Some mistakes in generating niid mnist data

[tjuxiaofeng]

Thanks to the author for modifying some old errors in the file two months ago, but there are still some errors that need attention.

1. Line 39: "l = (user * NUM_USERS + j) % 10" should be changed to "l = (user * NUM_LABELS + j) % 10". The former will cause data allocation errors, and all users are assigned data with the same label.
2. Line 81: The code to calculate "l" should be same in Line 39 and Line 81.
3. Line 86: "if idx[l] + num_samples < len(mnist_data[l]):" the "<" should be modified to "<=". The former will cause the last part of the data set of each label to not be correctly assigned to the user. （This problem occurs because the author modified an old error on line 87， which has changed "mnist_data[l][idx[l]:num_samples]" to "mnist_data[l][idx[l]:idx[l]+num_samples]". ）

[CharlieDinh]

Thank you for your suggestion and sorry for the late reply, I didn't get the notification for the issue. I haven't tested the new code. Could you make a pull request with your update and then I can check the correctness?

[sunbusi]

Hello, author. I run the code directly without change, but the loss value has always been 2.3. Why. I use cifar10, and every model is like this

[CapussoM]

same for me, loss stays on 2.3

[frontop137]

Hello, author. I run the code directly without change, but the loss value has always been 2.3. Why. I use cifar10, and every model is like this

Hello, I also ran into this problem, did you solve it afterwards? thx

[frontop137]

same for me, loss stays on 2.3

Hello, I also ran into this problem, did you solve it afterwards? thx

[kwunhang]

same for me, loss stays on 2.3

Hello, I also ran into this problem, did you solve it afterwards? thx

Did anyone solve this issue.....? I also stuck in this problem....

[Fruit0218]

same for me, loss stays on 2.3

Hello, I also ran into this problem, did you solve it afterwards? thx

Did anyone solve this issue.....? I also stuck in this problem....

What about u, buddy?I've been stuck in this for a long time

[kwunhang]

same for me, loss stays on 2.3

Hello, I also ran into this problem, did you solve it afterwards? thx

Did anyone solve this issue.....? I also stuck in this problem....

What about u, buddy?I've been stuck in this for a long time

I found that it just due to the model is too simple and unable to learn.

[Fruit0218]

same for me, loss stays on 2.3

Hello, I also ran into this problem, did you solve it afterwards? thx

Did anyone solve this issue.....? I also stuck in this problem....

What about u, buddy?I've been stuck in this for a long time

I found that it just due to the model is too simple and unable to learn.

u mean the model "CNNCifar" in models.py?I did think it cannot train because of the model, but I don't notice the simplicity and construction of it. Could you offer any good suggestions regarding this? Thank you very much.

[kwunhang]

same for me, loss stays on 2.3

Hello, I also ran into this problem, did you solve it afterwards? thx

Did anyone solve this issue.....? I also stuck in this problem....

What about u, buddy?I've been stuck in this for a long time

I found that it just due to the model is too simple and unable to learn.

u mean the model "CNNCifar" in models.py?I did think it cannot train because of the model, but I don't notice the simplicity and construction of it. Could you offer any good suggestions regarding this? Thank you very much.

I was working on federated learning for my final year project and also learning about it. I forked the pFedMe repo and did some simple work and tries on it. You may visit my work if you are interested. Just the commit message and code maybe a bit messy since the repo is for my personal use.

[Fruit0218]

same for me, loss stays on 2.3

Hello, I also ran into this problem, did you solve it afterwards? thx

Did anyone solve this issue.....? I also stuck in this problem....

What about u, buddy?I've been stuck in this for a long time

I found that it just due to the model is too simple and unable to learn.

u mean the model "CNNCifar" in models.py?I did think it cannot train because of the model, but I don't notice the simplicity and construction of it. Could you offer any good suggestions regarding this? Thank you very much.

I was working on federated learning for my final year project and also learning about it. I forked the pFedMe repo and did some simple work and tries on it. You may visit my work if you are interested. Just the commit message and code maybe a bit messy since the repo is for my personal use.
That would be great if I could learn from you! I'm currently studying the pFedMe paper in depth, but there are still many aspects I don't understand. I hope to have the opportunity to learn and exchange ideas with you when you have time.