flan-t5-base-extraction-cnndm_20000-all-hint_precision-ep50

This model is a fine-tuned version of google/flan-t5-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.6865
• Hint Hit Num: 2.3572
• Hint Precision: 0.4273
• Num: 5.5014
• Gen Len: 18.9996

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: 2e-05
• train_batch_size: 60
• eval_batch_size: 400
• seed: 1799
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Hint Hit Num	Hint Precision	Num	Gen Len
2.1368	1.2	400	1.7751	2.203	0.4143	5.3192	18.9994
1.9563	2.4	800	1.7394	2.2359	0.4182	5.3459	18.9999
1.8966	3.59	1200	1.7167	2.2493	0.4172	5.3808	19.0
1.8669	4.79	1600	1.6975	2.2785	0.4208	5.4041	18.9999
1.8311	5.99	2000	1.6944	2.2849	0.4202	5.4243	18.9999
1.8044	7.19	2400	1.6918	2.2822	0.4214	5.4014	18.9998
1.7837	8.38	2800	1.6875	2.3155	0.425	5.4314	18.9999
1.7622	9.58	3200	1.6832	2.3069	0.423	5.4366	18.9996
1.7424	10.78	3600	1.6861	2.3173	0.4246	5.4455	18.9999
1.7318	11.98	4000	1.6856	2.3389	0.4246	5.4939	18.9999
1.7084	13.17	4400	1.6865	2.3572	0.4273	5.5014	18.9996
1.6946	14.37	4800	1.6817	2.3177	0.4219	5.4749	18.9996
1.685	15.57	5200	1.6884	2.3499	0.4262	5.4982	18.9999
1.6686	16.77	5600	1.6879	2.3466	0.4242	5.5094	18.9995
1.6583	17.96	6000	1.6885	2.3547	0.4255	5.5123	18.9993
1.6462	19.16	6400	1.6874	2.3582	0.4259	5.5165	18.9993

Framework versions

• Transformers 4.18.0
• Pytorch 1.10.0+cu111
• Datasets 2.5.1
• Tokenizers 0.12.1