bengali_news_article_summarization_mt5

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

• Loss: 0.2111

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: 0.001
• train_batch_size: 20
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 160
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine_with_restarts
• lr_scheduler_warmup_steps: 100
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.99	83	0.8963
No log	2.0	167	0.3201
9.149	2.99	250	0.2583
9.149	3.99	334	0.2372
0.3009	5.0	418	0.2298
0.3009	5.99	501	0.2244
0.3009	7.0	585	0.2213
0.2524	8.0	669	0.2163
0.2524	8.99	752	0.2136
0.2306	10.0	836	0.2126
0.2306	10.99	919	0.2117
0.2176	11.99	1003	0.2120
0.2176	13.0	1087	0.2116
0.2176	13.99	1170	0.2111
0.2119	14.89	1245	0.2111

Framework versions

• Transformers 4.39.3
• Pytorch 2.1.2
• Datasets 2.18.0
• Tokenizers 0.15.2