bart-base-finetuned-steel-news-trading

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: 0.2259
• Rouge1: 47.0722
• Rouge2: 28.9579
• Rougel: 44.3914
• Rougelsum: 44.3290
• Rouge3: 16.3522

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: 3e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 8
• 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
• lr_scheduler_warmup_steps: 500
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Rouge3
No log	1.0	86	8.7376	30.0358	13.9991	26.2396	26.2230	5.2765
11.3777	2.0	172	2.9228	43.4801	24.3953	40.0007	40.0751	11.5791
4.5138	3.0	258	0.8541	47.1045	28.3175	44.0498	43.8975	14.6814
1.5225	4.0	344	0.2663	45.3376	27.1792	42.6611	42.6164	14.5913
0.341	5.0	430	0.2390	48.3841	30.2381	45.9038	45.8778	16.7329

Framework versions

• Transformers 4.52.2
• Pytorch 2.6.0+cu124
• Datasets 2.14.4
• Tokenizers 0.21.1