bart_seq2seq_finetune – AI Model by BKVNP | AlphaNeural AI

You can deploy this model and start earning money today!

BKVNP/
bart_seq2seq_finetune

like

0

text2text-generation

generated_from_trainer

facebook/bart-base

endpoints_compatible

Views

No views yet

bart_seq2seq_finetune

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

Loss: 1.7677
Rouge1: 0.4256
Rouge2: 0.1969
Rougel: 0.2938
Rougelsum: 0.3962
Gen Len: 83.5743

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: 0.0003
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 16
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
num_epochs: 2
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
2.3916	0.5573	10000	2.0244	0.4042	0.1802	0.2791	0.3765	71.6338
2.1308	1.1145	20000	1.8904	0.4195	0.1917	0.2875	0.3895	84.0658
1.8683	1.6718	30000	1.7677	0.4256	0.1969	0.2938	0.3962	83.5743

Framework versions

Transformers 4.48.3
Pytorch 2.6.0+cu126
Datasets 3.5.0
Tokenizers 0.21.1