I am trying to train facebook/bart-large-mnli with our dataset. I am getting this error when autotrain runs:
ValueError: could not broadcast input array from shape (2,2) into shape (2,)
I was trying this out, so I used a small dataset. The training dataset has about 65 rows, with each class having between 4-6 rows.
I am using default set of training params for this model:
{
"lr": 0.00005,
"epochs": 3,
"max_seq_length": 128,
"batch_size": 8,
"warmup_ratio": 0.1,
"gradient_accumulation": 1,
"optimizer": "adamw_torch",
"scheduler": "linear",
"weight_decay": 0,
"max_grad_norm": 1,
"seed": 42,
"logging_steps": -1,
"auto_find_batch_size": false,
"mixed_precision": "fp16",
"save_total_limit": 1,
"save_strategy": "epoch",
"evaluation_strategy": "epoch"
}