Regression_albert_4

This model is a fine-tuned version of albert-base-v2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1353
• Mse: 0.1353
• Mae: 0.3311
• R2: 0.0037
• Accuracy: 0.8421

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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Mse	Mae	R2	Accuracy
No log	1.0	33	0.0644	0.0644	0.1871	0.2305	0.9459
No log	2.0	66	0.1220	0.1220	0.2936	-0.4587	0.8919
No log	3.0	99	0.0755	0.0755	0.2180	0.0979	0.9459
No log	4.0	132	0.0662	0.0662	0.1757	0.2086	0.9189
No log	5.0	165	0.0827	0.0827	0.1978	0.0121	0.8919
No log	6.0	198	0.0962	0.0962	0.2147	-0.1498	0.9189
No log	7.0	231	0.0918	0.0918	0.1867	-0.0973	0.8919
No log	8.0	264	0.0955	0.0955	0.2075	-0.1419	0.8378
No log	9.0	297	0.0950	0.0950	0.2361	-0.1358	0.8649
No log	10.0	330	0.0875	0.0875	0.1819	-0.0455	0.8108
No log	11.0	363	0.0922	0.0922	0.2030	-0.1020	0.8649
No log	12.0	396	0.0976	0.0976	0.2194	-0.1666	0.8378
No log	13.0	429	0.0872	0.0872	0.2206	-0.0416	0.8649
No log	14.0	462	0.0810	0.0810	0.1818	0.0315	0.8919
No log	15.0	495	0.0877	0.0877	0.1861	-0.0485	0.9189
0.0535	16.0	528	0.0882	0.0882	0.1963	-0.0541	0.8919
0.0535	17.0	561	0.0814	0.0814	0.1869	0.0268	0.9189
0.0535	18.0	594	0.0902	0.0902	0.1953	-0.0775	0.8649
0.0535	19.0	627	0.0934	0.0934	0.1957	-0.1169	0.8649
0.0535	20.0	660	0.0923	0.0923	0.1928	-0.1027	0.8649

Framework versions

• Transformers 4.27.1
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2