canine_2303

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: 0.9987
• Recall: 0.9982
• F1: 0.9985
• Accuracy: 0.9999

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	244	0.0025	0.9819	0.9924	0.9871	0.9993
No log	2.0	488	0.0018	0.9855	0.9925	0.9890	0.9995
0.0382	3.0	732	0.0014	0.9923	0.9891	0.9907	0.9996
0.0382	4.0	976	0.0009	0.9930	0.9931	0.9931	0.9997
0.0017	5.0	1220	0.0009	0.9922	0.9949	0.9936	0.9997
0.0017	6.0	1464	0.0007	0.9940	0.9952	0.9946	0.9998
0.0012	7.0	1708	0.0005	0.9947	0.9952	0.9949	0.9998
0.0012	8.0	1952	0.0005	0.9947	0.9955	0.9951	0.9998
0.0009	9.0	2196	0.0003	0.9959	0.9960	0.9959	0.9998
0.0009	10.0	2440	0.0003	0.9958	0.9963	0.9961	0.9998
0.0007	11.0	2684	0.0003	0.9971	0.9958	0.9965	0.9999
0.0007	12.0	2928	0.0003	0.9971	0.9962	0.9967	0.9999
0.0005	13.0	3172	0.0002	0.9974	0.9967	0.9971	0.9999
0.0005	14.0	3416	0.0002	0.9980	0.9972	0.9976	0.9999
0.0004	15.0	3660	0.0002	0.9982	0.9980	0.9981	0.9999
0.0004	16.0	3904	0.0002	0.9984	0.9974	0.9979	0.9999
0.0004	17.0	4148	0.0001	0.9984	0.9975	0.9979	0.9999
0.0004	18.0	4392	0.0001	0.9988	0.9982	0.9985	0.9999
0.0003	19.0	4636	0.0001	0.9987	0.9982	0.9985	0.9999
0.0003	20.0	4880	0.0001	0.9987	0.9982	0.9985	0.9999

Framework versions

• Transformers 4.27.2
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2