canine_deasciifier_0305

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: 0.0007
• Precision: 0.9978
• Recall: 0.9983
• F1: 0.9981
• Accuracy: 0.9998

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	0.0672	0.7433	0.8204	0.7800	0.9735
No log	2.0	488	0.0445	0.8205	0.8889	0.8533	0.9832
0.1401	3.0	732	0.0256	0.8950	0.9259	0.9102	0.9906
0.1401	4.0	976	0.0165	0.9384	0.9483	0.9433	0.9943
0.0364	5.0	1220	0.0112	0.9597	0.9629	0.9613	0.9962
0.0364	6.0	1464	0.0089	0.9656	0.9736	0.9696	0.9970
0.021	7.0	1708	0.0073	0.9710	0.9797	0.9753	0.9976
0.021	8.0	1952	0.0060	0.9740	0.9838	0.9789	0.9980
0.0139	9.0	2196	0.0038	0.9856	0.9890	0.9873	0.9988
0.0139	10.0	2440	0.0030	0.9893	0.9912	0.9903	0.9991
0.01	11.0	2684	0.0024	0.9916	0.9932	0.9924	0.9993
0.01	12.0	2928	0.0021	0.9919	0.9941	0.9930	0.9993
0.0072	13.0	3172	0.0018	0.9938	0.9957	0.9947	0.9995
0.0072	14.0	3416	0.0016	0.9940	0.9958	0.9949	0.9995
0.0056	15.0	3660	0.0012	0.9955	0.9968	0.9962	0.9996
0.0056	16.0	3904	0.0012	0.9954	0.9969	0.9962	0.9996
0.0045	17.0	4148	0.0008	0.9975	0.9979	0.9977	0.9998
0.0045	18.0	4392	0.0008	0.9975	0.9981	0.9978	0.9998
0.0039	19.0	4636	0.0008	0.9974	0.9981	0.9977	0.9998
0.0039	20.0	4880	0.0007	0.9978	0.9983	0.9981	0.9998

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.0+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3