datasetsANDmodels/occupation-extraction

datasetsANDmodels/occupation_extraction

This model is a fine-tuned version of t5-large on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0260

Model description

This model extracts the cocupation's name from text.

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
5.0748	1.0	26	3.6708
1.2816	2.0	52	1.5886
1.422	3.0	78	0.7878
0.5729	4.0	104	0.4200
1.007	5.0	130	0.2706
0.2949	6.0	156	0.1751
0.2805	7.0	182	0.1193
0.1689	8.0	208	0.0948
0.1232	9.0	234	0.0717
0.0205	10.0	260	0.0656
0.1277	11.0	286	0.0600
0.0357	12.0	312	0.0550
0.0217	13.0	338	0.0469
0.0201	14.0	364	0.0377
0.0904	15.0	390	0.0320
0.0083	16.0	416	0.0289
0.1448	17.0	442	0.0284
0.2741	18.0	468	0.0276
0.0028	19.0	494	0.0261
0.015	20.0	520	0.0260

Framework versions

• Transformers 4.41.2
• Pytorch 2.2.2
• Datasets 2.19.2
• Tokenizers 0.19.1