led-risalah-v2

This model is a fine-tuned version of allenai/led-base-16384 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.2877
• Rouge1 Precision: 0.2686
• Rouge1 Recall: 0.3416
• Rouge1 Fmeasure: 0.2966

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: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 1
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1 Precision	Rouge1 Recall	Rouge1 Fmeasure
2.0939	0.1278	10	2.6270	0.2628	0.3474	0.2911
1.8688	0.2556	20	2.4999	0.2679	0.3473	0.2973
1.7662	0.3834	30	2.4115	0.2692	0.3391	0.2955
1.669	0.5112	40	2.3715	0.2677	0.336	0.2939
1.7041	0.6390	50	2.3470	0.2649	0.3496	0.2972
1.6741	0.7668	60	2.3200	0.2617	0.3359	0.2903
1.6621	0.8946	70	2.2998	0.2688	0.3448	0.2982

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1