oop-de-qg-flan-t5-base-v7

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: 0.8142
• Rouge1: 62.4362
• Rouge2: 49.6516
• Rougel: 60.4681
• Rougelsum: 60.5095
• Gen Len: 14.8550
• Bleu: 0.3895
• Precisions: [0.6839118825100133, 0.5106941838649156, 0.4258783204798629, 0.3599600599101348]
• Brevity Penalty: 0.8098
• Length Ratio: 0.8258
• Translation Length: 2996
• 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: 2
• total_train_batch_size: 20
• 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	116	0.9430	57.099	43.0882	55.3609	55.5231	14.5619	0.3190	[0.6260691070817653, 0.4359567901234568, 0.3485183547103052, 0.2854922279792746]	0.7857	0.8057	2923	3628
No log	2.0	233	0.8789	58.1933	45.0993	56.5574	56.6018	14.2840	0.3343	[0.6437931034482759, 0.45932269365511874, 0.37310098302055406, 0.30886208704771895]	0.7780	0.7993	2900	3628
No log	3.0	349	0.8464	60.5514	47.5045	59.0662	59.1036	14.4683	0.3590	[0.6698663009941721, 0.4895591647331787, 0.4017738359201774, 0.3341995841995842]	0.7837	0.8040	2917	3628
No log	4.0	466	0.8383	61.0697	48.057	59.3569	59.396	14.4894	0.3676	[0.6767537826685007, 0.5001940240589833, 0.41451469278717723, 0.3503916449086162]	0.7807	0.8015	2908	3628
0.9742	5.0	582	0.8179	61.3398	48.1544	59.4838	59.5751	14.6918	0.3696	[0.6702557200538358, 0.4926164331692541, 0.4043290043290043, 0.33804951995957555]	0.8019	0.8192	2972	3628
0.9742	6.0	699	0.8175	60.8548	47.6751	59.0342	58.9987	14.9033	0.3708	[0.6651178227680054, 0.4862043251304996, 0.3985538068906848, 0.3316831683168317]	0.8154	0.8305	3013	3628
0.9742	7.0	815	0.8163	62.9547	50.5344	61.1969	61.1641	14.7946	0.3919	[0.6915322580645161, 0.5202268431001891, 0.43301642178046673, 0.36359051941502774]	0.8033	0.8203	2976	3628
0.9742	7.97	928	0.8142	62.4362	49.6516	60.4681	60.5095	14.8550	0.3895	[0.6839118825100133, 0.5106941838649156, 0.4258783204798629, 0.3599600599101348]	0.8098	0.8258	2996	3628

Framework versions

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