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: 0.0886

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
2.9371	1.0	6	2.2809
2.0329	2.0	12	1.7417
1.5558	3.0	18	1.3281
1.2296	4.0	24	1.1153
1.0568	5.0	30	0.9804
0.935	6.0	36	0.8776
0.8432	7.0	42	0.7807
0.7598	8.0	48	0.7081
0.6977	9.0	54	0.6922
0.6587	10.0	60	0.6535
0.6083	11.0	66	0.5866
0.5913	12.0	72	0.5483
0.5361	13.0	78	0.4847
0.4944	14.0	84	0.4919
0.4768	15.0	90	0.4696
0.4375	16.0	96	0.4223
0.4205	17.0	102	0.3673
0.379	18.0	108	0.3210
0.3334	19.0	114	0.2762
0.3065	20.0	120	0.2556
0.2759	21.0	126	0.2456
0.2589	22.0	132	0.2186
0.2374	23.0	138	0.2009
0.2248	24.0	144	0.1943
0.2182	25.0	150	0.1972
0.2181	26.0	156	0.1695
0.1999	27.0	162	0.1829
0.2048	28.0	168	0.1786
0.1865	29.0	174	0.1371
0.1667	30.0	180	0.1323
0.161	31.0	186	0.1301
0.1442	32.0	192	0.1162
0.1365	33.0	198	0.1111
0.1298	34.0	204	0.1054
0.1295	35.0	210	0.1001
0.1223	36.0	216	0.0974
0.1222	37.0	222	0.0939
0.1227	38.0	228	0.0921
0.1198	39.0	234	0.0895
0.1139	40.0	240	0.0886

Framework versions

• Transformers 4.39.0.dev0
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2