-
-
Notifications
You must be signed in to change notification settings - Fork 50
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
Fix pytorch-{cpu,gpu} packages are built only for single python version #281
Comments
I think the following patch will work: diff --git a/recipe/meta.yaml b/recipe/meta.yaml
index 46a7507..e56535c 100644
--- a/recipe/meta.yaml
+++ b/recipe/meta.yaml
@@ -347,7 +347,9 @@ outputs:
- pytorch-cpu # [cuda_compiler_version == "None"]
requirements:
run:
- - {{ pin_subpackage("pytorch", exact=True) }}
+ - pytorch {{ version }}=cuda_{{ blas_impl }}*{{ PKG_BUILDNUM }} # [megabuild and cuda_compiler_version != "None"]
+ - pytorch {{ version }}=cpu_{{ blas_impl }}*{{ PKG_BUILDNUM }} # [megabuild and cuda_compiler_version == "None"]
+ - {{ pin_subpackage("pytorch", exact=True) }} # [not megabuild]
test:
imports:
- torch |
Hmm we might need the build string to also be "resolved" diff --git a/recipe/meta.yaml b/recipe/meta.yaml
index 46a7507..7fbee9f 100644
--- a/recipe/meta.yaml
+++ b/recipe/meta.yaml
@@ -338,8 +338,10 @@ outputs:
{% set pytorch_cpu_gpu = "pytorch-gpu" %} # [cuda_compiler_version != "None"]
- name: {{ pytorch_cpu_gpu }}
build:
- string: cuda{{ cuda_compiler_version | replace('.', '') }}py{{ CONDA_PY }}h{{ PKG_HASH }}_{{ PKG_BUILDNUM }} # [cuda_compiler_version != "None"]
- string: cpu_{{ blas_impl }}_py{{ CONDA_PY }}h{{ PKG_HASH }}_{{ PKG_BUILDNUM }} # [cuda_compiler_version == "None"]
+ string: cuda{{ cuda_compiler_version | replace('.', '') }}h{{ PKG_HASH }}_{{ PKG_BUILDNUM }} # [megabuild and cuda_compiler_version != "None"]
+ string: cpu_{{ blas_impl }}_h{{ PKG_HASH }}_{{ PKG_BUILDNUM }} # [megabuild and cuda_compiler_version == "None"]
+ string: cuda{{ cuda_compiler_version | replace('.', '') }}py{{ CONDA_PY }}h{{ PKG_HASH }}_{{ PKG_BUILDNUM }} # [not megabuild and cuda_compiler_version != "None"]
+ string: cpu_{{ blas_impl }}_py{{ CONDA_PY }}h{{ PKG_HASH }}_{{ PKG_BUILDNUM }} # [not megabuild and cuda_compiler_version == "None"]
detect_binary_files_with_prefix: false
skip: true # [cuda_compiler_version != "None" and linux64 and blas_impl != "mkl"]
# weigh down cpu implementation and give cuda preference
@@ -347,7 +349,9 @@ outputs:
- pytorch-cpu # [cuda_compiler_version == "None"]
requirements:
run:
- - {{ pin_subpackage("pytorch", exact=True) }}
+ - pytorch {{ version }}=cuda_{{ blas_impl }}*{{ PKG_BUILDNUM }} # [megabuild and cuda_compiler_version != "None"]
+ - pytorch {{ version }}=cpu_{{ blas_impl }}*{{ PKG_BUILDNUM }} # [megabuild and cuda_compiler_version == "None"]
+ - {{ pin_subpackage("pytorch", exact=True) }} # [not megabuild]
test:
imports:
- torch |
5 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
That said this seems to be a problem with the 2.4.1 package as well and likely for many of the
megabuild
packages....https://conda-metadata-app.streamlit.app/?q=conda-forge%2Flinux-64%2Fpytorch-cpu-2.4.1-cpu_mkl_py39h09a6fac_103.conda
Originally posted by @hmaarrfk in #277 (comment)
The text was updated successfully, but these errors were encountered: