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: 0.0047

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
2.7292	1.0	6	2.1173
1.7871	2.0	12	1.3767
1.1487	3.0	18	0.9345
0.8396	4.0	24	0.8010
0.741	5.0	30	0.6938
0.617	6.0	36	0.6118
0.5495	7.0	42	0.5396
0.4991	8.0	48	0.4896
0.4526	9.0	54	0.4333
0.3899	10.0	60	0.3730
0.331	11.0	66	0.3093
0.2617	12.0	72	0.2348
0.2112	13.0	78	0.1729
0.1642	14.0	84	0.1352
0.1285	15.0	90	0.1043
0.1063	16.0	96	0.0811
0.0806	17.0	102	0.0593
0.0617	18.0	108	0.0429
0.0519	19.0	114	0.0305
0.0372	20.0	120	0.0230
0.0323	21.0	126	0.0191
0.0255	22.0	132	0.0152
0.0219	23.0	138	0.0134
0.0202	24.0	144	0.0118
0.0161	25.0	150	0.0105
0.0154	26.0	156	0.0087
0.0126	27.0	162	0.0077
0.0115	28.0	168	0.0069
0.0104	29.0	174	0.0064
0.0121	30.0	180	0.0061
0.0113	31.0	186	0.0062
0.0105	32.0	192	0.0059
0.0103	33.0	198	0.0057
0.0097	34.0	204	0.0052
0.0088	35.0	210	0.0052
0.0083	36.0	216	0.0050
0.0086	37.0	222	0.0048
0.0075	38.0	228	0.0048
0.0075	39.0	234	0.0047
0.0073	40.0	240	0.0047

Framework versions

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