0

I am trying to create an environment using

conda env create -f environment.yaml

But I get this error

Found conflicts! Looking for incompatible packages.
This can take several minutes. Press CTRL-C to abort.
failed

UnsatisfiableError: The following specifications were found to be incompatible with each other:

Output in format: Requested package -> Available versions

Package pytorch conflicts for:
torchaudio=0.13.1 -> pytorch==1.13.1
torchvision=0.14.1 -> pytorch[version='1.13.1|>=1.13.1,<1.14.0a0']
pytorch=1.13.1

Package python conflicts for:
torchaudio=0.13.1 -> numpy[version='>=1.11'] -> python[version='>=3.10,<3.11.0a0|>=3.11,<3.12.0a0|>=3.9,<3.10.0a0|>=3.8,<3.9.0a0',build=_cpython]
torchvision=0.14.1 -> python[version='>=3.10,<3.11.0a0|>=3.10,<3.11.0a0|>=3.8,<3.9.0a0|>=3.9,<3.10.0a0|>=3.9,<3.10.0a0|>=3.8,<3.9.0a0',build=_cpython]
torchaudio=0.13.1 -> python[version='>=3.10,<3.11.0a0|>=3.9,<3.10.0a0|>=3.8,<3.9.0a0']
pip -> python[version='>=3.10,<3.11.0a0|>=3|>=3.6|>=3.7|>=3.9,<3.10.0a0|>=3.8,<3.9.0a0|>=3.11,<3.12.0a0']
torchvision=0.14.1 -> numpy[version='>=1.11'] -> python[version='3.10.|>=3.11,<3.12.0a0|>=3.7|>=3.7,<4.0|>=3.6,<4.0|>=3.6|3.9.|3.8.',build=_cpython]
pytorch=1.13.1 -> python[version='>=3.10,<3.11.0a0|>=3.10,<3.11.0a0|>=3.8,<3.9.0a0|>=3.9,<3.10.0a0|>=3.8,<3.9.0a0|>=3.9,<3.10.0a0|>=3.11,<3.12.0a0|>=3.11,<3.12.0a0',build=_cpython]
pytorch=1.13.1 -> typing_extensions -> python[version='3.10.|3.11.|3.8.|>=3.5|>=3.6|>=3.7|3.9.|3.9.16|3.8.16|3.9.10|3.8.12|3.7.12|3.7.10|3.7.10|3.6.12|3.7.9|3.6.12|3.6.9|3.6.9|3.6.9|3.6.9|>=3.8',build='0_73_pypy|1_73_pypy|2_73_pypy|3_73_pypy|0_73_pypy|1_73_pypy|0_73_pypy|5_73_pypy|5_73_pypy|4_73_pypy']
pip -> setuptools -> python[version='!=3.0,!=3.1,!=3.2,!=3.3,!=3.4|>=3.10,<3.11.0a0|>=3.8|>=3.8,<3.9.0a0|>=3.9,<3.10.0a0|>=3.11,<3.12.0a0|2.7.|>=3.6',build=*_cpython]

Package wheel conflicts for:
pip -> wheel
python=3.10 -> pip -> wheel

Package certifi conflicts for:
pip -> setuptools -> certifi[version='>=2016.9.26']
torchvision=0.14.1 -> requests -> certifi[version='>=2017.4.17']
pytorch=1.13.1 -> setuptools -> certifi[version='>=2016.9.26']

Package libiconv conflicts for:
torchvision=0.14.1 -> ffmpeg[version='>=4.2'] -> libiconv[version='>=1.16,<2.0.0a0|>=1.17,<2.0a0|>=1.16,<2.0a0']
pytorch=1.13.1 -> python[version='>=3.9,<3.10.0a0'] -> libiconv[version='>=1.16,<2.0a0']
pip -> python[version='>=3.7'] -> libiconv[version='>=1.16,<2.0a0']
torchaudio=0.13.1 -> python[version='>=3.9,<3.10.0a0'] -> libiconv[version='>=1.16,<2.0a0']

Package libexpat conflicts for:
pytorch=1.13.1 -> python[version='>=3.11,<3.12.0a0'] -> libexpat[version='>=2.5.0,<3.0a0']
pip -> python[version='>=3.7'] -> libexpat[version='>=2.5.0,<3.0a0']

Package python_abi conflicts for:
torchvision=0.14.1 -> numpy[version='>=1.11'] -> python_abi=3.11[build=_cp311]
torchvision=0.14.1 -> python_abi[version='3.10.|3.9.|3.8.',build='_cp38|_cp310|*_cp39']

Here is the yaml content

name: animatediff
channels:
  - pytorch
  - nvidia
  - conda-forge
  - anaconda
  - defaults
dependencies:
  - python=3.10
  - pytorch=1.13.1
  - torchvision=0.14.1
  - torchaudio=0.13.1
  - pytorch-cuda=11.7
  - pip
  - pip:
    - diffusers==0.11.1
    - transformers==4.25.1
    - xformers==0.0.16
    - imageio==2.27.0
    - gdown
    - einops
    - omegaconf
    - safetensors
    - gradio


What I tried

I've tried all the answers from the link below with no luck, any suggestions are highly appreciated

Create conda environment: "Found conflicts!" when solving environment and "Finding shortest conflict path" running forever

I am using conda 23.5.2 and Python 3.11.4

  • I guess the versions are very strict and it's having some problem to find a plausible solution. I think one possible alternative would be to create a new environment, [install pytorch following the instruction from the website](https://pytorch.org/get-started/locally/), ([or install previous versions](https://pytorch.org/get-started/previous-versions/)) and then install the rest via command (or worst case one by one). Also if you do not care about versions, removing the version numbering could allow conda to search for a more flexible solution.. – freerafiki Jul 25 '23 at 12:40

0 Answers0