Edit model card

bart_recommendation_sports_equipment_english

This model is a fine-tuned version of google/pegasus-xsum on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 3.9511
  • Rouge1: 29.2063
  • Rouge2: 9.5238
  • Rougel: 28.8889
  • Rougelsum: 29.3651
  • Gen Len: 3.4762

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: 2
  • eval_batch_size: 2
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 8
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Rouge1 Rouge2 Rougel Rougelsum Gen Len
No log 0.96 12 5.1428 25.7937 4.7619 26.1905 26.1905 3.3810
No log 2.0 25 4.5991 28.4127 4.7619 28.4127 28.4127 4.2857
No log 2.96 37 4.3659 30.0000 4.7619 30.0 30.2381 3.9048
No log 4.0 50 4.2992 23.2540 4.7619 23.1746 23.3333 3.9524
No log 4.96 62 4.1730 23.2540 4.7619 23.1746 23.3333 3.5714
No log 6.0 75 4.0884 29.2063 14.2857 29.2063 29.2063 3.4762
No log 6.96 87 4.0252 25.0 4.7619 24.9206 25.2381 3.4762
No log 8.0 100 4.0019 31.5873 14.2857 31.1905 31.4286 3.5714
No log 8.96 112 3.9648 24.2857 4.7619 24.2857 24.7619 3.4762
No log 9.6 120 3.9511 29.2063 9.5238 28.8889 29.3651 3.4762

Framework versions

  • Transformers 4.38.2
  • Pytorch 2.2.1+cu121
  • Datasets 2.18.0
  • Tokenizers 0.15.2
Downloads last month
8
Safetensors
Model size
570M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for subhamiiita/bart_recommendation_sports_equipment_english

Finetuned
(28)
this model