calculator_model_test

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

• Loss: 0.6102

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: 0.001
• train_batch_size: 512
• eval_batch_size: 512
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 40

Training results

Training Loss	Epoch	Step	Validation Loss
3.4688	1.0	5	2.8659
2.5395	2.0	10	2.1007
1.9592	3.0	15	1.7491
1.6895	4.0	20	1.5873
1.5985	5.0	25	1.5268
1.5444	6.0	30	1.5378
1.5074	7.0	35	1.4607
1.4651	8.0	40	1.4537
1.4525	9.0	45	1.4718
1.4429	10.0	50	1.3729
1.3852	11.0	55	1.3508
1.3357	12.0	60	1.2872
1.2972	13.0	65	1.2634
1.2483	14.0	70	1.1977
1.1983	15.0	75	1.1583
1.1462	16.0	80	1.0801
1.0846	17.0	85	1.0128
1.055	18.0	90	1.0017
1.0239	19.0	95	0.9550
0.9661	20.0	100	0.9190
0.9551	21.0	105	0.8978
0.9148	22.0	110	0.8516
0.887	23.0	115	0.8381
0.8872	24.0	120	0.8330
0.8642	25.0	125	0.8257
0.8441	26.0	130	0.7729
0.8151	27.0	135	0.7500
0.7906	28.0	140	0.7392
0.7797	29.0	145	0.7292
0.7707	30.0	150	0.7043
0.7475	31.0	155	0.6987
0.74	32.0	160	0.6696
0.7274	33.0	165	0.6536
0.7121	34.0	170	0.6459
0.7113	35.0	175	0.6505
0.7016	36.0	180	0.6299
0.6946	37.0	185	0.6234
0.6886	38.0	190	0.6157
0.6821	39.0	195	0.6146
0.6768	40.0	200	0.6102

Framework versions

• Transformers 4.38.1
• Pytorch 2.1.0+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2