Hugues THOMAS

> Where do I place the extracted zip file? Well in a folder of your choice, that does not really matter. Ideally you want to have this folder containing the...

Hi @Yakamoko, I am sorry, but you will have to train it from scratch. I don't have these logs anymore. Best, Hugues

> Can you please clarify what you mean by "the score was obtained during validation and with a progressive voting scheme"? I am confused since there is no separate validation...

As I told other people previously, I personally use Linux to run the code. I only tried Windows once a while ago to test if it worked, so I don't...

I am sorry, I have never worked with tf2. The version of cuda should not matter for the cpp_wrappers, and as for tf2, I don't know what changed so if...

Oh, that's strange, there might be something that was modified between the training parameters and the test parameters. Could you check that? Otherwise, there might be a problem with the...

Mmmmh I wonder why you have these results... Can you go here: https://github.com/HuguesTHOMAS/KPConv-PyTorch/blob/73e444d486cd6cb56122c3dd410e51c734064cfe/test_models.py#L138-L149 And change some parameters: ```python config.augment_noise = 0.0001 config.augment_symmetries = False #config.batch_num = 3 #config.in_radius = 4...

That being said I just thought of something, on the dataset website, there is an aligned version and a normal version. Which one did you download? The height of cloud...

Oh sorry make it ```python config.augment_symmetries = [False, False, False] ``` instead

See the [scene_segmentation_guide](https://github.com/HuguesTHOMAS/KPConv-PyTorch/blob/master/doc/scene_segmentation_guide.md): you should download the file `Stanford3dDataset_v1.2.zip`