distilbert-base-uncased-finetuned-scientific-eval

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: 0.1810
• Precision: 0.5820
• Recall: 0.6719
• F1: 0.6237
• Accuracy: 0.9483

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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	71	0.3580	0.3871	0.1893	0.2542	0.9068
No log	2.0	142	0.2235	0.5585	0.4669	0.5086	0.9332
No log	3.0	213	0.1870	0.6355	0.5994	0.6169	0.9456
No log	4.0	284	0.1857	0.5915	0.6120	0.6016	0.9474
No log	5.0	355	0.1810	0.5820	0.6719	0.6237	0.9483
No log	6.0	426	0.1969	0.6108	0.6435	0.6267	0.9505
No log	7.0	497	0.1914	0.5965	0.6530	0.6235	0.9513
0.1851	8.0	568	0.1964	0.6040	0.6593	0.6305	0.9517
0.1851	9.0	639	0.2026	0.6023	0.6593	0.6295	0.9517
0.1851	10.0	710	0.2027	0.6011	0.6751	0.6360	0.9508

Framework versions

• Transformers 4.27.2
• Pytorch 2.0.1+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3