climate-percentage-regression

This model is a fine-tuned version of alex-miller/ODABert on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0489

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-06
• train_batch_size: 24
• eval_batch_size: 24
• 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
0.0982	1.0	75	0.0878
0.0608	2.0	150	0.0590
0.0473	3.0	225	0.0535
0.0409	4.0	300	0.0521
0.0377	5.0	375	0.0506
0.0365	6.0	450	0.0506
0.0332	7.0	525	0.0497
0.0323	8.0	600	0.0495
0.0302	9.0	675	0.0490
0.0295	10.0	750	0.0491
0.0276	11.0	825	0.0486
0.0263	12.0	900	0.0487
0.026	13.0	975	0.0487
0.0246	14.0	1050	0.0486
0.0239	15.0	1125	0.0484
0.0234	16.0	1200	0.0486
0.0232	17.0	1275	0.0488
0.0229	18.0	1350	0.0487
0.0227	19.0	1425	0.0489
0.0222	20.0	1500	0.0489

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu121
• Datasets 3.0.0
• Tokenizers 0.19.1