Right after I start the training, I run into this error :
ERROR | 2025-01-16 20:42:47 | autotrain.trainers.common:wrapper:215 - train has failed due to an exception: Traceback (most recent call last):
File "/app/env/lib/python3.10/site-packages/autotrain/trainers/common.py", line 212, in wrapper
return func(*args, **kwargs)
File "/app/env/lib/python3.10/site-packages/autotrain/trainers/object_detection/__main__.py", line 203, in train
trainer.train()
File "/app/env/lib/python3.10/site-packages/transformers/trainer.py", line 2171, in train
return inner_training_loop(
File "/app/env/lib/python3.10/site-packages/transformers/trainer.py", line 2625, in _inner_training_loop
self._maybe_log_save_evaluate(tr_loss, grad_norm, model, trial, epoch, ignore_keys_for_eval, start_time)
File "/app/env/lib/python3.10/site-packages/transformers/trainer.py", line 3071, in _maybe_log_save_evaluate
metrics = self._evaluate(trial, ignore_keys_for_eval)
File "/app/env/lib/python3.10/site-packages/transformers/trainer.py", line 3025, in _evaluate
metrics = self.evaluate(ignore_keys=ignore_keys_for_eval)
File "/app/env/lib/python3.10/site-packages/transformers/trainer.py", line 4076, in evaluate
output = eval_loop(
File "/app/env/lib/python3.10/site-packages/transformers/trainer.py", line 4365, in evaluation_loop
metrics = self.compute_metrics(
File "/app/env/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
return func(*args, **kwargs)
File "/app/env/lib/python3.10/site-packages/autotrain/trainers/object_detection/utils.py", line 191, in object_detection_metrics
boxes = convert_bbox_yolo_to_pascal(boxes, image_target["orig_size"])
File "/app/env/lib/python3.10/site-packages/autotrain/trainers/object_detection/utils.py", line 146, in convert_bbox_yolo_to_pascal
height, width = image_size
ValueError: not enough values to unpack (expected 2, got 1)
ERROR | 2025-01-16 20:42:47 | autotrain.trainers.common:wrapper:216 - not enough values to unpack (expected 2, got 1)
I copied the metadata format from the documentation but it seems like this is an issue with AutoTrain not being able to get the original size of the document. I noticed that in other datasets, they define the width and height in the metadata file but even when I added that, it still did not work. @abhishek