Table 2 Loss and accuracy under different learning rates
From: Motor Fault Diagnosis Based on Short-time Fourier Transform and Convolutional Neural Network
Learning rate | Train-loss | Test-loss | Train-acc | Test-acc |
|---|---|---|---|---|
0.00001 | 1.3525 | 1.3109 | 0.5709 | 0.5562 |
0.00005 | 0.0625 | 0.0442 | 0.98 | 0.9962 |
0.0001 | 0.0137 | 0.0107 | 0.9996 | 0.9998 |
0.0002 | 0.0031 | 0.0033 | 0.9999 | 1 |
0.0003 | 0.0018 | 0.0016 | 1 | 1 |
0.0004 | 0.0012 | 0.0012 | 1 | 1 |
0.0005 | 0.00091 | 0.00056 | 1 | 1 |
0.0006 | 0.0015 | 0.00045 | 0.9995 | 1 |
0.0007 | 0.00065 | 0.0004 | 1 | 1 |
0.0008 | 0.00038 | 0.00035 | 1 | 1 |
0.0009 | 0.0003 | 0.00022 | 1 | 1 |
0.001 | 0.00026 | 0.00028 | 1 | 1 |
0.0015 | 0.00018 | 0.00018 | 1 | 1 |
0.005 | 0.000048 | 0.00018 | 1 | 1 |
0.007 | 1.6101 | 1.6097 | 0.2042 | 0.1994 |
0.009 | 1.6109 | 1.611 | 0.2002 | 0.1962 |
0.01 | 1.6101 | 1.6114 | 0.2072 | 0.2 |