How could I upgrade the accuracy?

[autumoon]

I have trained the model by the data you gived(https://pan.baidu.com/s/1fzNiOE7elGRmIo2h6MIhZw code:l7iv), while the accuracy is always lower the the pretrained model. i have upgrade the max_epochs to 500, while i have the same problem(Lastest model updated. Epoch_acc=0.8744, Historical_best_acc=0.8793 (at epoch 435)). how should i upgrade the accuracy? i have tried to change the model and so on but i don't think i got the point, would you like to give some suggestion? 3q!

[autumoon]

with the help of stiwfmjX ，i found that it is beacuse i don't cut the data into 256256 so the result is not good, even if i tried to resize the data into 256256, while it didn't work. I realized that it was not correct to resize the data instead of cutting the data into 256256 after the trial, so thanks a lot. above all, you should cut the data(https://pan.baidu.com/s/1fzNiOE7elGRmIo2h6MIhZw code:l7iv) into 256256 before trainning to get good result. mine is "Historical_best_acc=0.9531 (at epoch 458)".

[wulei1595]

with the help of stiwfmjX ，i found that it is beacuse i don't cut the data into 256_256 so the result is not good, even if i tried to resize the data into 256_256, while it didn't work. I realized that it was not correct to resize the data instead of cutting the data into 256_256 after the trial, so thanks a lot. above all, you should cut the data(https://pan.baidu.com/s/1fzNiOE7elGRmIo2h6MIhZw code:l7iv) into 256_256 before trainning to get good result. mine is "Historical_best_acc=0.9531 (at epoch 458)".

How did you cut data(https://pan.baidu.com/s/1fzNiOE7elGRmIo2h6MIhZw code:l7iv) into 256256 before trainning？

[wulei1595]

with the help of stiwfmjX ，i found that it is beacuse i don't cut the data into 256_256 so the result is not good, even if i tried to resize the data into 256_256, while it didn't work. I realized that it was not correct to resize the data instead of cutting the data into 256_256 after the trial, so thanks a lot. above all, you should cut the data(https://pan.baidu.com/s/1fzNiOE7elGRmIo2h6MIhZw code:l7iv) into 256_256 before trainning to get good result. mine is "Historical_best_acc=0.9531 (at epoch 458)".

Can you share it？

[autumoon]

with the help of stiwfmjX ，i found that it is beacuse i don't cut the data into 256_256 so the result is not good, even if i tried to resize the data into 256_256, while it didn't work. I realized that it was not correct to resize the data instead of cutting the data into 256_256 after the trial, so thanks a lot. above all, you should cut the data(https://pan.baidu.com/s/1fzNiOE7elGRmIo2h6MIhZw code:l7iv) into 256_256 before trainning to get good result. mine is "Historical_best_acc=0.9531 (at epoch 458)".

Can you share it？

256x256 with no overlap

[wulei1595]

在 stiwfmjX 的帮助下，我发现这是因为我没有将数据剪切成 256_256，所以结果不好，即使我尝试将数据大小调整为 256_256，而它不起作用。我意识到在试用后调整数据大小而不是将数据剪切到 256_256 是不正确的，所以非常感谢。最重要的是，您应该在训练之前将数据（https://pan.baidu.com/s/1fzNiOE7elGRmIo2h6MIhZw 代码：L7IV）剪切成 256_256 以获得良好的结果。我的是“Historical_best_acc=0.9531（在纪元 458）”。

你能分享一下吗？

256x256，无重叠

Can you share the code to split the image

[autumoon]

在 stiwfmjX 的帮助下，我发现这是因为我没有将数据剪切成 256_256，所以结果不好，即使我尝试将数据大小调整为 256_256，而它不起作用。我意识到在试用后调整数据大小而不是将数据剪切到 256_256 是不正确的，所以非常感谢。最重要的是，您应该在训练之前将数据（https://pan.baidu.com/s/1fzNiOE7elGRmIo2h6MIhZw 代码：L7IV）剪切成 256_256 以获得良好的结果。我的是“Historical_best_acc=0.9531（在纪元 458）”。

你能分享一下吗？

256x256，无重叠

Can you share the code to split the image

这部分代码不是用python写的，是用c++编写的，如果你需要这个小工具，请添加我的QQ4589968

[qianday]

在stiwfmjX的帮助下，我发现这是因为我没有将数据切成 256 256 所以结果不好，即使我尝试将数据大小调整为 256 256，但它不起作用。试用后发现调整数据大小而不是把数据切成256 256是不正确的，非常感谢。 最重要的是，你应该在训练之前将数据（https://pan.baidu.com/s/1fzNiOE7elGRmIo2h6MIhZw code:l7iv）切成256 256以获得良好的结果。我的是“Historical_best_acc=0.9531（在纪元458）”。

你好，切割后和论文结果比过高怎么理解