oop-de-qg-flan-t5-base-v4

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: 0.8388
• Rouge1: 59.6054
• Rouge2: 46.5045
• Rougel: 58.1566
• Rougelsum: 58.1981
• Gen Len: 14.5287
• Bleu: 0.3568
• Precisions: [0.6571719226856562, 0.4774637127578304, 0.3935286401399213, 0.32975460122699385]
• Brevity Penalty: 0.7943
• Length Ratio: 0.8128
• Translation Length: 2949
• Reference Length: 3628

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: 10
• eval_batch_size: 10
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 40
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 8

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len	Bleu	Precisions	Brevity Penalty	Length Ratio	Translation Length	Reference Length
No log	1.0	58	0.9821	56.9689	42.8079	55.2974	55.3949	14.6133	0.3163	[0.6255566974991436, 0.43508500772797526, 0.34603455914931325, 0.2808930425752856]	0.7844	0.8046	2919	3628
No log	1.99	116	0.9155	57.2587	43.5697	55.6919	55.8245	14.4018	0.3208	[0.629286694101509, 0.43945841392649904, 0.35226264418811004, 0.28861154446177845]	0.7834	0.8037	2916	3628
No log	2.99	174	0.8738	58.2245	44.9681	56.8015	56.9123	14.2719	0.3356	[0.6483021483021483, 0.461839530332681, 0.37634892086330934, 0.31484416270470156]	0.7733	0.7955	2886	3628
No log	4.0	233	0.8636	59.608	46.4603	58.1054	58.1991	14.4381	0.3494	[0.6566037735849056, 0.47794117647058826, 0.38925876608965826, 0.32466181061394384]	0.7830	0.8035	2915	3628
No log	5.0	291	0.8460	59.0765	45.7893	57.418	57.573	14.5196	0.3450	[0.6488601565158217, 0.46702453987730064, 0.3790074659639877, 0.31500513874614594]	0.7910	0.8101	2939	3628
No log	5.99	349	0.8427	58.3394	44.9653	56.6741	56.7472	14.6254	0.3439	[0.6412933647692826, 0.4586808188021228, 0.3740788903337668, 0.30870445344129555]	0.8009	0.8184	2969	3628
No log	6.99	407	0.8398	59.5629	46.3188	58.0534	58.1472	14.5257	0.3549	[0.65625, 0.47722923842326825, 0.39176161262050835, 0.32752434648898]	0.7927	0.8115	2944	3628
No log	7.97	464	0.8388	59.6054	46.5045	58.1566	58.1981	14.5287	0.3568	[0.6571719226856562, 0.4774637127578304, 0.3935286401399213, 0.32975460122699385]	0.7943	0.8128	2949	3628

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.1