ft-t5-small-nl-2-fol-v1

This model is a fine-tuned version of t5-small on the yuan-yang/MALLS-v0, alevkov95/text2log dataset. It achieves the following results on the evaluation set:

• Loss: 1.0732
• Top-1 accuracy: 0.0
• Bleu Score: 0.3056
• Rouge1: 0.5254
• Rouge2: 0.2795
• Rougel: 0.5082
• Rougelsum: 0.5083
• Exact Match: 0.0

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: 32
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Top-1 accuracy	Bleu Score	Rouge1	Rouge2	Rougel	Rougelsum	Exact Match
1.6921	1.0	3231	1.0767	0.0	0.3052	0.5249	0.2786	0.5076	0.5077	0.0
1.688	2.0	6462	1.0741	0.0	0.3056	0.5254	0.2795	0.5081	0.5082	0.0
1.679	3.0	9693	1.0734	0.0	0.3056	0.5254	0.2796	0.5081	0.5082	0.0
1.6846	4.0	12924	1.0733	0.0	0.3058	0.5255	0.2798	0.5083	0.5083	0.0
1.6889	5.0	16155	1.0734	0.0	0.3056	0.5253	0.2798	0.5082	0.5083	0.0
1.6725	6.0	19386	1.0733	0.0	0.3056	0.5254	0.2799	0.5084	0.5084	0.0
1.6771	7.0	22617	1.0733	0.0	0.3056	0.5254	0.2797	0.5083	0.5083	0.0
1.6843	8.0	25848	1.0734	0.0	0.3056	0.5255	0.2797	0.5084	0.5084	0.0
1.6651	9.0	29079	1.0733	0.0	0.3054	0.5252	0.2795	0.5081	0.5082	0.0
1.7005	10.0	32310	1.0732	0.0	0.3056	0.5254	0.2795	0.5082	0.5083	0.0

Framework versions

• Transformers 4.45.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.2
• Tokenizers 0.20.1