bart-billsum – AI Model by baek26 | AlphaNeural AI

You can deploy this model and start earning money today!

baek26/
bart-billsum

like

0

text2text-generation

generated_from_trainer

facebook/bart-base

endpoints_compatible

Views

No views yet

billsum_7999_bart-base

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: 2.4857
Rouge1: 0.1511
Rouge2: 0.0596
Rougel: 0.1229
Rougelsum: 0.1301
Gen Len: 20.0

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: 5e-05
train_batch_size: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 16
total_train_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
1.6272	1.69	500	2.4571	0.1622	0.071	0.1353	0.1414	20.0
1.3504	3.38	1000	2.4533	0.1562	0.0637	0.1272	0.1345	20.0
1.251	5.07	1500	2.4592	0.1489	0.0586	0.1215	0.1287	20.0
1.2374	6.75	2000	2.4967	0.1487	0.0588	0.1219	0.1286	20.0
1.15	8.44	2500	2.4857	0.1511	0.0596	0.1229	0.1301	20.0

Framework versions

Transformers 4.38.2
Pytorch 2.0.0+cu117
Datasets 2.18.0
Tokenizers 0.15.2