how to vc.widget() in my server #88
Comments
|
Hi Fadhl,
|
|
Thank you, @keller-mark
I also added 8002 as a tunnel port in puTTy. but still get the same. I think there is something wrong. Could you help? |
|
is it because: |
|
I apologize, I think our documentation is missing a step for bundling the JavaScript file for the widget. Can you try: |
|
thank you, npm run build BTW I installed nodeJs by |
|
@keller-mark any help is appreciated |
|
@FADHLyemen What version of |
|
@FADHLyemen I see you only have node 6. @keller-mark can correct me, but I believe you probably need node 14 |
|
See https://github.com/vitessce/vitessce-python/blob/master/.github/workflows/test.yml for how we set up an example environment in our CI workflow |
|
Thank you @ilan-gold. Yes, I would also check that you have NodeJS version 14. We should note this in the development section of the README, or point users to the JS repo's README for that step. Also, I just made this pull request #89 which allows using the vitessce.io web app in place of the Jupyter widget, in case you are interested. That would avoid the need to build the JS files locally, (but opens Vitessce in a separate tab rather than embedded in the notebook). |
BTW: sorry for this, I saw one of your team member presented at ISMB this tool: can you do the same for Viteesce, so we only need to upload our h5ad file? |
|
@FADHLyemen For Vitessce, we only accept |
That is in a PR currently #89 and will be merged eventually |
|
As for the node version, I would recommend upgrading. If you could provide that debug log file that NPM is pointing out ( |
|
I installed nodejs 14, please see below now I run it in jupyter notebook with option proxy =TRUE
However can not access did I miss any thing? |
|
This is correct. There is no web page to see at the 127.0.0.1:8000 URL, it is just the hostname of the web server where the raw data is being served from. Also, the http://127.0.0.1:8000/ URL is local to the HPC where the Jupyter server is running (rather than your local machine). I will make an issue to change the log message to make it clearer about the context of that URL for the remote/HPC use cases. |
|
you make it easy for me. |
|
@FADHLyemen Yes that is very doable. We have a notebook for doing just that, which starts with an |
|
Please see these instructions to make sure that your AWS S3 bucket permissions allow accessing the data on the web http://beta.vitessce.io/docs/data-hosting/index.html#aws-s3 |
|
Thank you, how to get the |
|
I still get some error, I appreciate your help |
|
You can replace the entire part after the equals sign with the string. like this: s3 = boto3.resource(
service_name="s3",
aws_access_key_id="my_access_key_id",
aws_secret_access_key="my_secret_access_key"
) |
|
when I run boto3.exceptions.S3UploadFailedError: An error occurred (AccessDenied) when calling the PutObject operation: Access Denied I read this [solution] and modify the bucket privacy to: it uploaded successfully. now when I run
|
|
The string contents of the "Out[7]" cell are the url you can copy, paste, share, and navigate into a web browser |
|
I just got this for long time, I do not know what is going on: |
|
Now, it gives this empty page |
|
@keller-mark I am trying his URL but getting a 403. Please correct me if I'm wrong @keller-mark but I believe you @FADHLyemen need to add |
|
Also |
|
Yes, sorry I thought it had been re-deployed. https://github.com/vitessce/vitessce/blob/keller-mark/docs/docs/docs/data-hosting.md#bucket-policy |
|
@keller-mark if I did not include s3:PutObject" in action, I got access denied. { I rerun vitessce_url is which is still not working. did I miss any thing? |
|
In your view config, you have "embeddingType": {
"A": "X_umap"
}which should be "embeddingType": {
"A": "UMAP"
}because your {
"type": "cells",
"fileType": "anndata-cells.zarr",
"url": "https://kidneyscrnaseq.s3.amazonaws.com/A/0/anndata.zarr",
"options": {
"mappings": {
"UMAP": {
"key": "obsm/X_umap",
"dims": [
0,
1
]
}
}
}
}i.e the mapping needs to match the It looks like this is a bug in our example that you followed. So replace scatterplot = vc.add_view(dataset, cm.SCATTERPLOT, mapping="X_umap")with scatterplot = vc.add_view(dataset, cm.SCATTERPLOT, mapping="UMAP")Once I fix that I am able to see your dataset - sorry about that! |
|
Wow, it works, thank you @keller-mark and @ilan-gold for your patience. |
|
@FADHLyemen Yes it is completely possible. Perhaps you can open a discussion for this new issue and we can resolve the issue there - especially if you have any data to share, I'd be happy to create an example notebook as this seems to be an increasingly common request. |
|
As for the URL length, you can use a URL shorten-er like tinyurl |
|
In the meantime I am going to close this issue @FADHLyemen but please do open a discussion post! |
Hi,
I follow steps in:
https://github.com/vitessce/vitessce-python/blob/master/docs/notebooks/widget_brain.ipynb
when I run on university hpc server using
I got
Running on http://127.0.0.1:8002and error in chrome:
This site can’t be reached127.0.0.1 refused to connect.Any suggestion how to access the vw?
Thank you
BTW:
I created the following conda env
then
jupyter labThe text was updated successfully, but these errors were encountered: