flan-t5-base-extraction-cnndm_2000-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: 1.8809
• Hint Hit Num: 2.3633
• Hint Precision: 0.4277
• Num: 5.5061
• Gen Len: 18.9963

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: 32
• eval_batch_size: 100
• 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
2.2998	3.17	200	1.8793	2.1611	0.41	5.2739	18.9821
2.005	6.35	400	1.8404	2.2977	0.4239	5.4206	18.9834
1.9122	9.52	600	1.8202	2.2932	0.4219	5.4262	18.9984
1.8362	12.7	800	1.8211	2.3419	0.4265	5.4797	18.9977
1.7743	15.87	1000	1.8206	2.3377	0.4256	5.4791	18.9975
1.722	19.05	1200	1.8310	2.3149	0.4242	5.4426	18.9983
1.6762	22.22	1400	1.8378	2.3138	0.4219	5.4592	18.9981
1.6396	25.4	1600	1.8457	2.3373	0.4261	5.4672	18.9975
1.6067	28.57	1800	1.8501	2.3552	0.4272	5.4914	18.9969
1.579	31.75	2000	1.8561	2.3567	0.4273	5.4933	18.9983
1.5516	34.92	2200	1.8662	2.3504	0.4262	5.4943	18.998
1.5368	38.1	2400	1.8755	2.3547	0.4264	5.4993	18.9966
1.5155	41.27	2600	1.8809	2.3633	0.4277	5.5061	18.9963
1.5123	44.44	2800	1.8801	2.3665	0.4271	5.5183	18.9972
1.5007	47.62	3000	1.8840	2.3535	0.4253	5.5096	18.9969

Framework versions

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