canine_sent_0504

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

• Loss: 0.0001
• Precision: 0.8919
• Recall: 0.9429
• F1: 0.9167
• Accuracy: 1.0000

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: 32
• eval_batch_size: 32
• 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	Precision	Recall	F1	Accuracy
No log	1.0	64	0.0037	0.1082	0.8286	0.1914	0.9988
No log	2.0	128	0.0019	0.2013	0.8857	0.3280	0.9993
No log	3.0	192	0.0015	0.3077	0.9143	0.4604	0.9994
No log	4.0	256	0.0012	0.3855	0.9143	0.5424	0.9995
No log	5.0	320	0.0011	0.2645	0.9143	0.4103	0.9995
No log	6.0	384	0.0010	0.2406	0.9143	0.3810	0.9996
No log	7.0	448	0.0008	0.2336	0.9143	0.3721	0.9996
0.008	8.0	512	0.0007	0.2623	0.9143	0.4076	0.9997
0.008	9.0	576	0.0006	0.3265	0.9143	0.4812	0.9998
0.008	10.0	640	0.0005	0.3137	0.9143	0.4672	0.9998
0.008	11.0	704	0.0004	0.4521	0.9429	0.6111	0.9999
0.008	12.0	768	0.0003	0.5	0.9429	0.6535	0.9999
0.008	13.0	832	0.0003	0.66	0.9429	0.7765	0.9999
0.008	14.0	896	0.0002	0.5238	0.9429	0.6735	0.9999
0.008	15.0	960	0.0002	0.7333	0.9429	0.8250	0.9999
0.0008	16.0	1024	0.0001	0.8684	0.9429	0.9041	1.0000
0.0008	17.0	1088	0.0001	0.8462	0.9429	0.8919	1.0000
0.0008	18.0	1152	0.0001	0.8684	0.9429	0.9041	1.0000
0.0008	19.0	1216	0.0001	0.8919	0.9429	0.9167	1.0000
0.0008	20.0	1280	0.0001	0.8919	0.9429	0.9167	1.0000

Framework versions

• Transformers 4.27.4
• Pytorch 2.0.0+cu118
• Datasets 2.11.0
• Tokenizers 0.13.2