Unable to upgrade to Bokeh 2.3.0

Hello Bokeh community.

I’m trying to upgrade from Bokeh 2.0.2 to 2.3.0 in Anaconda.
I can’t make the upgrade either from the Anaconda Navigator or from the Anaconda prompt.

I will appreciate any guidance on how can I do make the upgrade. I couldn’t find a discussion about this.

In case it matters, I’m using Windows 10.

Thank you!

I can’t make the upgrade either from the Anaconda Navigator or from the Anaconda prompt.

@ncachanosky You need to provide more details. What did you try (be precise) and what was the results (also be precise)? Typically it’s no more than running conda install bokeh

Thank you Bryan,

I tried:

  1. conda install bokeh
  2. conda update bokeh
  3. connda install bokeh=2.3.0
  4. conda update conda
  5. conda update --all

No matter what I try, Bokeh’s version doesn’t update. The message I get seems to indicate there are no updates available.

(base) C:\>conda install bokeh
Collecting package metadata (current_repodata.json): done
Solving environment: done

# All requested packages already installed.

(base) C:\>bokeh info
Python version      :  3.8.8 (default, Feb 24 2021, 15:54:32) [MSC v.1928 64 bit (AMD64)]
IPython version     :  7.22.0
Tornado version     :  6.1
Bokeh version       :  2.0.2
BokehJS static path :  C:\Users\ncach\anaconda3\lib\site-packages\bokeh\server\static
node.js version     :  v14.8.0

@ncachanosky We don’t have any control over the Anaconda defaults channel, or over issues with conda. The only offhand suggestions I can think of:

  • Try installing from conda-forge instead of Anaconda defauls

    conda install -c conda-forge bokeh=2.3
  • Try creating and using a new (non-root) conda env:

    conda create -n <envname> python=3.9 bokeh=2.3 ...
    conda activate <envname>
  • Install with pip instead of conda (it’s fine to pip install things into conda environments)

  • Make sure that there is not a system PATH issue or mixup

Otherwise, I am afraid this is not really a Bokeh question, per se, and should probably be directed at a support forum for Anaconda instead of here.

1 Like

Thank you Bryan,

I undestand. I appreciate your feedback.

Best wishes!