bart-large-fleece2instructions-r1

This model is a fine-tuned version of facebook/bart-large on the pszemraj/fleece2instructions dataset. It achieves the following results on the evaluation set:

• Loss: 1.0436
• Rouge1: 64.4538
• Rouge2: 47.8829
• Rougel: 62.5085
• Rougelsum: 62.6165
• Gen Len: 14.4780

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: 6e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• distributed_type: multi-GPU
• gradient_accumulation_steps: 16
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.03
• num_epochs: 2.0

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
1.0379	1.0	362	1.0932	62.4953	46.0277	60.6748	60.7667	14.5614
0.925	2.0	724	1.0436	64.4538	47.8829	62.5085	62.6165	14.4780

Framework versions

• Transformers 4.25.0.dev0
• Pytorch 1.13.0+cu117
• Datasets 2.7.0
• Tokenizers 0.13.2