errors when running fine tuning cmd

[YupengGao]

I download your dataset and run:
python train_tasks.py --bert_model bert-base-uncased --from_pretrained ./multi_task_model.bin --config_file config/bert_base_6layer_6conect.json --tasks 1 --lr_scheduler 'warmup_linear' --train_iter_gap 4 --task_specific_tokens --save_name finetune_from_multi_task_model

it will give error:
Traceback (most recent call last):
File "train_tasks.py", line 670, in
main()
File "train_tasks.py", line 529, in main
task_losses,
File "/dccstor/extrastore/vilbert-multi-task/vilbert/task_utils.py", line 197, in ForwardModelsTrain
batch
ValueError: too many values to unpack (expected 9)

[realTaki]

I got this one in eval_retrieval.py

The error is :
Traceback (most recent call last):
File "eval_retrieval.py", line 367, in
main()
File "eval_retrieval.py", line 309, in main
task_ids=task_tokens,
ValueError: too many values to unpack (expected 9)

and task_tokens is the type of tensor([[19]], device='cuda:0')

[shivangibithel]

@KAndHisC I got the same error in eval_retrieval.py.
How did you solve this?

[realTaki]

@KAndHisC I got the same error in eval_retrieval.py.
How did you solve this?

there is a little mistake in this line

-                    _, _, vil_logit, _, _, _, _, _, _ = model(
+                    _, _, vil_logit, _, _, _, _, _, _, _ = model(

And you can also use my repo as a reference. I'm still learning and maintaining this project recently.