ci(benchmark): Benchmark introduction (ydataai#753)

.github/workflows/benchmark.yml

@@ -0,0 +1,37 @@
+name: Performance Benchmarks
+
+on:
+  push:
+    branches:
+      - master
+      - develop
+
+jobs:
+  benchmark:
+    name: ${{ matrix.os }} x ${{ matrix.python }}
+    runs-on: ${{ matrix.os }}
+    strategy:
+      fail-fast: false
+      matrix:
+        os: [ ubuntu-latest ] #, macos-latest, windows-latest ]
+        python: ['3.8']
+    steps:
+      - uses: actions/checkout@v2
+        with:
+          fetch-depth: 0
+      - uses: actions/setup-python@v1
+        with:
+          python-version: ${{ matrix.python }}
+      - name: Run benchmark
+        run: |
+          pip install -r requirements.txt
+          pip install -r requirements-test.txt
+          pytest tests/benchmarks/bench.py --benchmark-json benchmark.json
+      - name: Store benchmark result
+        uses: rhysd/github-action-benchmark@v1
+        with:
+          name: Pandas Profiling Benchmarks
+          tool: 'pytest'
+          output-file-path: benchmark.json
+          github-token: ${{ secrets.GITHUB_TOKEN }}
+          auto-push: true

requirements-test.txt

@@ -3,6 +3,7 @@ coverage<5
 codecov
 pytest-mypy
 pytest-cov
+pytest-benchmark~=3.2.2
 nbval
 pyarrow
 flake8

tests/benchmarks/bench.py

@@ -0,0 +1,74 @@
+import pandas as pd
+
+from pandas_profiling import ProfileReport
+from pandas_profiling.utils.cache import cache_file
+
+
+def test_titanic_explorative(benchmark):
+    file_name = cache_file(
+        "titanic.parquet",
+        "https://github.com/pandas-profiling/pandas-profiling-data/raw/master/data/titanic.parquet",
+    )
+
+    data = pd.read_parquet(file_name)
+
+    def func(df):
+        profile = ProfileReport(
+            df, title="Titanic Dataset", explorative=True, progress_bar=False
+        )
+        report = profile.to_html()
+        return report
+
+    benchmark(func, data)
+
+
+def test_titanic_default(benchmark):
+    file_name = cache_file(
+        "titanic.parquet",
+        "https://github.com/pandas-profiling/pandas-profiling-data/raw/master/data/titanic.parquet",
+    )
+
+    data = pd.read_parquet(file_name)
+
+    def func(df):
+        profile = ProfileReport(df, title="Titanic Dataset", progress_bar=False)
+        report = profile.to_html()
+        return report
+
+    benchmark(func, data)
+
+
+def test_titanic_minimal(benchmark):
+    file_name = cache_file(
+        "titanic.parquet",
+        "https://github.com/pandas-profiling/pandas-profiling-data/raw/master/data/titanic.parquet",
+    )
+
+    data = pd.read_parquet(file_name)
+
+    def func(df):
+        profile = ProfileReport(
+            df, title="Titanic Dataset", minimal=True, progress_bar=False
+        )
+        report = profile.to_html()
+        return report
+
+    benchmark(func, data)
+
+
+def test_rdw_minimal(benchmark):
+    file_name = cache_file(
+        "rdw.parquet",
+        "https://github.com/pandas-profiling/pandas-profiling-data/raw/master/data/rdw.parquet",
+    )
+
+    data = pd.read_parquet(file_name)
+
+    def func(df):
+        profile = ProfileReport(
+            df, title="RDW Dataset", minimal=True, progress_bar=False
+        )
+        report = profile.to_html()
+        return report
+
+    benchmark(func, data)