jialicheng/imdb-roberta-large

roberta-large

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

• Loss: 0.5329
• Accuracy: 0.9121

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: 5e-05
• train_batch_size: 32
• eval_batch_size: 256
• 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
No log	1.0	782	0.2406	0.8990
0.3075	2.0	1564	0.2539	0.8962
0.2171	3.0	2346	0.2650	0.9031
0.1697	4.0	3128	0.3427	0.8973
0.1697	5.0	3910	0.3241	0.9031
0.1339	6.0	4692	0.4141	0.9049
0.1038	7.0	5474	0.4572	0.8946
0.0922	8.0	6256	0.4154	0.9054
0.0676	9.0	7038	0.5020	0.8982
0.0676	10.0	7820	0.5070	0.9071
0.0568	11.0	8602	0.4826	0.9067
0.0443	12.0	9384	0.5104	0.9086
0.0313	13.0	10166	0.5456	0.9088
0.0313	14.0	10948	0.4740	0.9078
0.0245	15.0	11730	0.4977	0.9071
0.0227	16.0	12512	0.5136	0.9098
0.0175	17.0	13294	0.5131	0.9108
0.0173	18.0	14076	0.5370	0.9109
0.0173	19.0	14858	0.5344	0.9126
0.0152	20.0	15640	0.5329	0.9121

Framework versions

• Transformers 4.37.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.0
• Tokenizers 0.15.2