test

This model is a fine-tuned version of facebook/mbart-large-cc25 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.1215
• Gen Len: 17.3533
• Rouge-1: 39.1861
• Rouge-2: 22.0975
• Rouge-l: 38.4014

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: 0.0001
• 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: polynomial
• lr_scheduler_warmup_steps: 1000
• num_epochs: 50
• label_smoothing_factor: 0.1

Training results

Training Loss	Epoch	Step	Validation Loss	Gen Len	Rouge-1	Rouge-2	Rouge-l
No log	1.0	642	3.3523	22.5622	28.8044	14.6813	28.1959
No log	2.0	1284	2.9887	22.7422	36.6353	19.681	35.9403
No log	3.0	1926	2.9367	20.0578	38.6433	21.0943	37.9327
No log	4.0	2568	2.9503	18.5644	38.6509	21.3031	37.8452
No log	5.0	3210	2.9366	17.1689	38.8973	21.9518	38.3012
No log	6.0	3852	2.9782	19.2489	39.5578	22.3324	38.9385
No log	7.0	4494	3.0080	19.0422	38.1388	21.5059	37.4054
2.8286	8.0	5136	3.0908	18.4667	38.7921	21.3614	38.0183
2.8286	9.0	5778	3.1191	18.2978	39.3199	22.3807	38.6943
2.8286	10.0	6420	3.1215	17.3533	39.1861	22.0975	38.4014

Framework versions

• Transformers 4.35.2
• Pytorch 2.2.1+cu121
• Datasets 2.19.1
• Tokenizers 0.15.2