MatSciBERT_ST_DA_100

This model is a fine-tuned version of m3rg-iitd/matscibert on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2043
• Precision: 0.9627
• Recall: 0.9693
• F1: 0.9660
• Accuracy: 0.9561

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	59	0.2685	0.9263	0.9420	0.9341	0.9213
No log	2.0	118	0.1935	0.9477	0.9573	0.9524	0.9429
No log	3.0	177	0.2043	0.9558	0.9669	0.9613	0.9506
No log	4.0	236	0.1769	0.9596	0.9701	0.9648	0.9554
No log	5.0	295	0.1789	0.9619	0.9686	0.9652	0.9561
No log	6.0	354	0.1916	0.9620	0.9683	0.9651	0.9557
No log	7.0	413	0.1955	0.9623	0.9685	0.9654	0.9559
No log	8.0	472	0.2002	0.9627	0.9713	0.9670	0.9575
0.1044	9.0	531	0.2033	0.9632	0.9698	0.9665	0.9566
0.1044	10.0	590	0.2043	0.9627	0.9693	0.9660	0.9561

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1