mt5-small-finetuned-cnn_dailymail

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: 1.5244
• Rouge1: 23.8806
• Rouge2: 11.7122
• Rougel: 20.1043
• Rougelsum: 22.5041
• Bleu 1: 3.5889
• Bleu 2: 2.411
• Bleu 3: 1.7466
• Meteor: 11.8919
• Lungime rezumat: 11.496
• Lungime original: 46.991

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Bleu 1	Bleu 2	Bleu 3	Meteor	Lungime rezumat	Lungime original
2.9989	1.0	3583	1.6617	21.745	9.6834	17.6	20.1315	3.1902	2.0591	1.4759	10.57	11.408	46.991
1.8552	2.0	7166	1.5640	22.5336	10.3837	18.3609	20.9449	3.2826	2.1341	1.5187	11.0138	11.3677	46.991
1.7715	3.0	10749	1.5354	23.5705	11.4281	19.7129	22.1588	3.5276	2.3649	1.7132	11.7397	11.4513	46.991
1.7385	4.0	14332	1.5244	23.8806	11.7122	20.1043	22.5041	3.5889	2.411	1.7466	11.8919	11.496	46.991

Framework versions

• Transformers 4.40.0
• Pytorch 2.2.2+cu118
• Datasets 2.19.0
• Tokenizers 0.19.1