mt5-bambara-resumer-boost1

This model is a fine-tuned version of Fatoumataa/mt5-bambara-resumer-final on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.4747
• Rouge1: 0.5265
• Rouge2: 0.2725
• Rougel: 0.3724
• Rougelsum: 0.3723

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: 2e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 50
• num_epochs: 8
• label_smoothing_factor: 0.05

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum
44.6779	1.0	564	2.5577	0.5194	0.2639	0.3621	0.3621
44.1834	2.0	1128	2.5343	0.5229	0.2675	0.3667	0.3666
43.8472	3.0	1692	2.5124	0.524	0.2695	0.3689	0.3688
43.0695	4.0	2256	2.4924	0.5232	0.2701	0.37	0.3698
43.0494	5.0	2820	2.4903	0.5249	0.2709	0.3713	0.3712
42.8678	6.0	3384	2.4828	0.5279	0.2721	0.3723	0.372
42.8899	7.0	3948	2.4752	0.5265	0.2721	0.3719	0.3718
42.9689	8.0	4512	2.4747	0.5265	0.2725	0.3724	0.3723

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2