search

bethgelab / siamese-mask-rcnn

Siamese Mask R-CNN model for one-shot instance segmentation
Other
346 stars 60 forks source link

Problem occurred in evaluate.ipynb #8

Closed DL-Alva closed 4 years ago

DL-Alva commented 5 years ago

When I run this part of code:

Load and evaluate model

Create model object in inference mode.

model = siamese_model.SiameseMaskRCNN(mode="inference", model_dir=MODEL_DIR, config=config) model.load_checkpoint(checkpoint, training_schedule=train_schedule)

Evaluate only active classes

active_class_idx = np.array(coco_val.ACTIVE_CLASSES) - 1

Evaluate on the validation set

print('starting evaluation ...') siamese_utils.evaluate_dataset(model, coco_val, coco_object, eval_type="bbox", dataset_type='coco', limit=0, image_ids=None, class_index=active_class_idx, verbose=1)

It reminds me ValueError: Subtract layer should be called on inputs of the same shape. However, it can run successfully when I convert the evaluate.jpynb into the file of evaluate.py.

michaelisc commented 4 years ago

I was not able to reproduce this issue and will thus close it now! Please let me know if anyone else runs into the same problem!