Merge pull request #47 from bernalde/pr-23

Pr 23

Author: bernalde

README.md

@@ -18,6 +18,7 @@ Details of the implementation and an illustrative example for Wishart instances
 - [Background](#background)
 - [Installation](#installation)
 - [Examples](#examples)
+- [Testing](#testing)
 - [Contributors](#contributors)
 - [Acknowledgements](#acknowledgements)
 - [License](#license)
@@ -46,8 +47,8 @@ The current package implements the following functionality:
 
 1. **Clone the Repository**:
     ```bash
-    git clone https://github.com/usra-riacs/stochastic-benchmarking.git
-    cd stochastic-benchmarking
+    git clone https://github.com/usra-riacs/stochastic-benchmark.git
+    cd stochastic-benchmark
     ```
 
 2. **Set up a Virtual Environment (Recommended)**:
@@ -64,7 +65,7 @@ The current package implements the following functionality:
 ### Method 2: Downloading as a Zip Archive
 
 1. **Download the Repository**:
-    - Navigate to the [stochastic-benchmarking GitHub page](https://github.com/usra-riacs/stochastic-benchmarking).
+    - Navigate to the [stochastic-benchmark GitHub page](https://github.com/usra-riacs/stochastic-benchmark).
     - Click on the `Code` button.
     - Choose `Download ZIP`.
     - Once downloaded, extract the ZIP archive and navigate to the extracted folder in your terminal or command prompt.
@@ -82,11 +83,36 @@ The current package implements the following functionality:
 
 <!-- the following `pip` command can install this package -->
 
-<!-- ``pip install -i https://test.pypi.org/simple/ stochastic-benchmark==0.0.1`` -->
+<!-- ``pip install -i https://test.pypi.org/simple/ stochastic-benchmark==0.1.0`` -->
 
 ## Examples
 
-For a full demonstration of the stochastic-benchmarking analysis in action, refer to the example notebooks located in the `examples` folder of this repository.
+For a full demonstration of the stochastic-benchmark analysis in action, refer to the example notebooks located in the `examples` folder of this repository.
+
+## Testing
+
+Tests can be executed using the helper script `run_tests.py`. Specify the type of
+tests to run along with any optional flags:
+
+```bash
+python run_tests.py [unit|integration|smoke|all|coverage] [--verbose] [--fast]
+```
+
+Example commands:
+
+- Run the unit test suite:
+
+  ```bash
+  python run_tests.py unit
+  ```
+
+- Generate a coverage report:
+
+  ```bash
+  python run_tests.py coverage
+  ```
+
+For additional details see [TESTING.md](TESTING.md).
 
 ## Contributors
 - [@robinabrown](https://github.com/robinabrown) Robin Brown

examples/QAOA_iterative/qaoa_demo.ipynb

@@ -0,0 +1,4 @@
+-r requirements.txt
+pytest
+pytest-cov
+pytest-xdist

requirements-dev.txt

@@ -1,13 +1,15 @@
-cloudpickle==2.2.0
-dill==0.3.5.1
-fonttools==4.25.0
-future==0.18.3
-hyperopt==0.2.7
-mkl-service==2.4.0
-multiprocess==0.70.13
-munkres==1.1.4
-networkx==2.8.6
-numpy==1.22.0
-pandas==1.4.3
-seaborn==0.12.0rc0
-tqdm==4.66.3
+cloudpickle>=2.2
+dill>=0.3.5
+fonttools>=4.25
+future>=0.18
+hyperopt>=0.2.7
+mkl-service>=2.5.2
+multiprocess>=0.70.18
+munkres>=1.1.4
+networkx>=3.5
+numpy>=2.0
+pandas>=2.3
+scipy>=1.11
+matplotlib>=3.7
+seaborn>=0.13.2
+tqdm>=4.66

requirements.txt

@@ -9,6 +9,7 @@
 import subprocess
 import sys
 import os
+import glob
 
 def run_command(cmd, description):
     """Run a command and handle errors."""
@@ -60,12 +61,18 @@ def main():
     success = True
 
     if args.test_type == "unit" or args.test_type == "all":
-        cmd = ["python", "-m", "pytest", "tests/test_*.py"]
-        if args.verbose:
-            cmd.append("-v")
-        if args.fast:
-            cmd.extend(["-m", "not slow"])
-        success &= run_command(cmd, "Unit tests")
+        # Expand glob pattern manually so pytest receives explicit file paths
+        unit_files = sorted(glob.glob(os.path.join("tests", "test_*.py")))
+        if not unit_files:
+            print("No unit test files found")
+            success &= False
+        else:
+            cmd = ["python", "-m", "pytest", *unit_files]
+            if args.verbose:
+                cmd.append("-v")
+            if args.fast:
+                cmd.extend(["-m", "not slow"])
+            success &= run_command(cmd, "Unit tests")
 
     if args.test_type == "integration" or args.test_type == "all":
         cmd = ["python", "-m", "pytest", "tests/integration/"]

run_tests.py

@@ -5,7 +5,7 @@
 
 setuptools.setup(
     name="stochastic-benchmark",
-    version="0.0.1",
+    version="0.1.0",
     author="David Bernal",
     author_email="dbernalneira@usra.edu",
     description="A package to analyze benchmarking results of stochastic optimization solvers",
@@ -17,10 +17,10 @@
     },
     classifiers=[
         "Programming Language :: Python :: 3",
-        "License :: OSI Approved :: MIT License",
+        "License :: OSI Approved :: Apache Software License",
         "Operating System :: OS Independent",
     ],
     package_dir={"": "src"},
     packages=setuptools.find_packages(where="src"),
-    python_requires=">=3.6",
+    python_requires=">=3.9",
 )

setup.py

@@ -1,6 +1,7 @@
 from collections import defaultdict
 import copy
 from dataclasses import dataclass, field
+import logging
 import df_utils
 from itertools import product
 from multiprocess import Process, Pool, Manager
@@ -17,6 +18,8 @@
 confidence_level = 68
 gap = 1.0
 
+logger = logging.getLogger(__name__)
+
 tqdm.pandas()
 
 
@@ -272,7 +275,7 @@ def Bootstrap(df, group_on, bs_params_list, progress_dir=None):
         df = pd.read_pickle(df)
 
     if type(df) != pd.DataFrame:
-        print("Unsupport type as bootstrap input")
+        logger.error("Unsupported type as bootstrap input")
 
     def f(bs_params):
         if progress_dir is not None:
@@ -365,12 +368,12 @@ def bs_params_eval(bs_params):
 
     elif type(df) == list:
         if type(df[0]) == str:
-            print("calling list of names method")
+            logger.debug("calling list of names method")
 
             def upper_f(upper_group_filename, bs_params_list):
                 df_group = pd.read_pickle(upper_group_filename)
                 group_name = name_fcn(upper_group_filename)
-                print("evaluation bs for {}".format(group_name))
+                logger.info("evaluation bs for %s", group_name)
                 filename = os.path.join(
                     bootstrap_dir, "bootstrapped_results_{}.pkl".format(group_name)
                 )  # TODO fix filename

src/bootstrap.py

@@ -268,7 +268,7 @@ def interpolate_raw_performance(
 
     # For each split_ind, do interpolation separately over the resource grid
     temp_df_interp = df_many_splits_performance_raw.groupby(group_on).progress_apply(
-        lambda df: interpolate.InterpolateSingle(df, iParams, group_on)
+        lambda df: interpolate.InterpolateSingle(df, iParams, group_on), include_groups=False
     )
     temp_df_interp.reset_index(inplace=True)
     return temp_df_interp

src/cross_validation.py

@@ -1,6 +1,7 @@
 import glob
 from multiprocess import Pool
 import os
+import logging
 import pandas as pd
 import names
 import numpy as np
@@ -10,6 +11,8 @@
 s = 0.99
 gap = 1.0
 
+logger = logging.getLogger(__name__)
+
 
 def applyParallel(dfGrouped, func):
     """
@@ -91,8 +94,7 @@ def monotone_df(
                     )
                 ]
                 if len(matched_row) == 0:
-                    print("No matched row found for")
-                    print(row)
+                    logger.debug("No matched row found for row: %s", row)
                     continue
                 else:
                     matched_row = matched_row.iloc[0]
@@ -101,7 +103,7 @@ def monotone_df(
                 df.loc[idx, rolling_cols] = matched_row[rolling_cols].copy()
         else:
             prev_row = row.copy()
-    df.drop(columns=[c for c in df.columns if "level" in c])
+    df.drop(columns=[c for c in df.columns if "level" in c], inplace=True)
     return df
 
 
@@ -123,17 +125,18 @@ def eval_cumm(df, group_on, resource_col, response_col, opt_sense):
     Returns
     -------
     df : pd.DataFrame
-        Datafram with the cummulative evaluation
+        Dataframe with the cumulative evaluation
     """
 
     def cummSingle(single_df):
         single_df = monotone_df(single_df.copy(), resource_col, response_col, 1)
-        single_df.loc[:, "cummulative" + resource_col] = (
+        single_df.loc[:, "cumulative" + resource_col] = (
             single_df[resource_col].expanding(min_periods=1).sum()
         )
         return single_df
 
-    cumm_df = df.groupby(group_on).apply(cummSingle)
+    cumm_df = df.groupby(group_on).apply(cummSingle, include_groups=False)
+    cumm_df.reset_index(inplace=True)
     return cumm_df
 
 

src/df_utils.py

@@ -40,13 +40,13 @@ def __post_init__(self):
             self.resource_value_type = "log"
 
         elif self.resource_value_type in ["data", "log"] and (
-            len(self.resource_values) >= 0
+            len(self.resource_values) > 0
         ):
             warn_str = "Resource value type {} does not support passing in values. Removing.".format(
                 self.resource_value_type
             )
             warnings.warn(warn_str)
-            self.resource_value = []
+            self.resource_values = []
 
 
 def generateResourceColumn(df: pd.DataFrame, interp_params: InterpolationParameters):
@@ -124,15 +124,12 @@ def InterpolateSingle(
     df_out = pd.DataFrame(index=interpolate_resource)
     df_out.index.name = "resource"
 
-    for colname, col in df_single.iteritems():
-        col = pd.to_numeric(col, errors="ignore")
+    for colname, col in df_single.items():
         if colname in group_on:
             continue
         elif colname in interp_params.ignore_cols:
             df_out[colname] = df_single[colname].iloc[0]
-        elif np.issubdtype(
-            col, np.number
-        ):  # An alternative: pd.api.types.is_numeric_dtype(col)
+        elif np.issubdtype(col.dtype, np.number):
             df_out[colname] = np.interp(
                 interpolate_resource, df_single.index, col, left=np.nan
             )
@@ -165,8 +162,7 @@ def Interpolate(df: pd.DataFrame, interp_params: InterpolationParameters, group_
     def dfInterp(df):
         return InterpolateSingle(df, interp_params, group_on)
 
-    df_interp = df.groupby(group_on).progress_apply(dfInterp)
-    df_interp.reset_index(inplace=True)
+    df_interp = df.groupby(group_on).progress_apply(dfInterp, include_groups=False)
     return df_interp
 
 
@@ -195,7 +191,8 @@ def Interpolate_reduce_mem(
         df = pd.read_pickle(df_name)
         generateResourceColumn(df, interp_params)
         temp_df_interp = df.groupby(group_on).progress_apply(
-            lambda df: InterpolateSingle(df, interp_params, group_on)
+            lambda df: InterpolateSingle(df, interp_params, group_on),
+            include_groups=False
         )
         temp_df_interp.reset_index(inplace=True)
         df_interp_list.append(temp_df_interp)