N_distilbert_sst2_padding20model

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: 0.8979
• Accuracy: 0.9017

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	0.2748	0.8869
0.353	2.0	866	0.3545	0.8830
0.1798	3.0	1299	0.3683	0.9066
0.0854	4.0	1732	0.5158	0.8951
0.05	5.0	2165	0.6938	0.8825
0.0209	6.0	2598	0.7370	0.8880
0.0223	7.0	3031	0.6240	0.9044
0.0223	8.0	3464	0.6566	0.9105
0.0145	9.0	3897	0.7591	0.9028
0.008	10.0	4330	0.7470	0.9066
0.0089	11.0	4763	0.7930	0.9039
0.0057	12.0	5196	0.8187	0.8995
0.0036	13.0	5629	0.8465	0.9061
0.0078	14.0	6062	0.8228	0.9105
0.0078	15.0	6495	0.8533	0.9044
0.0067	16.0	6928	0.8553	0.9012
0.0016	17.0	7361	0.8933	0.9023
0.0042	18.0	7794	0.8994	0.9017
0.0	19.0	8227	0.8964	0.9028
0.0025	20.0	8660	0.8979	0.9017

Framework versions

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