distilbert_sst2_padding100model

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

• Loss: 0.8470
• Accuracy: 0.9050

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
No log	1.0	433	0.2619	0.9012
0.3643	2.0	866	0.2941	0.8973
0.1913	3.0	1299	0.3841	0.8951
0.0956	4.0	1732	0.5066	0.9006
0.0531	5.0	2165	0.7628	0.8786
0.0336	6.0	2598	0.6779	0.8940
0.0141	7.0	3031	0.7910	0.8913
0.0141	8.0	3464	0.8459	0.8885
0.0159	9.0	3897	0.6874	0.9023
0.0151	10.0	4330	0.9174	0.8819
0.0159	11.0	4763	0.7183	0.9033
0.012	12.0	5196	0.8621	0.8880
0.0063	13.0	5629	0.8841	0.8957
0.0064	14.0	6062	0.7985	0.9066
0.0064	15.0	6495	0.8537	0.9028
0.0075	16.0	6928	0.8202	0.9055
0.0042	17.0	7361	0.8655	0.9023
0.0029	18.0	7794	0.8738	0.9006
0.0034	19.0	8227	0.8642	0.9023
0.0012	20.0	8660	0.8470	0.9050

Framework versions

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