flan-t5-small-samsum

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

• Loss: 1.6754
• Rouge1: 42.6766
• Rouge2: 18.3823
• Rougel: 35.2377
• Rougelsum: 38.9432
• Gen Len: 16.8474

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: 52
• eval_batch_size: 52
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
1.8824	0.35	100	1.7015	42.4882	18.3559	35.0843	38.8515	16.6532
1.8578	0.7	200	1.6878	42.0007	18.2614	34.9474	38.4992	16.7216
1.835	1.06	300	1.6823	42.7535	18.6418	35.3807	39.0098	16.9048
1.8144	1.41	400	1.6786	42.6326	18.4197	35.2995	38.9165	16.6618
1.8094	1.76	500	1.6754	42.6766	18.3823	35.2377	38.9432	16.8474

Framework versions

• Transformers 4.36.0
• Pytorch 2.0.0
• Datasets 2.15.0
• Tokenizers 0.15.0