Xmm/background-summaries-flan-t5-xl

background-summaries-flan-t5-large

This model is a fine-tuned version of google/flan-t5-xl on the hf_dataset_script dataset. It achieves the following results on the evaluation set:

• Loss: 2.1489
• Rouge1: 43.0
• Rouge2: 20.2
• Rougel: 28.9
• Rougelsum: 39.5

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: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 2
• total_train_batch_size: 16
• total_eval_batch_size: 16
• 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
No log	1.0	45	1.7449	37.9	17.2	25.4	34.5
No log	2.0	90	1.7964	40.8	19.0	27.5	37.3
No log	3.0	135	1.8705	39.5	18.2	26.7	36.1
No log	4.0	180	1.9253	40.1	18.7	27.0	36.6
No log	5.0	225	1.9471	41.8	19.6	28.0	38.4
No log	6.0	270	2.0004	42.5	20.0	28.5	39.0
No log	7.0	315	1.9927	43.2	20.6	29.1	39.7
No log	8.0	360	2.0119	42.6	20.4	28.8	39.1
No log	9.0	405	2.0653	42.7	20.3	28.7	39.1
No log	10.0	450	2.1489	43.0	20.2	28.9	39.5

Framework versions

• Transformers 4.27.4
• Pytorch 2.0.0+cu118
• Datasets 2.11.0
• Tokenizers 0.13.3