[Solved] why acc can not reach at 93%

[Yuquan-Lu]

acc can't reach at 93.%;this code i have run two times,i don't change any parameters.

[KuangenZhang]

acc can't reach at 93.%;this code i have run two times,i don't change any parameters.

Thanks for your message. I will check the code recently. Moreover, how much did you achieve for classification accuracy?

[grxiao]

@KuangenZhang Hi,I run the baseline,and I got the result of 92.2%

[KuangenZhang]

@KuangenZhang Hi,I run the baseline,and I got the result of 92.2% Hi, I will run the code again and check this issue. However, it is possible that the accuracy varies sometimes, and you can try to run the train.py for several times.

[grxiao]

@KuangenZhang Hi,I run the baseline,and I got the result of 92.2% Hi, I will run the code again and check this issue. However, it is possible that the accuracy varies sometimes, and you can try to run the train.py for several times. Thanks for your reply and the sharing of the code.Thanks!

[KuangenZhang]

@grxiao @Yuquan-Lu Hi, I have run the code again and shown the results in the above picture. The highest classification accuracy is 92.73%. The accuracy does vary around 92.5%. If you try to run the train.py for 3 to 4 times, you may get a satisfying training result.

[KuangenZhang]

@Yuquan-Lu @grxiao Moreover, I changed the batch_size to 16, which seems to increase the stability of the training result.

[grxiao]

@Yuquan-Lu @grxiao Moreover, I changed the batch_size to 16, which seems to increase the stability of the training result.

thanks！

[Yuquan-Lu]

@grxiao @Yuquan-Lu Hi, I have run the code again and shown the results in the above picture. The highest classification accuracy is 92.73%. The accuracy does vary around 92.5%. If you try to run the train.py for 3 to 4 times, you may get a satisfying training result.

I am very grateful for your reply, I will repeat the experiment several times.