google-t5-small

This model is a fine-tuned version of google-t5/t5-small on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.9027
• Accuracy: 0.7963
• Precision: 0.7873
• Recall: 0.7963
• Precision Macro: 0.7130
• Recall Macro: 0.7178
• Macro Fpr: 0.0186
• Weighted Fpr: 0.0179
• Weighted Specificity: 0.9724
• Macro Specificity: 0.9846
• Weighted Sensitivity: 0.7963
• Macro Sensitivity: 0.7178
• F1 Micro: 0.7963
• F1 Macro: 0.7139
• F1 Weighted: 0.7913

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	Precision Macro	Recall Macro	Macro Fpr	Weighted Fpr	Weighted Specificity	Macro Specificity	Weighted Sensitivity	Macro Sensitivity	F1 Micro	F1 Macro	F1 Weighted
1.9743	1.0	643	1.2581	0.6197	0.5444	0.6197	0.2733	0.2987	0.0432	0.0420	0.9378	0.9705	0.6197	0.2987	0.6197	0.2816	0.5736
1.2712	2.0	1286	0.9250	0.7049	0.6888	0.7049	0.4124	0.4222	0.0296	0.0290	0.9631	0.9779	0.7049	0.4222	0.7049	0.3987	0.6876
0.9455	3.0	1929	0.8416	0.7312	0.7170	0.7312	0.4418	0.4789	0.0262	0.0256	0.9682	0.9800	0.7312	0.4789	0.7312	0.4515	0.7214
0.7104	4.0	2572	0.8019	0.7576	0.7395	0.7576	0.4638	0.5140	0.0232	0.0223	0.9695	0.9818	0.7576	0.5140	0.7576	0.4805	0.7460
0.642	5.0	3215	0.7784	0.7668	0.7539	0.7668	0.5402	0.5477	0.0220	0.0213	0.9703	0.9825	0.7668	0.5477	0.7668	0.5288	0.7578
0.5814	6.0	3858	0.7890	0.7800	0.7781	0.7800	0.6857	0.6053	0.0205	0.0197	0.9706	0.9834	0.7800	0.6053	0.7800	0.5979	0.7728
0.4982	7.0	4501	0.8016	0.7808	0.7758	0.7808	0.6895	0.6541	0.0202	0.0197	0.9723	0.9835	0.7808	0.6541	0.7808	0.6581	0.7762
0.4402	8.0	5144	0.8413	0.7862	0.7813	0.7862	0.6899	0.6867	0.0196	0.0191	0.9737	0.9840	0.7862	0.6867	0.7862	0.6828	0.7823
0.4405	9.0	5787	0.8244	0.7955	0.7848	0.7955	0.7088	0.7061	0.0188	0.0180	0.9719	0.9845	0.7955	0.7061	0.7955	0.7059	0.7898
0.397	10.0	6430	0.8535	0.8025	0.7928	0.8025	0.7169	0.7202	0.0179	0.0173	0.9731	0.9850	0.8025	0.7202	0.8025	0.7173	0.7972
0.3596	11.0	7073	0.8741	0.7940	0.7839	0.7940	0.7110	0.7174	0.0189	0.0182	0.9720	0.9844	0.7940	0.7174	0.7940	0.7126	0.7883
0.3343	12.0	7716	0.8837	0.7971	0.7883	0.7971	0.7123	0.7161	0.0185	0.0179	0.9730	0.9847	0.7971	0.7161	0.7971	0.7130	0.7922
0.3422	13.0	8359	0.8903	0.8002	0.7907	0.8002	0.7166	0.7201	0.0182	0.0175	0.9728	0.9849	0.8002	0.7201	0.8002	0.7168	0.7949
0.3264	14.0	9002	0.9004	0.7978	0.7890	0.7978	0.7140	0.7185	0.0184	0.0178	0.9727	0.9847	0.7978	0.7185	0.7978	0.7147	0.7929
0.3096	15.0	9645	0.9027	0.7963	0.7873	0.7963	0.7130	0.7178	0.0186	0.0179	0.9724	0.9846	0.7963	0.7178	0.7963	0.7139	0.7913

Framework versions

• Transformers 4.38.2
• Pytorch 2.1.2
• Datasets 2.1.0
• Tokenizers 0.15.2