README.md

metadata

library_name: transformers
license: apache-2.0
base_model: facebook/bart-base
tags:
  - generated_from_trainer
metrics:
  - rouge
  - bleu
model-index:
  - name: lc-to-event-BART
    results: []

lc-to-event-BART

This model is a fine-tuned version of facebook/bart-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2962
• Rouge1: 0.0905
• Rouge2: 0.0195
• Rougel: 0.0903
• Rougelsum: 0.0904
• Bleu: 0.0
• Gen Len: 7.2644

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: 128
• eval_batch_size: 128
• seed: 42
• 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: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Bleu	Gen Len
0.327	0.3618	1000	0.3132	0.0821	0.0112	0.0820	0.0820	0.0	7.0490
0.3177	0.7236	2000	0.3049	0.0833	0.0137	0.0833	0.0833	0.0	7.0730
0.2963	1.0854	3000	0.3009	0.0916	0.0154	0.0915	0.0914	0.0	6.9973
0.2943	1.4472	4000	0.2975	0.0811	0.0159	0.0811	0.0811	0.0	7.2202
0.2933	1.8090	5000	0.2952	0.0880	0.0152	0.0880	0.0880	0.0	6.9883
0.2765	2.1708	6000	0.2956	0.0871	0.0180	0.0870	0.0871	0.0	7.1586
0.2761	2.5326	7000	0.2951	0.0880	0.0167	0.0879	0.0880	0.0	7.1181
0.2743	2.8944	8000	0.2928	0.0921	0.0181	0.0920	0.0921	0.0	7.0337
0.2599	3.2562	9000	0.2960	0.0900	0.0192	0.0900	0.0899	0.0	7.2215
0.2612	3.6179	10000	0.2944	0.0907	0.0189	0.0905	0.0906	0.0	7.2409
0.261	3.9797	11000	0.2940	0.0881	0.0203	0.0878	0.0878	0.0	7.2495
0.2487	4.3415	12000	0.2967	0.0881	0.0194	0.0878	0.0879	0.0	7.2486
0.2484	4.7033	13000	0.2962	0.0905	0.0195	0.0903	0.0904	0.0	7.2644

Framework versions

• Transformers 4.52.4
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.2