t_5_classifier

This model is a fine-tuned version of google-t5/t5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5350
• F1: 0.7367
• Accuracy: 0.7299

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: 128
• eval_batch_size: 128
• 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	Accuracy
No log	1.0	49	0.6857	0.6233	0.4126
No log	2.0	98	0.6695	0.6567	0.5429
No log	3.0	147	0.6445	0.6898	0.6202
No log	4.0	196	0.6087	0.7053	0.6680
No log	5.0	245	0.5762	0.7122	0.6944
No log	6.0	294	0.5601	0.7180	0.7054
No log	7.0	343	0.5512	0.7281	0.7189
No log	8.0	392	0.5471	0.7303	0.7189
No log	9.0	441	0.5457	0.7311	0.7195
No log	10.0	490	0.5405	0.7315	0.7234
0.607	11.0	539	0.5386	0.7319	0.7234
0.607	12.0	588	0.5391	0.7321	0.7240
0.607	13.0	637	0.5378	0.7357	0.7286
0.607	14.0	686	0.5362	0.7368	0.7305
0.607	15.0	735	0.5352	0.7392	0.7324
0.607	16.0	784	0.5360	0.7344	0.7292
0.607	17.0	833	0.5360	0.7358	0.7292
0.607	18.0	882	0.5353	0.7359	0.7305
0.607	19.0	931	0.5351	0.7374	0.7305
0.607	20.0	980	0.5350	0.7367	0.7299

Framework versions

• Transformers 4.41.1
• Pytorch 1.13.1+cu117
• Datasets 2.19.1
• Tokenizers 0.19.1