Regression_roberta

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

• Loss: 0.1625
• Mse: 0.1625
• Mae: 0.3187
• R2: 0.9161
• Accuracy: 0.5714

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: 1e-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: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Mse	Mae	R2	Accuracy
No log	1.0	7	3.0379	3.0379	1.3888	-1.8504	0.4286
No log	2.0	14	2.4610	2.4610	1.2703	-1.3091	0.4286
No log	3.0	21	1.9135	1.9135	1.3077	-0.7954	0.0
No log	4.0	28	1.7647	1.7647	1.1897	-0.6557	0.1429
No log	5.0	35	2.2432	2.2432	1.1115	-1.1047	0.5714
No log	6.0	42	2.3279	2.3279	1.1562	-1.1842	0.5714
No log	7.0	49	1.9694	1.9694	1.0216	-0.8478	0.5714
No log	8.0	56	1.6951	1.6951	0.9216	-0.5905	0.5714
No log	9.0	63	1.5986	1.5986	0.8898	-0.4999	0.5714
No log	10.0	70	1.2021	1.2021	0.7820	-0.1279	0.5714
No log	11.0	77	1.0724	1.0724	0.8114	-0.0062	0.5714
No log	12.0	84	0.7198	0.7198	0.6954	0.3246	0.4286
No log	13.0	91	0.4436	0.4436	0.5758	0.5838	0.4286
No log	14.0	98	0.2337	0.2337	0.4422	0.7807	0.5714
No log	15.0	105	0.1429	0.1429	0.3187	0.8659	0.7143
No log	16.0	112	0.1090	0.1090	0.2643	0.8977	0.8571
No log	17.0	119	0.1228	0.1228	0.2882	0.8848	0.8571
No log	18.0	126	0.1318	0.1318	0.2713	0.8763	0.8571
No log	19.0	133	0.1270	0.1270	0.2451	0.8809	0.8571
No log	20.0	140	0.1181	0.1181	0.2174	0.8892	0.8571
No log	21.0	147	0.1441	0.1441	0.2630	0.8648	0.8571
No log	22.0	154	0.1749	0.1749	0.3027	0.8359	0.8571
No log	23.0	161	0.1465	0.1465	0.2596	0.8626	0.8571
No log	24.0	168	0.1699	0.1699	0.2918	0.8406	0.7143
No log	25.0	175	0.1877	0.1877	0.3154	0.8239	0.7143

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu116
• Datasets 2.9.0
• Tokenizers 0.13.2