SOMD-train-scibert

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

• Loss: 0.0000
• F1: 1.0

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: 5e-05
• train_batch_size: 32
• eval_batch_size: 256
• 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	F1
No log	1.0	1243	0.0053	0.7304
No log	2.0	2486	0.0029	0.8024
No log	3.0	3729	0.0017	0.8700
No log	4.0	4972	0.0011	0.9302
No log	5.0	6215	0.0010	0.9305
No log	6.0	7458	0.0008	0.9545
No log	7.0	8701	0.0007	0.9629
No log	8.0	9944	0.0004	0.9698
No log	9.0	11187	0.0002	0.9854
No log	10.0	12430	0.0002	0.9871
No log	11.0	13673	0.0002	0.9934
No log	12.0	14916	0.0001	0.9898
No log	13.0	16159	0.0000	0.9985
No log	14.0	17402	0.0001	0.9940
No log	15.0	18645	0.0000	0.9986
No log	16.0	19888	0.0000	0.9966
No log	17.0	21131	0.0000	0.9994
No log	18.0	22374	0.0000	1.0
No log	19.0	23617	0.0000	1.0
No log	20.0	24860	0.0000	1.0

Framework versions

• Transformers 4.37.0
• Pytorch 2.1.2
• Datasets 2.1.0
• Tokenizers 0.15.1