N_roberta_twitterfin_padding100model

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

• Loss: 0.8729
• Accuracy: 0.9075

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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6304	1.0	597	0.3671	0.8719
0.3345	2.0	1194	0.2625	0.9020
0.2763	3.0	1791	0.4097	0.8932
0.2033	4.0	2388	0.3877	0.9020
0.1785	5.0	2985	0.4475	0.9037
0.1018	6.0	3582	0.4814	0.9087
0.0923	7.0	4179	0.6967	0.8907
0.0676	8.0	4776	0.6585	0.8961
0.0619	9.0	5373	0.7497	0.8945
0.0436	10.0	5970	0.7177	0.9049
0.0301	11.0	6567	0.7729	0.9066
0.0292	12.0	7164	0.7846	0.9028
0.0223	13.0	7761	0.8207	0.9041
0.0242	14.0	8358	0.7909	0.9049
0.0195	15.0	8955	0.8201	0.9037
0.0171	16.0	9552	0.8230	0.9070
0.0097	17.0	10149	0.8850	0.8970
0.0084	18.0	10746	0.8350	0.9075
0.0047	19.0	11343	0.8665	0.9075
0.003	20.0	11940	0.8729	0.9075

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.13.3