Mehran Ghandehari

vnet_2d: Inputs have incompatible shapes. Received shapes (32, 32, 512) and (32, 32, 256)

2

I tested all of the models and they all worked fine, except the vnet_2d: model.vnet_2d((128, 128, 3), filter_num=[64, 128, 256, 512], n_labels=7, res_num_ini=1, res_num_max=3, activation='ReLU', output_activation="Softmax", batch_norm=True, pool=True, unpool=True, name='vnet')...

### Search before asking - [X] I have searched the YOLOv8 [issues](https://github.com/ultralytics/ultralytics/issues) and found no similar bug report. ### YOLOv8 Component Training ### Bug The model starts training all files...

dp vs ddp training

1

### Search before asking - [X] I have searched the YOLOv8 [issues](https://github.com/ultralytics/ultralytics/issues) and [discussions](https://github.com/ultralytics/ultralytics/discussions) and found no similar questions. ### Question Is it possible to switch between dp and ddp...

### Search before asking - [X] I have searched the YOLOv8 [issues](https://github.com/ultralytics/ultralytics/issues) and found no similar bug report. ### YOLOv8 Component Training ### Bug I wanted to report an issue...

adding warning when projection or resolution of items are different

1

The stac_vrt.build_vrt function determines the projection and resolution based on the first item in the dataset. This approach can sometimes lead to complications. A notable example is the case of...

Update predict.py for segmentation

2

running the segment/predict.py on an image folder gives the following error. AttributeError: 'list' object has no attribute 'shape'