-
Notifications
You must be signed in to change notification settings - Fork 56
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
FileNotFoundError: [Errno 2] No such file or directory: 'ckpt.tar.gz.tmp' -> 'ckpt.tar.gz' #241
Comments
I encountered this error when launching multiple cellbender jobs from the same directory. My guess is that |
I also have a similar experience and second what @chris-rands says. When running multiple jobs from the same directory I get this error and when I create subdirs and run each job from a separate dir I don't run into this issue. |
@chris-rands you are quite correct! For some reason, I did not think of this. I typically work in the cloud where one machine is designated to run one job. Let me explain what the tool is currently doing:
Two options moving forward:
|
@sjfleming I think giving the option to run everything from the same directory and specifying the checkpoint name would be nice to have down the line. For the time being it might be worth warning the users not to run multiple jobs from the same directory. |
Penny @pengxin2019 , does this solve your problem as well? |
I actually misunderstood the existing My two cents: Most users probably don't care about the checkpoint file and would be ideal if the default options handle this. So, could add some unique identifier (like a hash/job id) to the checkpoint file name for each run to avoid these name collisions. Or make the default write location the same as the output directory (I'm assuming users would not launch multiple runs with the same |
@chris-rands yeah that I think you're right. I'll do a bit of thinking about the best way to handle this so that users really don't have to think about it. The reason I wanted to always call the saved checkpoint file |
I was looking for this option as well. We typically run jobs in parallel on a HPC using snakemake. @sjfleming From my point of view, it would be enough to just save the checkpoint file in a user-specified output directory and only start from a checkpoint when the |
Hi sjfleming, here are my commands for the job of sample 6204:
However the expected output file sample.6204_1.cellbender.output.h5 was not generated. Here is my output file sample.6204_1.cellbender.output.log:
This is the end of the output file:
Briefly speaking, Thanks for your patience. |
Hi Penny @pengxin2019 , Okay, the issue you're having there is #212 which is also mentioned here The problem is that you probably have pytorch version 2.0+ (and probably python > 3.7). Currently cellbender needs pytorch < 2.0 (and python 3.7), because they changed something in pytorch in v2.0 and I have not quite figured out how to make cellbender work with it. (Progress on that is being tracked here #203 ) Can you try this to install cellbender (assuming you are using
|
Thanks for the input @makrez ! |
(please update the documentation in the README.md ) |
Dear cellbender team,
Thanks for developing this great tool. I am using cellbender to remove some ambient RNA but ran into some errors that I do not know how to tackle.
I installed cellbender by issuing:
Here is my command:
Here is my output showing the error:
As you can see above, there are some problems like
Can you give me some suggestions on how to fix it?
Thanks
Best
Penny
The text was updated successfully, but these errors were encountered: