mt5-small-finetuned-digikala-titleGen

This model is a fine-tuned version of google/mt5-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.8801
• Rouge1: 70.3489
• Rouge2: 43.245
• Rougel: 34.6608
• Rougelsum: 34.6608

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: 5.6e-05
• 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: 7

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
7.5555	1.0	847	3.2594	45.6729	19.6446	31.5974	31.5974
4.1386	2.0	1694	3.0347	58.3021	32.8172	33.9012	33.9012
3.7449	3.0	2541	2.9665	66.731	40.8991	34.2203	34.2203
3.5575	4.0	3388	2.9102	65.598	39.4081	34.5116	34.5116
3.4062	5.0	4235	2.8944	69.6081	42.8707	34.6622	34.6622
3.3408	6.0	5082	2.8888	70.2123	42.8639	34.5669	34.5669
3.3025	7.0	5929	2.8801	70.3489	43.245	34.6608	34.6608

Framework versions

• Transformers 4.26.0
• Pytorch 1.13.1+cu116
• Datasets 2.9.0
• Tokenizers 0.13.2