi92]

Hi, I've been making a prediction using the following command "python train.py --task=test"

below is a partial output:

Optimization information Optimal -234.7307451974677
wall 17 [0, 2] [[200, 9], [199, 51]] dict_keys([]) dict_keys([55, 56, 57, 58, 59, 60])
wall 23 [1, 3] [[115, 85], [115, 107]] dict_keys([55, 56, 57, 60, 142, 143, 144, 146, 150, 151, 152, 153]) dict_keys([11, 12, 14, 70, 71, 73, 147, 148, 149])
wall 24 [0, 8] [[242, 107], [243, 143]] dict_keys([]) dict_keys([19, 20, 27, 28, 30, 31, 76, 77, 79, 80, 90, 167, 168])
.........

I would like to understand the output above, I need a file like the representation prediction represented below:

81  162 81  188 door    1   1   
81  67  81  100 door    1   1   
15  228.666666667   77  228.666666667   door    1   1   
19  41.5    33  41.5    door    1   1   
87  41  116 41  door    1   1   
97  228.666666667   148 228.666666667   door    1   1   
47  387 115 387 door    1   1   
49  104.5   75  104.5   door    1   1   
52  41  71  41  door    1   1   

Can you help me, please? Thanks, Giovanni.

[weavermonkey]

@Giogi92 - can you please share how you obtained the Pytorch pretrained model?

[Giogi92]

hi, I don't remember how but some time ago I managed to download it from the Google drive link. you can download it from here: https://mega.nz/file/ux1jDD6R#IdrjIb9R1LR4PTrfn9pWVKuWIdA0wJCGW5n_-5F3H1U

[weavermonkey]

hi, I don't remember how but some time ago I managed to download it from the Google drive link. you can download it from here: https://mega.nz/file/ux1jDD6R#IdrjIb9R1LR4PTrfn9pWVKuWIdA0wJCGW5n_-5F3H1U

Thanks a ton, very helpful!