bert-base-uncased-sst-2-16-13-30

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

• Loss: 0.5710
• Accuracy: 0.7188

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	1	0.6730	0.5938
No log	2.0	2	0.6718	0.625
No log	3.0	3	0.6692	0.6562
No log	4.0	4	0.6657	0.6875
No log	5.0	5	0.6616	0.6562
No log	6.0	6	0.6567	0.7188
No log	7.0	7	0.6514	0.6875
No log	8.0	8	0.6462	0.75
No log	9.0	9	0.6407	0.75
0.6558	10.0	10	0.6354	0.75
0.6558	11.0	11	0.6311	0.6562
0.6558	12.0	12	0.6277	0.625
0.6558	13.0	13	0.6244	0.5938
0.6558	14.0	14	0.6203	0.5938
0.6558	15.0	15	0.6158	0.5938
0.6558	16.0	16	0.6109	0.5938
0.6558	17.0	17	0.6066	0.5938
0.6558	18.0	18	0.6016	0.5938
0.6558	19.0	19	0.5968	0.5938
0.4973	20.0	20	0.5924	0.6562
0.4973	21.0	21	0.5882	0.6875
0.4973	22.0	22	0.5843	0.6875
0.4973	23.0	23	0.5812	0.6875
0.4973	24.0	24	0.5785	0.7188
0.4973	25.0	25	0.5762	0.7188
0.4973	26.0	26	0.5744	0.7188
0.4973	27.0	27	0.5730	0.7188
0.4973	28.0	28	0.5720	0.7188
0.4973	29.0	29	0.5713	0.7188
0.3872	30.0	30	0.5710	0.7188

Framework versions

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