mt5-small-finetuned-beer-ctg-en

This model is a fine-tuned version of google/mt5-small on the beer_reviews_label_drift_neg dataset. It achieves the following results on the evaluation set:

• Loss: 1.5562
• Rouge1: 26.1727
• Rouge2: 15.5176
• Rougel: 26.1813
• Rougelsum: 26.5943

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: 0.01
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 8

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
5.6484	1.0	115	3.7051	14.9617	2.9719	14.9901	15.104
3.7622	2.0	230	3.2168	21.9807	0.0	21.8357	21.8277
3.4237	3.0	345	2.8767	8.9642	5.2358	8.9597	9.0307
2.8204	4.0	460	2.4836	21.9807	0.0	21.8357	21.8277
2.3335	5.0	575	2.2340	8.9642	5.2358	8.9597	9.0307
2.1118	6.0	690	1.9121	8.9642	5.2358	8.9597	9.0307
1.7533	7.0	805	1.6571	25.9752	13.5161	25.8609	26.2153
1.5755	8.0	920	1.5562	26.1727	15.5176	26.1813	26.5943

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2