pyro-meta.yaml

{% set name = "pyro-ppl" %}
{% set version = "0.3.1" %}
{% set sha256 = "c776ddc41c218e1d86f4ea177218688d458226ce6987559b818771b943b940c0" %}

package:
  name: {{ name|lower }}
  version: {{ version }}

source:
  url: https://github.com/pyro-ppl/pyro/archive/{{ version }}.tar.gz
  sha256: {{ sha256 }}

build:
  noarch: python
  number: 0
  script: "{{ PYTHON }} -m pip install . --no-deps -vv"

requirements:
  host:
    - python >=2.7,<3|>=3.6
    - pip
  run:
    - python >=2.7,<3|>=3.6
    - contextlib2
    - opt_einsum >=2.2
    - python-graphviz >=0.8
    - six >=1.10
    - pytorch >=1.0
    - tqdm >=4.28

test:
  imports:
    - pyro
    - pyro.contrib
    - pyro.distributions
    - pyro.infer
    - pyro.nn
    - pyro.ops
    - pyro.optim
    - pyro.params
    - pyro.poutine

about:
  home: http://pyro.ai/
  license: MIT
  license_family: MIT
  license_file: LICENSE.md
  summary: 'A Python library for probabilistic modeling and inference'

  description: |
    Pyro is a universal probabilistic programming language (PPL) written in
    Python and supported by PyTorch on the backend. Pyro enables flexible and
    expressive deep probabilistic modeling, unifying the best of modern deep
    learning and Bayesian modeling.
  doc_url: http://docs.pyro.ai/
  dev_url: https://github.com/pyro-ppl/pyro

extra:
  recipe-maintainers:
    - fehiepsi