metadata

license: mit
base_model: roberta-base
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: roberta-base-sst-2-32-13-30
    results: []

roberta-base-sst-2-32-13-30

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

Loss: 0.5496
Accuracy: 0.75

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: 1.5e-05
train_batch_size: 32
eval_batch_size: 32
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 5
num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	2	0.6930	0.5156
No log	2.0	4	0.6929	0.5
No log	3.0	6	0.6928	0.5
No log	4.0	8	0.6925	0.5
0.6955	5.0	10	0.6920	0.5
0.6955	6.0	12	0.6914	0.5156
0.6955	7.0	14	0.6904	0.5469
0.6955	8.0	16	0.6891	0.5312
0.6955	9.0	18	0.6872	0.5156
0.6791	10.0	20	0.6845	0.5312
0.6791	11.0	22	0.6805	0.5312
0.6791	12.0	24	0.6751	0.5312
0.6791	13.0	26	0.6654	0.5625
0.6791	14.0	28	0.6525	0.625
0.6052	15.0	30	0.6347	0.6406
0.6052	16.0	32	0.6130	0.6719
0.6052	17.0	34	0.5903	0.6875
0.6052	18.0	36	0.5770	0.6875
0.6052	19.0	38	0.5569	0.7031
0.3501	20.0	40	0.5333	0.75
0.3501	21.0	42	0.5251	0.7344
0.3501	22.0	44	0.5137	0.75
0.3501	23.0	46	0.5118	0.7656
0.3501	24.0	48	0.5151	0.7656
0.137	25.0	50	0.5202	0.7656
0.137	26.0	52	0.5299	0.7656
0.137	27.0	54	0.5379	0.7656
0.137	28.0	56	0.5433	0.75
0.137	29.0	58	0.5477	0.75
0.0715	30.0	60	0.5496	0.75

Framework versions

Transformers 4.32.0.dev0
Pytorch 2.0.1+cu118
Datasets 2.4.0
Tokenizers 0.13.3