md_mt5_base_boun_split_second_v1_retrain_on_second_imst_v2_1112

This model is a fine-tuned version of Buseak/md_mt5_base_boun_split_second_v1_retrain_on_second_imst on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1052
• Bleu: 2.2599
• Gen Len: 18.7745

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: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
0.2959	1.0	916	0.1507	2.123	18.7726
0.2784	2.0	1832	0.1420	2.1538	18.771
0.2741	3.0	2748	0.1376	2.1514	18.7701
0.2635	4.0	3664	0.1304	2.1802	18.7745
0.2551	5.0	4580	0.1255	2.1975	18.7775
0.2572	6.0	5496	0.1225	2.2077	18.7775
0.2504	7.0	6412	0.1188	2.2158	18.7781
0.2449	8.0	7328	0.1152	2.2287	18.7781
0.2425	9.0	8244	0.1121	2.2366	18.7748
0.2383	10.0	9160	0.1100	2.2437	18.777
0.2458	11.0	10076	0.1081	2.2543	18.7786
0.2334	12.0	10992	0.1075	2.2505	18.7786
0.2415	13.0	11908	0.1060	2.2519	18.775
0.2401	14.0	12824	0.1052	2.2579	18.7742
0.2435	15.0	13740	0.1052	2.2599	18.7745

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0