Merge pull request #16 from thomasthaddeus/dev

Dev
thomasthaddeus · May 6, 2024 · 65e5b96 · 65e5b96
2 parents 7fc3e04 + 0fd5d37
commit 65e5b96
Show file tree

Hide file tree

Showing 33 changed files with 132 additions and 6 deletions.
diff --git a/.vscode/settings.json b/.vscode/settings.json
@@ -0,0 +1,7 @@
+{
+    "python.testing.pytestArgs": [
+        "tests"
+    ],
+    "python.testing.unittestEnabled": false,
+    "python.testing.pytestEnabled": true
+}
diff --git a/pyproject.toml b/pyproject.toml
@@ -32,10 +32,66 @@ requires-python = ">=3.8"
 keywords = ["data analysis", "CSV", "statistics", "data cleaning", "data visualization"]
 # include = ["src/**/*.py", "docs/**/*.md"]
 # exclude = ["tests/", "examples/"]
+dependencies = [
+    "backports.tarfile==1.1.1",
+    "build==1.2.1",
+    "certifi==2024.2.2",
+    "charset-normalizer==3.3.2",
+    "click==8.1.7",
+    "colorama==0.4.6",
+    "contourpy==1.2.1",
+    "cycler==0.12.1",
+    "docutils==0.21.2",
+    "fonttools==4.51.0",
+    "greenlet==3.0.3",
+    "idna==3.7",
+    "importlib_metadata==7.1.0",
+    "jaraco.classes==3.4.0",
+    "jaraco.context==5.3.0",
+    "jaraco.functools==4.0.1",
+    "joblib==1.4.2",
+    "keyring==25.2.0",
+    "kiwisolver==1.4.5",
+    "markdown-it-py==3.0.0",
+    "matplotlib==3.8.4",
+    "mdurl==0.1.2",
+    "more-itertools==10.2.0",
+    "nh3==0.2.17",
+    "nltk==3.8.1",
+    "numpy==1.26.4",
+    "packaging==24.0",
+    "pandas==2.2.2",
+    "pillow==10.3.0",
+    "pkginfo==1.10.0",
+    "Pygments==2.18.0",
+    "pyparsing==3.1.2",
+    "pyproject_hooks==1.1.0",
+    "python-dateutil==2.9.0.post0",
+    "pytz==2024.1",
+    "pywin32-ctypes==0.2.2",
+    "readme_renderer==43.0",
+    "regex==2024.4.28",
+    "requests==2.31.0",
+    "requests-toolbelt==1.0.0",
+    "rfc3986==2.0.0",
+    "rich==13.7.1",
+    "scikit-learn==1.4.2",
+    "scipy==1.13.0",
+    "seaborn==0.13.2",
+    "six==1.16.0",
+    "SQLAlchemy==2.0.30",
+    "threadpoolctl==3.5.0",
+    "tqdm==4.66.4",
+    "twine==5.0.0",
+    "typing_extensions==4.11.0",
+    "tzdata==2024.1",
+    "urllib3==2.2.1",
+    "zipp==3.18.1"
+]
 
 [project.optional-dependencies]
 dev = ["pytest", "pytest-mock", "check-manifest"]
-docs = ["sphinx>=2.2"]
+docs = ["myst-parser", "sphinx", "nbsphinx", "sphinx-rtd-theme", "sphinxcontrib-websupport"]
 
 [tool.setuptools.dynamic]
 version = {attr = "dataanalysistoolkit.__version__"}

diff --git a/requirements2.txt b/requirements2.txt
@@ -0,0 +1,54 @@
+backports.tarfile==1.1.1
+build==1.2.1
+certifi==2024.2.2
+charset-normalizer==3.3.2
+click==8.1.7
+colorama==0.4.6
+contourpy==1.2.1
+cycler==0.12.1
+docutils==0.21.2
+fonttools==4.51.0
+greenlet==3.0.3
+idna==3.7
+importlib_metadata==7.1.0
+jaraco.classes==3.4.0
+jaraco.context==5.3.0
+jaraco.functools==4.0.1
+joblib==1.4.2
+keyring==25.2.0
+kiwisolver==1.4.5
+markdown-it-py==3.0.0
+matplotlib==3.8.4
+mdurl==0.1.2
+more-itertools==10.2.0
+nh3==0.2.17
+nltk==3.8.1
+numpy==1.26.4
+packaging==24.0
+pandas==2.2.2
+pillow==10.3.0
+pkginfo==1.10.0
+Pygments==2.18.0
+pyparsing==3.1.2
+pyproject_hooks==1.1.0
+python-dateutil==2.9.0.post0
+pytz==2024.1
+pywin32-ctypes==0.2.2
+readme_renderer==43.0
+regex==2024.4.28
+requests==2.31.0
+requests-toolbelt==1.0.0
+rfc3986==2.0.0
+rich==13.7.1
+scikit-learn==1.4.2
+scipy==1.13.0
+seaborn==0.13.2
+six==1.16.0
+SQLAlchemy==2.0.30
+threadpoolctl==3.5.0
+tqdm==4.66.4
+twine==5.0.0
+typing_extensions==4.11.0
+tzdata==2024.1
+urllib3==2.2.1
+zipp==3.18.1
diff --git a/src/.conf/plot_config.json → ...ataanalysistoolkit/.conf/plot_config.json b/src/.conf/plot_config.json → ...ataanalysistoolkit/.conf/plot_config.json
diff --git a/src/__init__.py → src/dataanalysistoolkit/__init__.py b/src/__init__.py → src/dataanalysistoolkit/__init__.py
diff --git a/src/data_analysis_toolkit.py → ...aanalysistoolkit/data_analysis_toolkit.py b/src/data_analysis_toolkit.py → ...aanalysistoolkit/data_analysis_toolkit.py
@@ -83,10 +83,12 @@ def __init__(self, filename):
             filename (_type_): _description_
         """
         self.data = self.load_data(filename)
-        self.visualizer = DataVisualizer(self.data)
+        self.visualizer = DataVisualizer(self.data, config_path='.conf/plot_config.json')
         self.imputer = DataImputer(self.data)
         self.preprocessor = DataPreprocessor(self.data)
         self.feature_engineer = FeatureEngineer(self.data)
+        # No value for argument 'X_test' in constructor callPylintE1120:no-value-for-parameter
+        # No value for argument 'y_test' in constructor callPylintE1120:no-value-for-parameter
         self.evaluator = ModelEvaluator(self.data)
         self.report_generator = ReportGenerator(self.data)
 

diff --git a/src/data_sources/__init__.py → ...aanalysistoolkit/data_sources/__init__.py b/src/data_sources/__init__.py → ...aanalysistoolkit/data_sources/__init__.py
diff --git a/src/data_sources/api_connector.py → ...ysistoolkit/data_sources/api_connector.py b/src/data_sources/api_connector.py → ...ysistoolkit/data_sources/api_connector.py
diff --git a/src/data_sources/excel_connector.py → ...istoolkit/data_sources/excel_connector.py b/src/data_sources/excel_connector.py → ...istoolkit/data_sources/excel_connector.py
diff --git a/src/data_sources/sql_connector.py → ...ysistoolkit/data_sources/sql_connector.py b/src/data_sources/sql_connector.py → ...ysistoolkit/data_sources/sql_connector.py
diff --git a/src/formatters/__init__.py → ...ataanalysistoolkit/formatters/__init__.py b/src/formatters/__init__.py → ...ataanalysistoolkit/formatters/__init__.py
diff --git a/src/formatters/data_formatter.py → ...lysistoolkit/formatters/data_formatter.py b/src/formatters/data_formatter.py → ...lysistoolkit/formatters/data_formatter.py
diff --git a/src/generators/__init__.py → ...ataanalysistoolkit/generators/__init__.py b/src/generators/__init__.py → ...ataanalysistoolkit/generators/__init__.py
diff --git a/src/generators/csv_data_generator.py → ...stoolkit/generators/csv_data_generator.py b/src/generators/csv_data_generator.py → ...stoolkit/generators/csv_data_generator.py
diff --git a/src/generators/generate_data.py → ...alysistoolkit/generators/generate_data.py b/src/generators/generate_data.py → ...alysistoolkit/generators/generate_data.py
diff --git a/src/generators/report_generator.py → ...sistoolkit/generators/report_generator.py b/src/generators/report_generator.py → ...sistoolkit/generators/report_generator.py
diff --git a/src/integrators/__init__.py → ...taanalysistoolkit/integrators/__init__.py b/src/integrators/__init__.py → ...taanalysistoolkit/integrators/__init__.py
diff --git a/src/integrators/data_integrator.py → ...sistoolkit/integrators/data_integrator.py b/src/integrators/data_integrator.py → ...sistoolkit/integrators/data_integrator.py
diff --git a/src/model/__init__.py → src/dataanalysistoolkit/model/__init__.py b/src/model/__init__.py → src/dataanalysistoolkit/model/__init__.py
diff --git a/src/model/feature_engineer.py → ...analysistoolkit/model/feature_engineer.py b/src/model/feature_engineer.py → ...analysistoolkit/model/feature_engineer.py
diff --git a/src/model/model_evaluator.py → ...aanalysistoolkit/model/model_evaluator.py b/src/model/model_evaluator.py → ...aanalysistoolkit/model/model_evaluator.py
diff --git a/src/preprocessor/__init__.py → ...aanalysistoolkit/preprocessor/__init__.py b/src/preprocessor/__init__.py → ...aanalysistoolkit/preprocessor/__init__.py
diff --git a/src/preprocessor/data_prep.py → ...analysistoolkit/preprocessor/data_prep.py b/src/preprocessor/data_prep.py → ...analysistoolkit/preprocessor/data_prep.py
diff --git a/src/utils/__init__.py → src/dataanalysistoolkit/utils/__init__.py b/src/utils/__init__.py → src/dataanalysistoolkit/utils/__init__.py
diff --git a/src/utils/data_imputer.py → ...dataanalysistoolkit/utils/data_imputer.py b/src/utils/data_imputer.py → ...dataanalysistoolkit/utils/data_imputer.py
diff --git a/src/visualizer/__init__.py → ...ataanalysistoolkit/visualizer/__init__.py b/src/visualizer/__init__.py → ...ataanalysistoolkit/visualizer/__init__.py
diff --git a/src/visualizer/data_visualizer.py → ...ysistoolkit/visualizer/data_visualizer.py b/src/visualizer/data_visualizer.py → ...ysistoolkit/visualizer/data_visualizer.py
diff --git a/tests/__init__.py b/tests/__init__.py
@@ -0,0 +1,7 @@
+from .data_sources import test_api_connector, test_excel_connector, test_sql_connector
+from .generators import test_csv_data_generator, test_generate_data, test_report_generator
+from .integrators import test_data_integrator
+from .model import test_feature_engineer, test_model_evaluator
+from .preprocessor import test_data_prep
+from .utils import test_data_imputer
+from .visualizer import test_data_visualizer
diff --git a/data/gen_test.csv → tests/data/gen_test.csv b/data/gen_test.csv → tests/data/gen_test.csv
diff --git a/data/test_random.csv → tests/data/test_random.csv b/data/test_random.csv → tests/data/test_random.csv
diff --git a/tests/data_sources/test_api_connector.py b/tests/data_sources/test_api_connector.py
@@ -1,7 +1,7 @@
 import pytest
 import requests
 import requests_mock
-from src.data_sources.api_connector import APIConnector
+from dataanalysistoolkit.data_sources.api_connector import APIConnector
 
 @pytest.fixture
 def api_connector():

diff --git a/tests/data_sources/test_excel_connector.py b/tests/data_sources/test_excel_connector.py
@@ -1,6 +1,6 @@
 import pytest
 import pandas as pd
-from src.data_sources.excel_connector import ExcelConnector
+from dataanalysistoolkit.data_sources.excel_connector import ExcelConnector
 
 # Create a sample Excel file for testing
 @pytest.fixture(scope="module")

diff --git a/tests/test_data_analysis_toolkit.py b/tests/test_data_analysis_toolkit.py
@@ -23,12 +23,12 @@
 
 import pytest
 import pandas as pd
-from src.data_analysis_toolkit import DataAnalysisToolkit
+from dataanalysistoolkit.data_analysis_toolkit import DataAnalysisToolkit
 
 
 @pytest.fixture
 def analyzer():
-    return DataAnalysisToolkit("../data/test.csv")
+    return DataAnalysisToolkit("./data/test_random.csv")
 
 
 def test_calculate_budget_statistics(analyzer):