Holmeister/TURNA_spell_correction_product_search_100k

TURNA_spell_correction_product_search

This model is a fine-tuned version of boun-tabi-LMG/TURNA on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0575
• Rouge1: 0.9181
• Rouge2: 0.8777
• Rougel: 0.9176
• Rougelsum: 0.9179
• Bleu: 0.9894
• Precisions: [0.961639874117547, 0.9979280745893148, 0.998756734355574, 1.0]
• Brevity Penalty: 1.0
• Length Ratio: 1.0006
• Translation Length: 11757
• Reference Length: 11750
• Meteor: 0.8549
• Score: 3.8959
• Num Edits: 449
• Ref Length: 11525.0

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: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Bleu	Precisions	Brevity Penalty	Length Ratio	Translation Length	Reference Length	Meteor	Score	Num Edits	Ref Length
No log	0.3333	469	0.0635	0.9159	0.8709	0.9161	0.9159	0.9793	[0.9593921703296703, 0.9948856799037304, 0.993993993993994, 0.9692307692307692]	1.0	1.0001	11648	11647	0.8496	4.2023	481	11446.0
No log	0.6667	938	0.0604	0.9179	0.8724	0.9181	0.9178	0.9821	[0.9607103028223385, 0.9954934655250113, 0.9931506849315068, 0.979381443298969]	1.0	1.0009	11657	11647	0.8518	4.0101	459	11446.0
0.398	1.0	1407	0.0559	0.9189	0.8734	0.9189	0.9186	0.9883	[0.9620503133854211, 0.9983451180983902, 0.9982854693527646, 0.9948453608247423]	1.0	1.0	11647	11647	0.8527	3.8791	444	11446.0
0.398	1.3333	1876	0.0572	0.9166	0.8738	0.9167	0.9165	0.9889	[0.9610233516483516, 0.9975932611311673, 0.9974271012006861, 1.0]	1.0	1.0001	11648	11647	0.8518	3.9839	456	11446.0
0.398	1.6667	2345	0.0562	0.9165	0.8743	0.9165	0.9164	0.9894	[0.9612842304060434, 0.9981952173259137, 0.9987135506003431, 1.0]	1.0	1.0002	11649	11647	0.8519	3.9490	452	11446.0
0.0414	2.0	2814	0.0563	0.9158	0.8738	0.9159	0.9158	0.9896	[0.9610200051515412, 0.9986460057168648, 0.9991419991419992, 1.0]	1.0	1.0	11647	11647	0.8517	3.9752	455	11446.0

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.1+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1