kchen-lab / scmer Goto Github PK

Hi, I am trying to run the example in
https://htmlpreview.github.io/?https://github.com/KChen-lab/SCMER/blob/master/notebooks/melanoma-gpu-with-batch-in-r.nb.html

I wonder where I can get the input files.

'melanoma.txt.gz'
"genes.txt"
"melanoma_batches.txt"

Thank you!

Hi!
Running into next error while trying to install SCMER:

(base) C02ZC54JLVDL:Develop alsu$ git clone https://github.com/KChen-lab/SCMER.git
Cloning into 'SCMER'...
remote: Enumerating objects: 485, done.
fatal: the remote end hung up unexpectedlyiB | 2.90 MiB/s
fatal: early EOF
fatal: index-pack failed

Hi! I am running SCMER on a single-cell dataset with 30000+ cells and 8000 genes with the function

model = UmapL1(lasso=0.005, n_threads=16, use_gpu=True).fit(adata.X)

Below is the error I receive. I wonder is there any solution (like setting some parameters) to prevent the memory overflow issue? I will hugely appreciate any help! Thank you in advance!!!!
(Remark: SCMER can be run on gpu with no issue on datasets less than 10000 cells. )

OutOfMemoryError: CUDA out of memory. Tried to allocate 14.02 GiB. GPU 0 has a total capacity of 23.99 GiB of which 0 bytes is free. Of the allocated memory 30.76 GiB is allocated by PyTorch, and 10.58 GiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)
Output is truncated. View as a scrollable element or open in a text editor. Adjust cell output settings...

Hi,

Thank you so much for sharing this tool. I am not sure how to deal batch effect before running this pipeline? Thank you so much.

Best,

Zhen

I tried in several machines install scmer but I get this error:

ERROR: Failed cleaning build dir for numpy
      Failed to build numpy
      ERROR: Could not build wheels for numpy, which is required to install pyproject.toml-based projects
      [end of output]

  note: This error originates from a subprocess, and is likely not a problem with pip.
error: subprocess-exited-with-error

× pip subprocess to install build dependencies did not run successfully.
│ exit code: 1
╰─> See above for output.

note: This error originates from a subprocess, and is likely not a problem with pip.

I want to know which requirements we need, in the requirements file we have only the scikit-learn version, could you please provide the correct versions for torch, numpy, scipy, pandas, and seaborn?

Thanks a lot! 😄