What format of s and psx should be inputted into get_noise_indices() in a multi-label scenario

[KongMingxi]

I have 20 samples with multi-label and 5 classes, such as:
[[2, 3, 4], [1, 3, 4, 5], [1, 3, 4], [1, 2, 3, 4, 5], [2, 3, 5], [1, 2, 4], [1, 3, 4, 5], [1, 3], [1, 5], [1, 3, 4, 5], [2, 3, 4], [3, 4], [4], [1, 3, 4], [2, 3, 4, 5], [1, 4], [3, 4], [3, 5], [2, 3, 5], [2, 5]]
I inputted this label list and a probabilities matrix as psx (shape=(20,5)) into get_noise_indices().
However, the error is:
File "C:\Users\Anaconda2\envs\tf18\lib\site-packages\cleanlab\pruning.py", line 342, in get_noise_indices
multi_label=multi_label,
File "C:\Users\Anaconda2\envs\tf18\lib\site-packages\cleanlab\latent_estimation.py", line 303, in compute_confident_joint
calibrate=calibrate,
File "C:\Users\Anaconda2\envs\tf18\lib\site-packages\cleanlab\latent_estimation.py", line 216, in _compute_confident_joint_multi_label
multi_label=True,
File "C:\Users\Anaconda2\envs\tf18\lib\site-packages\cleanlab\latent_estimation.py", line 121, in calibrate_confident_joint
confident_joint.T / confident_joint.sum(axis=1) * s_counts
ValueError: operands could not be broadcast together with shapes (5,5) (6,)

Is there any wrong with my inputs?
What format of s and psx are correct in this multi-label scenario?

[cgnorthcutt]

Hi @KongMingxi. Easy fix. The labels need to start at 0. So, use labels 0, 1, 2, 3, 4 instead of 1, 2, 3, 4, 5.

[KongMingxi]

Hi @KongMingxi. Easy fix. The labels need to start at 0. So, use labels 0, 1, 2, 3, 4 instead of 1, 2, 3, 4, 5.

Hi @cgnorthcutt
Many thanks for your reply. Besides the above format, can the one-hot labels be inputted directly?
Such as numpy array:
array([[0, 0, 1, 0, 0],
[1, 1, 1, 0, 0],
[1, 1, 1, 1, 0],
[0, 1, 0, 0, 1]])

[cgnorthcutt]

Hi @KongMingxi

You can use cleanlab.util.onehot2int to convert your onehots to the appropriate format before using cleanlab. View the implementation here: https://github.com/cgnorthcutt/cleanlab/blob/master/cleanlab/util.py#L264

For example,

from cleanlab.util import onehot2int
correctly_formatted_labels = onehot2int(onehot_matrix)

[nuaayuyang]

when i am using it to mutil lable i have got this problem , how can i solve it

[cgnorthcutt]

Please share the function call and the first few lines of all inputs.

…

On Thu, Jul 9, 2020, 7:21 AM nuaayuyang ***@***.***> wrote: [image: image] <https://user-images.githubusercontent.com/67412250/87033619-1de8fa00-c219-11ea-8e95-abd1867f0e87.png> when i am using it to mutil lable i have got this problem , how can i solve it — You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub <https://github.com/cgnorthcutt/cleanlab/issues/23#issuecomment-656069595>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AGOHF7T4U6XWXXJVDNE2HSLR2WR3PANCNFSM4LDUR6SQ> .

[nuaayuyang]

hi@cgnorthcutt

then I got this output

the number(above the image) which I print is s, my psx is a nxm matrix