ACTS_feedback1

This model is a fine-tuned version of roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2357
• Accuracy: 0.8936
• Balanced accuracy: 0.8897
• Precision: 0.8951
• Recall: 0.8936
• F1: 0.8915

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Balanced accuracy	Precision	Recall	F1
1.0881	1.0	12	1.0513	0.5532	0.5119	0.4004	0.5532	0.4645
0.9933	2.0	24	0.9257	0.5319	0.4952	0.3852	0.5319	0.4463
0.8065	3.0	36	0.7059	0.7234	0.7295	0.7607	0.7234	0.7184
0.5504	4.0	48	0.4259	0.8511	0.8474	0.8486	0.8511	0.8472
0.3262	5.0	60	0.3703	0.8511	0.8654	0.8624	0.8511	0.8499
0.1877	6.0	72	0.2518	0.8723	0.8731	0.8719	0.8723	0.8703
0.1094	7.0	84	0.2283	0.9362	0.9410	0.9415	0.9362	0.9365
0.0721	8.0	96	0.2246	0.9149	0.9244	0.9233	0.9149	0.9149
0.0521	9.0	108	0.2215	0.8936	0.8897	0.8951	0.8936	0.8915
0.0455	10.0	120	0.2357	0.8936	0.8897	0.8951	0.8936	0.8915

Framework versions

• Transformers 4.18.0
• Pytorch 1.10.0+cu111
• Datasets 2.0.0
• Tokenizers 0.11.6