compute_columns3d not implemented for Half #8104
-
|
Hello I was trying to train a lung nodule detection model with a small dataset using the command line --------- python -m monai.bundle run --config_file configs/train.json and came across the following error log:
If I could get help to overcome this error log, it would be great. Thanking you Rajesh |
Beta Was this translation helpful? Give feedback.
Replies: 2 comments
-
|
Hi @Rajesh-ParaxialTech, based on the error message, it appears that the data dtype is incompatible with the network implementation. A possible workaround would be to add an EnsureType transform before passing the data into the network to ensure it's in float32 format. Hope it helps, thanks. |
Beta Was this translation helpful? Give feedback.
-
|
Thank you YunLiu
…On Mon, Sep 23, 2024 at 8:02 AM YunLiu ***@***.***> wrote:
Hi @Rajesh-ParaxialTech <https://github.com/Rajesh-ParaxialTech>, based
on the error message, it appears that the data dtype is incompatible with
the network implementation. A possible workaround would be to add an
EnsureType transform before passing the data into the network to ensure
it's in float32 format.
Hope it helps, thanks.
—
Reply to this email directly, view it on GitHub
<#8104 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/BF63OZUAZDIACLTKFFI46UTZX54TDAVCNFSM6AAAAABOUXXKM6VHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTANZSGE2DAMI>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
Beta Was this translation helpful? Give feedback.
Hi @Rajesh-ParaxialTech, based on the error message, it appears that the data dtype is incompatible with the network implementation. A possible workaround would be to add an EnsureType transform before passing the data into the network to ensure it's in float32 format.
Hope it helps, thanks.