oop-de-qg-flan-t5-base-v5

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: 0.8305
• Rouge1: 60.2858
• Rouge2: 47.0551
• Rougel: 58.5541
• Rougelsum: 58.5986
• Gen Len: 14.6254
• Bleu: 0.3585
• Precisions: [0.6612685560053981, 0.4800607671857197, 0.39139878366637704, 0.3257229832572298]
• Brevity Penalty: 0.7993
• Length Ratio: 0.8170
• Translation Length: 2964
• 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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• 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	0.99	72	0.9838	58.281	44.4811	56.6252	56.6047	14.6042	0.3304	[0.6428324697754749, 0.4543681747269891, 0.367666815942678, 0.30546792849631965]	0.7763	0.7980	2895	3628
No log	1.99	145	0.9010	55.8534	42.0605	54.3596	54.3148	14.6586	0.3076	[0.6021433355659745, 0.41167608286252355, 0.3253012048192771, 0.26241846462619167]	0.8065	0.8230	2986	3628
No log	3.0	218	0.8767	57.7174	44.1283	56.4402	56.3292	14.5136	0.3323	[0.6361781706902414, 0.4509578544061303, 0.36287845546292236, 0.2982546201232033]	0.7917	0.8106	2941	3628
No log	4.0	291	0.8583	60.2113	47.3135	58.8257	58.7408	14.3233	0.3580	[0.6711758584807492, 0.49490595611285265, 0.4074741107609185, 0.3412698412698413]	0.7723	0.7947	2883	3628
No log	4.99	363	0.8396	59.8588	46.8718	58.3234	58.2478	14.4894	0.3539	[0.6580469547465124, 0.47929447852760737, 0.39042599912165127, 0.32528263103802674]	0.7910	0.8101	2939	3628
No log	5.99	436	0.8316	59.7653	46.5459	58.066	58.1354	14.4804	0.3548	[0.6613342409802587, 0.4798619102416571, 0.3914762741652021, 0.3264781491002571]	0.7907	0.8098	2938	3628
0.9411	7.0	509	0.8305	60.2858	47.0551	58.5541	58.5986	14.6254	0.3585	[0.6612685560053981, 0.4800607671857197, 0.39139878366637704, 0.3257229832572298]	0.7993	0.8170	2964	3628
0.9411	7.92	576	0.8309	60.2226	47.1068	58.611	58.5902	14.6526	0.3605	[0.6590450571620713, 0.4801362088535755, 0.39273356401384085, 0.3276123170116103]	0.8026	0.8197	2974	3628

Framework versions

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