flan-t5-qa

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

• Loss: 1.9501
• Rouge1: 32.6382
• Rouge2: 18.8113
• Rougel: 31.4434
• Rougelsum: 31.55
• Gen Len: 17.333

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: 5
• eval_batch_size: 5
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len
2.2112	1.0	2000	2.2398	31.2944	16.8097	29.7158	29.856	17.726
1.9424	2.0	4000	2.0614	32.2914	17.2352	30.6559	30.7999	17.264
1.8168	3.0	6000	1.9951	32.1343	17.6839	30.9136	31.0279	17.385
1.7924	4.0	8000	1.9566	32.6305	18.4737	31.4423	31.5984	17.302
1.6977	5.0	10000	1.9501	32.6382	18.8113	31.4434	31.55	17.333

Framework versions

• Transformers 4.27.4
• Pytorch 1.13.0
• Datasets 2.1.0
• Tokenizers 0.13.2