Julien Cohen-Adad

Strangely, nilearn's install seem to point to "scikit-learn": https://github.com/nilearn/nilearn/blob/8b0657bfe5338c63923323f3b2ff5c45b5a953d9/pyproject.toml#L32 So how come I'm seeing this during SCT's install: ~~~ Collecting sklearn (from nilearn->-r requirements.txt (line 20)) Downloading sklearn-0.0.post12.tar.gz (2.6 kB)...

I tried running: ~~~ SKLEARN_ALLOW_DEPRECATED_SKLEARN_PACKAGE_INSTALL=True ~~~ Before installation, but that didn't make any difference. No time to dig further.

Workaround until this issue gets resolved: remove "nilearn" from the requirements.txt. This is a dirty workaround as related functionalities won't work, but in my case i'm not using nilearn and...

Thank you @po09i , this is a much better temp fix than my suggestion to get rid of nilearn 😅

both Joshua and Mathieu are on holidays. I see more pros than cons in doing the fix you propose (pin dipy to 1.5.x) so that your project can move forward...

This is very nice! I'm wondering how we should standardize our color mapping across functions. Eg: in other functions (eg: sct_deepseg_sc), the spinal cord appear red in the QC report....

I'm not a big fan of random colors-- using a color code for label number can help QCing more rapidly

Yeah. I _think_ ivadomed plays a big role in the slow imports. Which is partly the reason we would like to remove that dependency.

I renamed this issue because it concerns most SCT functions. Eg: it takes 15s (!) to run help on `sct_maths`: Terminal output ```console julien-macbook:~ $ time sct_maths -- Spinal Cord...

ah! super interesting and encouraging, given that `ivadomed` is on the way out, and `wandb`... well... there is no reason to have `wandb` here 😅 (it's only used for monitoring...