TypeError: 'JpegImageFile' object is not iterable

[lmx760581375]

2025-01-26 16:24:34.358 | ERROR | main:cli_evaluate:349 - Error during evaluation: 'JpegImageFile' object is not iterable. Please set --verbosity=DEBUG to get more information.
Token indices sequence length is longer than the specified maximum sequence length for this model (2699 > 2048). Running this sequence through the model will result in indexing errors
Traceback (most recent call last):
File "lmms-eval/lmms_eval/main.py", line 330, in cli_evaluate
results, samples = cli_evaluate_single(args)
File "/lmms-eval/lmms_eval/main.py", line 471, in cli_evaluate_single
results = evaluator.simple_evaluate(
File "lmms-eval/lmms_eval/utils.py", line 533, in _wrapper
return fn(*args, **kwargs)
File "lmms-eval/lmms_eval/evaluator.py", line 243, in simple_evaluate
results = evaluate(
File "/lmms-eval/lmms_eval/utils.py", line 533, in _wrapper
return fn(*args, **kwargs)
File "/lmms-eval/lmms_eval/evaluator.py", line 453, in evaluate
resps = getattr(lm, reqtype)(cloned_reqs) # Choiszt run generate until
File "/lmms-eval/lmms_eval/models/llava.py", line 314, in generate_until
batched_visuals = [doc_to_visual0 for ids in doc_id] # [B, N]
File "lmms-eval/lmms_eval/models/llava.py", line 314, in
batched_visuals = [doc_to_visual0 for ids in doc_id] # [B, N]
File "/lmms-eval/lmms_eval/api/task.py", line 1347, in doc_to_visual
else self.config.doc_to_visual(
File "lmms-eval/lmms_eval/tasks/mmt/utils.py", line 53, in mmt_doc_to_visual
return [image.convert("RGB") for image in doc["image"]]
TypeError: 'JpegImageFile' object is not iterable
2025-01-26 16:28:41.128 | ERROR | main:cli_evaluate:349 - Error during evaluation: 'JpegImageFile' object is not iterable. Please set --verbosity=DEBUG to get more information.
Token indices sequence length is longer than the specified maximum sequence length for this model (2681 > 2048). Running this sequence through the model will result in indexing errors
Traceback (most recent call last):
File "lmms-eval/lmms_eval/main.py", line 330, in cli_evaluate
results, samples = cli_evaluate_single(args)
File "lmms-eval/lmms_eval/main.py", line 471, in cli_evaluate_single
results = evaluator.simple_evaluate(
File "lmms-eval/lmms_eval/utils.py", line 533, in _wrapper
return fn(*args, **kwargs)
File "lmms-eval/lmms_eval/evaluator.py", line 243, in simple_evaluate
results = evaluate(
File "lmms-eval/lmms_eval/utils.py", line 533, in _wrapper
return fn(*args, **kwargs)
File "lmms-eval/lmms_eval/evaluator.py", line 453, in evaluate
resps = getattr(lm, reqtype)(cloned_reqs) # Choiszt run generate until
File "/lmms-eval/lmms_eval/models/llava.py", line 314, in generate_until
batched_visuals = [doc_to_visual0 for ids in doc_id] # [B, N]
File "/lmms-eval/lmms_eval/models/llava.py", line 314, in
batched_visuals = [doc_to_visual0 for ids in doc_id] # [B, N]
File "lmms-eval/lmms_eval/api/task.py", line 1347, in doc_to_visual
else self.config.doc_to_visual(
File "lmms-eval/lmms_eval/tasks/mmt/utils.py", line 53, in mmt_doc_to_visual
return [image.convert("RGB") for image in doc["image"]]
TypeError: 'JpegImageFile' object is not iterable
2025-01-26 16:33:44.349 | ERROR | main:cli_evaluate:349 - Error during evaluation: 'JpegImageFile' object is not iterable. Please set --verbosity=DEBUG to get more information.

[BlackflashJKL]

I meet the same problem.