metadata

license: mit
base_model: roberta-base
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: roberta-base-sst-2-64-13-30
    results: []

roberta-base-sst-2-64-13-30

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

Loss: 0.6400
Accuracy: 0.8984

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	4	0.6936	0.5
No log	2.0	8	0.6928	0.5156
0.6938	3.0	12	0.6921	0.6328
0.6938	4.0	16	0.6911	0.6328
0.6895	5.0	20	0.6894	0.5859
0.6895	6.0	24	0.6866	0.625
0.6895	7.0	28	0.6818	0.6641
0.6758	8.0	32	0.6727	0.6953
0.6758	9.0	36	0.6495	0.7656
0.615	10.0	40	0.5773	0.8125
0.615	11.0	44	0.4229	0.875
0.615	12.0	48	0.3311	0.8906
0.3514	13.0	52	0.3047	0.8906
0.3514	14.0	56	0.3420	0.8828
0.0929	15.0	60	0.4113	0.8906
0.0929	16.0	64	0.4550	0.8906
0.0929	17.0	68	0.5299	0.8906
0.0206	18.0	72	0.6554	0.8594
0.0206	19.0	76	0.7213	0.8594
0.007	20.0	80	0.7860	0.8516
0.007	21.0	84	0.8466	0.8438
0.007	22.0	88	0.8522	0.8516
0.0037	23.0	92	0.8023	0.8516
0.0037	24.0	96	0.6670	0.8828
0.0028	25.0	100	0.6224	0.8984
0.0028	26.0	104	0.6283	0.8906
0.0028	27.0	108	0.6333	0.8906
0.0026	28.0	112	0.6307	0.8906
0.0026	29.0	116	0.6348	0.8984
0.003	30.0	120	0.6400	0.8984

Framework versions

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