t5_base_autotagging

This model is a fine-tuned version of google-t5/t5-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5004

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss
0.6489	1.0	1250	0.5840
0.5754	2.0	2500	0.5296
0.5182	3.0	3750	0.5059
0.4823	4.0	5000	0.4930
0.4643	5.0	6250	0.4826
0.4418	6.0	7500	0.4763
0.4379	7.0	8750	0.4739
0.4106	8.0	10000	0.4728
0.4045	9.0	11250	0.4729
0.3846	10.0	12500	0.4727
0.3825	11.0	13750	0.4719
0.3747	12.0	15000	0.4734
0.3621	13.0	16250	0.4744
0.3524	14.0	17500	0.4770
0.3446	15.0	18750	0.4785
0.344	16.0	20000	0.4811
0.3379	17.0	21250	0.4836
0.3342	18.0	22500	0.4838
0.3294	19.0	23750	0.4866
0.3159	20.0	25000	0.4867
0.3171	21.0	26250	0.4899
0.312	22.0	27500	0.4925
0.3007	23.0	28750	0.4943
0.3114	24.0	30000	0.4962
0.295	25.0	31250	0.4978
0.2956	26.0	32500	0.4981
0.289	27.0	33750	0.4981
0.2934	28.0	35000	0.4992
0.2932	29.0	36250	0.5006
0.2941	30.0	37500	0.5004

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0

Evaluation Metrics

The model was evaluated on the auto-tagging task using the following metrics:

Metric	Score
ROUGE-1	0.6923
ROUGE-2	0.3731
ROUGE-L	0.6226
BLEU	0.2578