Model save

a5aad1d verified 9 months ago

2.83 kB

library_name: transformers
tags:
  - generated_from_trainer
model-index:
  - name: patchtst-tsmixup
    results: []

patchtst-tsmixup

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

Model description

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 0.0001
train_batch_size: 256
eval_batch_size: 512
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 512
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 1000
num_epochs: 5

Training Loss	Epoch	Step	Validation Loss	Mse	Mae	Rmse	Smape
0.4699	0.3170	100	0.3986	1.9127	0.5988	1.3830	85.7261
0.3548	0.6339	200	0.2949	1.4500	0.4838	1.2042	106.9687
0.2584	0.9509	300	0.2033	1.0556	0.3760	1.0274	110.8115
0.2056	1.2662	400	0.1819	0.9678	0.3455	0.9838	98.2755
0.1933	1.5832	500	0.1795	0.9547	0.3400	0.9771	82.1879
0.1904	1.9002	600	0.1797	0.9553	0.3421	0.9774	80.7883
0.1865	2.2155	700	0.1757	0.9618	0.3338	0.9807	94.2195
0.1872	2.5325	800	0.1749	0.9400	0.3335	0.9695	202.7448
0.1861	2.8494	900	0.1737	0.9284	0.3323	0.9635	134.4096
0.1838	3.1648	1000	0.1740	0.9250	0.3337	0.9618	87.0757
0.1846	3.4818	1100	0.1714	0.9415	0.3278	0.9703	76.6578
0.1806	3.7987	1200	0.1703	0.9292	0.3265	0.9640	69.0125
0.1806	4.1141	1300	0.1674	0.9126	0.3213	0.9553	67.0837
0.1776	4.4311	1400	0.1691	0.9056	0.3268	0.9516	55.4593
0.177	4.7480	1500	0.1667	0.9109	0.3200	0.9544	66.8504