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: 0.0909

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
2.928	1.0	6	2.2384
1.994	2.0	12	1.7009
1.5069	3.0	18	1.2937
1.1935	4.0	24	1.1334
1.0548	5.0	30	0.9714
0.9261	6.0	36	0.8736
0.8355	7.0	42	0.7772
0.7459	8.0	48	0.7135
0.6821	9.0	54	0.6394
0.6268	10.0	60	0.5985
0.5906	11.0	66	0.5664
0.5521	12.0	72	0.5647
0.5535	13.0	78	0.5673
0.5531	14.0	84	0.4928
0.4741	15.0	90	0.4800
0.4655	16.0	96	0.4641
0.4527	17.0	102	0.4285
0.4072	18.0	108	0.3956
0.39	19.0	114	0.3738
0.3573	20.0	120	0.3478
0.3423	21.0	126	0.3087
0.3111	22.0	132	0.2840
0.2909	23.0	138	0.2555
0.2574	24.0	144	0.2241
0.2423	25.0	150	0.2157
0.2212	26.0	156	0.1817
0.2042	27.0	162	0.1823
0.1849	28.0	168	0.1592
0.1764	29.0	174	0.1436
0.1626	30.0	180	0.1327
0.1617	31.0	186	0.1239
0.1441	32.0	192	0.1220
0.1451	33.0	198	0.1132
0.1327	34.0	204	0.1062
0.1276	35.0	210	0.1023
0.1237	36.0	216	0.1011
0.1183	37.0	222	0.0959
0.1163	38.0	228	0.0949
0.1107	39.0	234	0.0915
0.116	40.0	240	0.0909

Framework versions

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