eyaltrabelsi

def test_model_cpu_gpu_same():
  model_gpu = get_trained_model(model)
  model_cpu = export_to_cpu(model_gpu)
  
  base = get_production_sample(pct=0.1)
  base = feature_engineering(input)
  gpu_preds = model_gpu.predict(base)
  cpu_preds = model_cpu.predict(base)
  assert all(abs(gpu_preds - cpu_preds)) < 0.1

eyaltrabelsi

eyaltrabelsi

@given(
    df=data_frames(
        columns=[
            column("Pclass", dtype=int, elements=integers(min_value=1, max_value=3)),
            column("Fare", dtype=int, elements=integers(min_value=0, max_value=1000)),
            column("SibSp", dtype=int, elements=integers(min_value=0, max_value=10)),
            column("Parch", dtype=int, elements=integers(min_value=0, max_value=10)),
            column("Embarked", dtype=int, elements=integers(min_value=0, max_value=3))
        ],
        index=indexes(min_size=1, max_size=10)

eyaltrabelsi

import pandera as pa
from pandera import Column, DataFrameSchema

def test_all_feature_engineering():
    output_schema = DataFrameSchema({
        "Pclass": Column(int),
        "Fare": Column(int),
        "FamilySize": Column(int),
        "Embarked": Column(int)
    })

eyaltrabelsi

// .github/workflows/pull-request.yml
Name: <name>
on:
  pull_request:
    branches:
      - main
jobs:
  tests:
    runs-on: ubuntu-latest
    steps:

eyaltrabelsi

# .pre-commit-config.yaml
repos:
  - repo: https://github.com/pre-commit/pre-commit-hooks
    rev: v4.0.1  # Update to the latest version as necessary
    hooks:
      ...
      - id: check-yaml
        name: Check YAML syntax
      - id: check-json
        name: Check JSON syntax

eyaltrabelsi

# .pre-commit-config.yaml
...
repos:
- repo: https://github.com/sqlfluff/sqlfluff
  rev: stable_version
  hooks:
    - id: sqlfluff-lint

eyaltrabelsi

# .pre-commit-config.yaml
repos:
  - repo: https://github.com/pre-commit/pre-commit-hooks
    rev: v4.0.1 
    hooks:
      - id: check-ast # Testing valid python code
  - repo: https://github.com/pre-commit/mirrors-mypy
    rev: v0.910  # Use the latest version available
    hooks:
      - id: mypy # Testing valid python static types code

eyaltrabelsi

def test_get_family_size():
   df = get_production_sample(pct=0.1)
   func = lambda x: get_family_size(x['SibSp'], x['Parch'])
   sizes = df.apply(func, axis=1)

   assert all(size >= 1 for size in sizes)

eyaltrabelsi

@given(st.integers(), st.integers())
@example(-1, 0)
def test_get_family_size(sib, parch):
   fs = get_family_size(sib, parch)
   assert fs >= 1