autonlab / auton-survival

Auton Survival - an open source package for Regression, Counterfactual Estimation, Evaluation and Phenotyping with Censored Time-to-Events
http://autonlab.github.io/auton-survival
MIT License
315 stars 74 forks source link

requirement text file #141

Open sbmubs opened 5 months ago

sbmubs commented 5 months ago

Hi, thank you for your valuable package, After clone the mentioned repository, to install requirements(pip install -r requirements.txt), I found this error: ERROR: Could not open requirements file: [Errno 2] No such file or directory: 'requirements.txt'

matteo4diani commented 3 months ago

Hi @sbmubs, while the answer from @FeckNeck is technically correct, those requirements.txt files are outdated and could lead to undesired behaviors: the auton-survival project is now available on the Python Package Index so, if you just want to use auton-survival as a dependency in your own project, you only need to run pip install auton-survival.

If instead you want to contribute a fix or a feature to the project you should clone the repo, install poetry and use that for development. You can install the needed dependencies by running poetry install in the root of the project.

If you prefer working with a requirements.txt file in your own fork, then clone the repo and run poetry export --output requirements.txt. This will generate a version-locked and hashed requirements.txt file.

Hope this helps you out 🙂

Jrhasse commented 3 months ago

Hi @matteo4diani, I was trying to download the package via pip like you suggested doing. However, when doing so it starts out downloading and then encounters an error. I am including an screenshot of the bottom bit of the error. I will happily provide more information in order to get the package working on my system. Thanks!

image
matteo4diani commented 3 months ago

Hi @Jrhasse, I need these info to help you troubleshoot the problem:

While you gather this info, at a glance it is most likely a problem with Python versions, try using a more recent version (3.10/3.11) and let me know if that fixes the problem.

Jrhasse commented 3 months ago

I am using Windows 11 (version 23H2). I just installed python via anaconda so I am using 3.12.3. I am relatively inexperienced with Python so my terminology may be incorrect, but my understanding is that it is not a virtual environment but is a conda environment (that is not the "base" environment). Below is a list of my modules/libraries in the environment I am trying to download your package into: blas 1.0 mkl bottleneck 1.3.7 py312he558020_0 brotli 1.0.9 h2bbff1b_8 brotli-bin 1.0.9 h2bbff1b_8 bzip2 1.0.8 h2bbff1b_6 ca-certificates 2024.3.11 haa95532_0 cffi 1.16.0 py312h2bbff1b_1 contourpy 1.2.0 py312h59b6b97_0 cycler 0.11.0 pyhd3eb1b0_0 expat 2.6.2 hd77b12b_0 filelock 3.13.1 py312haa95532_0 fonttools 4.51.0 py312h2bbff1b_0 freetype 2.12.1 ha860e81_0 fsspec 2024.3.1 py312haa95532_0 git 2.40.1 haa95532_1 anaconda icu 73.1 h6c2663c_0 intel-openmp 2023.1.0 h59b6b97_46320 jinja2 3.1.4 py312haa95532_0 jpeg 9e h2bbff1b_1 kiwisolver 1.4.4 py312hd77b12b_0 krb5 1.20.1 h5b6d351_0 lcms2 2.12 h83e58a3_0 lerc 3.0 hd77b12b_0 libbrotlicommon 1.0.9 h2bbff1b_8 libbrotlidec 1.0.9 h2bbff1b_8 libbrotlienc 1.0.9 h2bbff1b_8 libclang 14.0.6 default_hb5a9fac_1 libclang13 14.0.6 default_h8e68704_1 libdeflate 1.17 h2bbff1b_1 libffi 3.4.4 hd77b12b_1 libpng 1.6.39 h8cc25b3_0 libpq 12.17 h906ac69_0 libtiff 4.5.1 hd77b12b_0 libuv 1.44.2 h2bbff1b_0 libwebp-base 1.3.2 h2bbff1b_0 lz4-c 1.9.4 h2bbff1b_1 markupsafe 2.1.3 py312h2bbff1b_0 matplotlib 3.8.4 py312haa95532_0 matplotlib-base 3.8.4 py312hc7c4135_0 mkl 2023.1.0 h6b88ed4_46358 mkl-service 2.4.0 py312h2bbff1b_1 mkl_fft 1.3.8 py312h2bbff1b_0 mkl_random 1.2.4 py312h59b6b97_0 mpmath 1.3.0 py312haa95532_0 networkx 3.1 py312haa95532_0 ninja 1.10.2 haa95532_5 ninja-base 1.10.2 h6d14046_5 numexpr 2.8.7 py312h96b7d27_0 numpy 1.26.4 py312hfd52020_0 numpy-base 1.26.4 py312h4dde369_0 openjpeg 2.4.0 h4fc8c34_0 openssl 3.0.13 h2bbff1b_2 packaging 23.2 py312haa95532_0 pandas 2.2.1 py312h0158946_0 pillow 10.3.0 py312h2bbff1b_0 pip 24.0 py312haa95532_0 ply 3.11 py312haa95532_1 pycparser 2.21 pyhd3eb1b0_0 pyparsing 3.0.9 py312haa95532_0 pyqt 5.15.10 py312hd77b12b_0 pyqt5-sip 12.13.0 py312h2bbff1b_0 python 3.12.3 h1d929f7_1 python-dateutil 2.9.0post0 py312haa95532_2 python-tzdata 2023.3 pyhd3eb1b0_0 pytorch 2.2.0 cpu_py312h746a3fd_0 pytz 2024.1 py312haa95532_0 qt-main 5.15.2 h19c9488_10 seaborn 0.12.2 py312haa95532_0 setuptools 69.5.1 py312haa95532_0 sip 6.7.12 py312hd77b12b_0 six 1.16.0 pyhd3eb1b0_1 sqlite 3.45.3 h2bbff1b_0 sympy 1.12 py312haa95532_0 tbb 2021.8.0 h59b6b97_0 tk 8.6.14 h0416ee5_0 tornado 6.3.3 py312h2bbff1b_0 typing_extensions 4.11.0 py312haa95532_0 tzdata 2024a h04d1e81_0 unicodedata2 15.1.0 py312h2bbff1b_0 vc 14.2 h2eaa2aa_1 vs2015_runtime 14.29.30133 h43f2093_3 wheel 0.43.0 py312haa95532_0 xz 5.4.6 h8cc25b3_1 zlib 1.2.13 h8cc25b3_1 zstd 1.5.5 hd43e919_2

matteo4diani commented 3 months ago

Ok, I can see a couple of issues now, namely:

  1. auton-survival does not support PyTorch 2.x yet
  2. We didn't test auton-survival with Python 3.12 yet

You got two options:

  1. I prepared a PR for another user that wanted to use auton-survival with PyTorch 2, have a look at my response on this issue. Since you are using Conda, you should install auton-survival as a VCS (Version Control System) dependency, you can find the repo URL and the branch name in the pyproject.toml in the linked response.
  2. Create a fresh Conda env with Python 3.11, install auton-survival first so you'll get compatible versions of torch, sklearn, sksurv, etc., then add other libraries not included in auton-survival's dependencies as needed.

Let me know if one of these options is viable for you, and if it fixed your problem.

Jrhasse commented 3 months ago

I would rather use option 2 and use the versions that the module is meant to work with. Makes no difference to me as I need to learn it regardless. I created a fresh conda environment using 3.11.7 but now encountered this issue. It looks like pip cant find an old version of pytorch, but I am not sure. See below.

image
matteo4diani commented 3 months ago

Try a new env with Python 3.10.

Jrhasse commented 3 months ago

Success! For anyone else who runs into similar issues. I ran the following in the anaconda prompt: conda create --name myenv python=3.10 conda activate myenv Python -V pip install auton_survival

Thank you @matteo4diani for your help!

matteo4diani commented 3 months ago

🎉 You're welcome @Jrhasse , happy to help. Thanks for reaching out and enjoy auton-survival.

If you find anything that needs improvement or you need a missing feature please open a new issue here on GitHub :)

MaurizioFi commented 2 months ago

Thank you both! I was going crazy with this installation problem. It worked perfectly (provided one pays attention to run anaconda in myenv too :) ).