led-base-16384-finetuned-cnn_dailymail

This model is a fine-tuned version of allenai/led-base-16384 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.0670
• Rouge1: 26.5966
• Rouge2: 13.4937
• Rougel: 22.1204
• Rougelsum: 25.0057
• Bleu 1: 4.81
• Bleu 2: 3.2976
• Bleu 3: 2.4273
• Meteor: 13.4385
• Lungime rezumat: 12.5033
• Lungime original: 48.674

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: 8
• eval_batch_size: 8
• 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
1.1488	1.0	7165	1.0638	26.6263	13.2802	22.0654	25.0416	4.7464	3.1681	2.2958	13.3027	12.5003	48.674
0.9202	2.0	14330	1.0475	26.5843	13.4795	22.083	25.0206	4.7096	3.1931	2.3493	13.3342	12.4533	48.674
0.7778	3.0	21495	1.0465	26.5754	13.4585	22.0522	24.9943	4.729	3.2152	2.3491	13.3759	12.455	48.674
0.6729	4.0	28660	1.0670	26.5966	13.4937	22.1204	25.0057	4.81	3.2976	2.4273	13.4385	12.5033	48.674

Framework versions

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