diff --git a/.gitignore b/.gitignore
index 96737d1..a7d3817 100644
--- a/.gitignore
+++ b/.gitignore
@@ -5,7 +5,36 @@ data/
docs/build/*
docs/source/api/
-### Python execution ###
-scripts/__pycache__/
-tests/output/
+### Python ###
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+pip-wheel-metadata/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# test results
+rr/tests/output/
diff --git a/.travis.yml b/.travis.yml
index 0674d0c..f20551b 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -6,27 +6,19 @@ python:
- "3.8"
install:
# Install phase of our CI pipeline
- - wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh -O miniconda.sh
- - bash miniconda.sh -b -p $HOME/miniconda
- - source "$HOME/miniconda/etc/profile.d/conda.sh"
- - hash -r
- - conda config --set always_yes yes --set changeps1 no
- - conda update -q conda
- - conda info -a
- - conda create -q -n project python=$TRAVIS_PYTHON_VERSION -c plotly --file requirements.txt --file build-requirements.txt
- - conda activate project
+ - pip install .
+ - pip install -r build-requirements.txt
script:
- # Run phase of our CI pipeline
# Run unitTests
- - nosetests --nocapture -v tests/test.py --with-coverage --cover-package=scripts
+ - nosetests --nocapture -v rr --with-coverage --cover-package=rr
# Generate rst files from docstring
- - sphinx-apidoc -fMeT -o docs/source/api scripts
+ - sphinx-apidoc -fMeT -o docs/source/api rr rr/tests
# Build documentation
- sphinx-build docs/source docs/build
- touch docs/build/.nojekyll
# Black lint checking
- - black --check scripts/ tests/
+ - black --check rr/
after_success:
# update coveralls report
- coveralls
@@ -39,4 +31,12 @@ deploy:
local_dir: ./docs/build
on:
branch: master
- condition: $TRAVIS_PYTHON_VERSION = 3.8
+ condition: "$TRAVIS_PYTHON_VERSION = 3.8"
+# - provider: pypi
+# server: https://test.pypi.org/legacy/ # Remove to deployment on pypi.org
+# username: "__token__"
+# password:
+# secure: AgENdGVzdC5weXBpLm9yZwIkNDkzMjE2ZDMtYjlhYy00NTViLTg1MzYtZGU4N2YyNjA0NDY1AAIleyJwZXJtaXNzaW9ucyI6ICJ1c2VyIiwgInZlcnNpb24iOiAxfQAABiCE1UhRrnGVriZfsUQEp0YRaWTuAOLsDggCBBsOB-PbMQ
+# on:
+# tags: true
+# condition: "$TRAVIS_PYTHON_VERSION = 3.8"
diff --git a/MANIFEST.in b/MANIFEST.in
new file mode 100644
index 0000000..c189475
--- /dev/null
+++ b/MANIFEST.in
@@ -0,0 +1,3 @@
+include LICENSE README.rst buildout.cfg requirements.txt
+recursive-include doc conf.py *.rst *.ico *.png
+graft tests/inputs
diff --git a/build-requirements.txt b/build-requirements.txt
index c362dd0..65015d0 100644
--- a/build-requirements.txt
+++ b/build-requirements.txt
@@ -4,3 +4,4 @@ nose
coverage
coveralls
black
+twine
diff --git a/docs/source/conf.py b/docs/source/conf.py
index 8375dfc..2297d5f 100644
--- a/docs/source/conf.py
+++ b/docs/source/conf.py
@@ -13,7 +13,7 @@
import os
import sys
import subprocess
-sys.path.insert(0, os.path.abspath('../../scripts'))
+sys.path.insert(0, os.path.abspath('../../experiment'))
# -- Project information -----------------------------------------------------
diff --git a/experiment.ini b/experiment.ini
new file mode 100644
index 0000000..604e7dd
--- /dev/null
+++ b/experiment.ini
@@ -0,0 +1,7 @@
+[nb_trees_experiment]
+nb_trees = 1, 2
+tree_depth = 10
+
+[tree_depth_experiment]
+nb_trees = 10
+tree_depth = 1, 2
diff --git a/scripts/__ini__.py b/rr/__init__.py
similarity index 100%
rename from scripts/__ini__.py
rename to rr/__init__.py
diff --git a/tests/inputs/not_a_zip.txt b/rr/download_data/__init__.py
similarity index 100%
rename from tests/inputs/not_a_zip.txt
rename to rr/download_data/__init__.py
diff --git a/rr/download_data/__main__.py b/rr/download_data/__main__.py
new file mode 100644
index 0000000..6dfd740
--- /dev/null
+++ b/rr/download_data/__main__.py
@@ -0,0 +1,22 @@
+import sys
+import argparse
+import os.path
+from rr.download_data import download_data
+
+
+def main():
+
+ parser = argparse.ArgumentParser(
+ description="M05 mini-project: Download dataset.zip online"
+ )
+ parser.add_argument("source", type=str, help="Data zip url")
+ parser.add_argument("destination", type=str, help="Destination folder")
+ args = parser.parse_args()
+
+ download_destination = os.path.join(args.destination + "/dataset.zip")
+ download_data.download_url(args.source, download_destination)
+ download_data.unzip_file(download_destination, args.destination)
+
+
+if __name__ == "__main__":
+ sys.exit(main())
diff --git a/scripts/download_data.py b/rr/download_data/download_data.py
similarity index 70%
rename from scripts/download_data.py
rename to rr/download_data/download_data.py
index 7a5e977..88c175b 100644
--- a/scripts/download_data.py
+++ b/rr/download_data/download_data.py
@@ -3,7 +3,7 @@
import zipfile
-# Function to download a file through http.get using requests
+# Function to download_data a file through http.get using requests
def download_url(url, save_path):
"""Download a file from the given url using http
@@ -16,7 +16,7 @@ def download_url(url, save_path):
None
"""
with open(save_path, "wb") as f:
- print("Downloading {} from {}".format(save_path, url))
+ print("Downloading {} from {}...".format(save_path, url))
response = requests.get(url, stream=True)
total_length = response.headers.get("content-length")
@@ -31,7 +31,7 @@ def download_url(url, save_path):
done = int(50 * dl / total_length)
sys.stdout.write("\r[%s%s]" % ("=" * done, " " * (50 - done)))
sys.stdout.flush()
- print()
+ print("Download succes.\n")
# Function to unzip files
@@ -46,17 +46,7 @@ def unzip_file(path_to_zip_file, directory_to_extract_to):
Raises:
None
"""
- print("Unzip files..")
+ print("Unzip {} to {}...".format(path_to_zip_file, directory_to_extract_to))
with zipfile.ZipFile(path_to_zip_file, "r") as zip_ref:
zip_ref.extractall(directory_to_extract_to)
-
-
-if __name__ == "__main__":
- url = "https://archive.ics.uci.edu/ml/machine-learning-databases/00506/casas-dataset.zip"
- url_test = "https://archive.ics.uci.edu/ml/machine-learning-databases/00405/Postures.zip" # Smaller zip to test
- save_path = "../data/casas-dataset.zip"
- # Download zip file
- download_url(url, save_path)
- # Unzip it
- unzip_file(save_path, "../data_test/")
- print("Done")
+ print("Unzip succes.\n")
diff --git a/rr/experiment/__init__.py b/rr/experiment/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/rr/experiment/__main__.py b/rr/experiment/__main__.py
new file mode 100644
index 0000000..eb0c752
--- /dev/null
+++ b/rr/experiment/__main__.py
@@ -0,0 +1,51 @@
+import sys
+import argparse
+import configparser
+from rr.experiment import experiments, database
+import os.path
+
+
+def main():
+ parser = argparse.ArgumentParser(description="M05 mini-project: experiments")
+ parser.add_argument("datapath", type=str, help="Dataset file in .csv")
+ parser.add_argument("output", type=str, help="Destination folder for the results")
+ parser.add_argument(
+ "config", type=str, help="Filepath for experiments configuration file in .ini"
+ )
+ args = parser.parse_args()
+
+ config = configparser.ConfigParser()
+ config.read(args.config)
+
+ print(
+ "M05 mini-project on Human Activity Recognition with Random Forest classifier"
+ )
+ tabnum = 1
+ experiment_results = experiments.experiment_impact_nb_trees(
+ tabnum,
+ filepath=args.datapath,
+ nb_trees=[int(n) for n in config["nb_trees_experiment"]["nb_trees"].split(",")],
+ max_depth=int(config["nb_trees_experiment"]["tree_depth"]),
+ plot_path=args.output,
+ )
+
+ tabnum += len(config["nb_trees_experiment"]["nb_trees"].split(",")) * len(
+ database.PROTOCOLS
+ )
+ experiment_results += experiments.experiment_impact_tree_depth(
+ tabnum,
+ filepath=args.datapath,
+ nb_trees=int(config["tree_depth_experiment"]["nb_trees"]),
+ max_depths=[
+ int(d) for d in config["tree_depth_experiment"]["tree_depth"].split(",")
+ ],
+ plot_path=args.output,
+ )
+
+ with open(os.path.join(args.output, "experiment_results.txt"), "w+") as fout:
+ fout.write(experiment_results)
+ print("Experiments done\n")
+
+
+if __name__ == "__main__":
+ sys.exit(main())
diff --git a/rr/experiment/algorithm.py b/rr/experiment/algorithm.py
new file mode 100644
index 0000000..8173a02
--- /dev/null
+++ b/rr/experiment/algorithm.py
@@ -0,0 +1,51 @@
+from sklearn.ensemble import RandomForestClassifier
+
+import logging
+
+logger = logging.getLogger()
+
+
+class Model:
+ def __init__(self, nb_tree_per_forest=50, max_depth=10):
+ """Create a new ML model (Random forest classifier from scikitlearn)
+
+ Args:
+ nb_tree_per_forest (int): number of decision trees in the forest
+ max_depth (int): max depth of the trees
+ Returns:
+ None
+ Raises:
+ None
+ """
+ self.model = RandomForestClassifier(
+ n_estimators=nb_tree_per_forest, max_depth=max_depth, random_state=0
+ )
+
+ def train(self, X, y):
+ """Train the model using the given data
+
+ Args:
+ X (numpy.ndarray):A NxM 2D-array where each row corresponds to a sample and each column to a feature
+ y (numpy.ndarray): A 1D-array of length N, where each element corresponds to a sample label
+ Returns:
+ None
+ Raises:
+ None
+ """
+ self.model.fit(X, y)
+
+ def predict(self, X):
+ """Make a prediction on the data using the trained model
+
+ Args:
+ X (numpy.ndarray):A NxM 2D-array where each row corresponds to a sample and each column to a feature
+ Returns:
+ numpy.ndarray: A 1D array (with a dtype of int) containing the predicted
+ label for each sample
+
+ Raises:
+ None
+ """
+ prediction = self.model.predict(X)
+
+ return prediction
diff --git a/scripts/analysis.py b/rr/experiment/analysis.py
similarity index 91%
rename from scripts/analysis.py
rename to rr/experiment/analysis.py
index a5509b4..d42abe7 100644
--- a/scripts/analysis.py
+++ b/rr/experiment/analysis.py
@@ -1,7 +1,6 @@
import plotly.express as px
from sklearn.metrics import confusion_matrix
import numpy as np
-import itertools
def get_confusion_matrix(prediction_label, true_label):
@@ -9,7 +8,7 @@ def get_confusion_matrix(prediction_label, true_label):
Args:
prediction_label (list): Estimated targets as returned by a classifier.
- true_label (list): Ground truth (correct) target values.
+ true_label (list): Ground truth (correct) target values.
Returns:
numpy.ndarray: A 2D array (with a dtype of int) containing the confusion matrix.
Raises:
@@ -26,7 +25,7 @@ def plot_confusion_matrix(
Args:
cm (numpy.ndarray): A 2D array (with a dtype of int) containing the confusion matrix.
classes (numpy.ndarray): A 1D array (with a dtype of str) containing the lable name for each class
- normalize (boolean): Flag to normalize the data
+ normalize (boolean): Flag to normalize the data
title (str): The title of the plot
file_name (str): File name to export the graph
Returns:
@@ -34,7 +33,6 @@ def plot_confusion_matrix(
Raises:
None
"""
- # Normalize if wanted
if normalize:
cm = cm / np.sum(cm)
@@ -47,6 +45,4 @@ def plot_confusion_matrix(
color_continuous_scale="Blues",
title=title,
)
- # fig.show()
- # Export graph
fig.write_html("{}.html".format(file_name))
diff --git a/scripts/database.py b/rr/experiment/database.py
similarity index 99%
rename from scripts/database.py
rename to rr/experiment/database.py
index 854ac71..f7f756f 100644
--- a/scripts/database.py
+++ b/rr/experiment/database.py
@@ -1,4 +1,3 @@
-#!/usr/bin/env python
import numpy as np
import csv
from sklearn.model_selection import train_test_split
diff --git a/scripts/main.py b/rr/experiment/experiments.py
similarity index 56%
rename from scripts/main.py
rename to rr/experiment/experiments.py
index 1889345..c9aeb68 100644
--- a/scripts/main.py
+++ b/rr/experiment/experiments.py
@@ -1,7 +1,7 @@
-#!/usr/bin/env python
from tabulate import tabulate
-from scripts import algorithm, database, analysis, config
+from rr.experiment import algorithm, database, analysis
import numpy as np
+import os.path
def base_experiment(protocol, variables, filepath, nb_tree_per_forest=50, max_depth=10):
@@ -48,27 +48,30 @@ def pretty_confusion_matrix(cm):
return table
-def experiment_impact_nb_trees(tabnum, filepath, nb_trees, max_depth):
- """Evaluates and print the impact of the number of trees per forest on the classifiers performance
+def experiment_impact_nb_trees(tabnum, filepath, nb_trees, max_depth, plot_path):
+ """Evaluates the impact of the number of trees per forest on the classifiers performance
Args:
tabnum (int): first confusion matrix numbering
filepath (str): path to the file containing the dataset to load
nb_trees (list): list of number of trees to evaluate
+ max_depth (int): trees maximum depth
+ plot_path (str): folder where to store confusion matrix plots
Returns:
- None
+ str : experiment results
Raises:
None
"""
- print("\nImpact of number of trees per forest")
+ print("Starting experiment on number of trees impact...")
+ result = "\nImpact of number of trees per forest"
for n, p in enumerate(database.PROTOCOLS):
+ print("Processing `protocol` {}...".format(p))
for m, nb_tree_per_forest in enumerate(nb_trees):
- print(
- "\nTable {table_number}: Confusion matrix with {nb_trees} tree(s) for Protocol `{protocol}`".format(
- table_number=(n * len(nb_trees)) + m + tabnum,
- protocol=p,
- nb_trees=nb_tree_per_forest,
- )
+ num = (n * len(nb_trees)) + m + tabnum
+ result += "\nTable {table_number}: Confusion matrix with {nb_trees} tree(s) for Protocol `{protocol}`".format(
+ table_number=num,
+ protocol=p,
+ nb_trees=nb_tree_per_forest,
)
cm = base_experiment(
p,
@@ -77,29 +80,40 @@ def experiment_impact_nb_trees(tabnum, filepath, nb_trees, max_depth):
max_depth=max_depth,
filepath=filepath,
)
- print(pretty_confusion_matrix(cm))
+ result += pretty_confusion_matrix(cm)
+ analysis.plot_confusion_matrix(
+ cm,
+ database.CLASSES,
+ file_name=os.path.join(plot_path, "table_{}".format(num)),
+ )
+ print("Experiment completed\n")
+ return result
-def experiment_impact_tree_depth(tabnum, filepath, nb_trees, max_depths):
- """Evaluates and print the impact of the trees depth on the classifiers performance
+def experiment_impact_tree_depth(tabnum, filepath, nb_trees, max_depths, plot_path):
+ """Evaluates and the impact of the trees depth on the classifiers performance
Args:
tabnum (int): first confusion matrix numbering
filepath (str): path to the file containing the dataset to load
+ nb_trees (int): number of trees in forest
+ max_depths (list): list of trees maximum depths to evaluate
+ plot_path (str): folder where to store confusion matrix plots
Returns:
- None
+ str : experiment results
Raises:
None
"""
- print("\nImpact of trees maximum depth")
+ print("Starting experiment on tree depth impact...")
+ result = "\nImpact of trees maximum depth"
for n, p in enumerate(database.PROTOCOLS):
+ print("Processing `protocol` {}".format(p))
for m, max_depth in enumerate(max_depths):
- print(
- "\nTable {table_number}: Confusion matrix with trees maximum depth of {max_depth} for Protocol `{protocol}`".format(
- table_number=(n * len(max_depths)) + m + tabnum,
- protocol=p,
- max_depth=max_depth,
- )
+ num = (n * len(max_depths)) + m + tabnum
+ result += "\nTable {table_number}: Confusion matrix with trees maximum depth of {max_depth} for Protocol `{protocol}`".format(
+ table_number=num,
+ protocol=p,
+ max_depth=max_depth,
)
cm = base_experiment(
p,
@@ -108,22 +122,11 @@ def experiment_impact_tree_depth(tabnum, filepath, nb_trees, max_depths):
max_depth=max_depth,
filepath=filepath,
)
- print(pretty_confusion_matrix(cm))
-
-
-if __name__ == "__main__":
- print("Main script for Human Activity Recognition with Random Forest classifier")
- tabnum = 1
- experiment_impact_nb_trees(
- tabnum,
- filepath=config.data_path,
- nb_trees=config.nb_trees_experiment["nb_trees"],
- max_depth=config.nb_trees_experiment["tree_depth"],
- )
- tabnum += len(config.nb_trees_experiment["nb_trees"]) * len(database.PROTOCOLS)
- experiment_impact_tree_depth(
- tabnum,
- filepath=config.data_path,
- nb_trees=config.tree_depth_experiment["nb_trees"],
- max_depths=config.tree_depth_experiment["tree_depth"],
- )
+ result += pretty_confusion_matrix(cm)
+ analysis.plot_confusion_matrix(
+ cm,
+ database.CLASSES,
+ file_name=os.path.join(plot_path, "table_{}".format(num)),
+ )
+ print("Experiment completed\n")
+ return result
diff --git a/rr/tests/inputs/not_a_zip.txt b/rr/tests/inputs/not_a_zip.txt
new file mode 100644
index 0000000..e69de29
diff --git a/tests/inputs/simpleZip.zip b/rr/tests/inputs/simpleZip.zip
similarity index 100%
rename from tests/inputs/simpleZip.zip
rename to rr/tests/inputs/simpleZip.zip
diff --git a/tests/inputs/test_set.csv b/rr/tests/inputs/test_set.csv
similarity index 100%
rename from tests/inputs/test_set.csv
rename to rr/tests/inputs/test_set.csv
diff --git a/rr/tests/output/confusion_matrix.html b/rr/tests/output/confusion_matrix.html
new file mode 100644
index 0000000..692387c
--- /dev/null
+++ b/rr/tests/output/confusion_matrix.html
@@ -0,0 +1,67 @@
+
+
+
+
+
+
\ No newline at end of file
diff --git a/rr/tests/output/f1.txt b/rr/tests/output/f1.txt
new file mode 100644
index 0000000..e69de29
diff --git a/rr/tests/output/invalid_dl.zip b/rr/tests/output/invalid_dl.zip
new file mode 100644
index 0000000..e69de29
diff --git a/rr/tests/output/test_dl.zip b/rr/tests/output/test_dl.zip
new file mode 100644
index 0000000..914ee19
Binary files /dev/null and b/rr/tests/output/test_dl.zip differ
diff --git a/tests/test.py b/rr/tests/test.py
similarity index 97%
rename from tests/test.py
rename to rr/tests/test.py
index 12b846c..8c72882 100644
--- a/tests/test.py
+++ b/rr/tests/test.py
@@ -1,8 +1,9 @@
import os
import sys
-sys.path.insert(0, os.path.dirname(os.path.realpath(__file__)) + "/../scripts/")
-from scripts import algorithm, database, analysis, download_data
+sys.path.insert(0, os.path.dirname(os.path.realpath(__file__)) + "/../experiment/")
+from rr.experiment import database, analysis
+from rr.download_data import download_data
import zipfile
import requests
import numpy as np
@@ -98,7 +99,6 @@ def test_load():
def test_split_data_test():
datapath = base_path + "/inputs/test_set.csv"
data_len = 1225
- num_columns = len(database.VARIABLES)
protocol = "proto1"
subset = "test"
expected_test_len = np.ceil(database.PROTOCOLS[protocol][subset] * data_len)
diff --git a/scripts/algorithm.py b/scripts/algorithm.py
deleted file mode 100644
index 3df618b..0000000
--- a/scripts/algorithm.py
+++ /dev/null
@@ -1,54 +0,0 @@
-from sklearn.ensemble import RandomForestClassifier
-from sklearn.metrics import accuracy_score, confusion_matrix
-import numpy as np
-
-import logging
-
-logger = logging.getLogger()
-
-
-class Model:
- def __init__(self, nb_tree_per_forest=50, max_depth=10):
- """Create a new ML model (Random forest classifier from scikitlearn)
-
- Args:
- nb_tree_per_forest: number of decision trees in the forest
- max_depth: max depth of the trees
- Returns:
- None
- Raises:
- None
- """
- # Create a random forest model
- self.model = RandomForestClassifier(
- n_estimators=nb_tree_per_forest, max_depth=max_depth, random_state=0
- )
-
- def train(self, X, y):
- """Train the model using the given data
-
- Args:
- X (numpy.ndarray):A NxM 2D-array where each row corresponds to a sample and each column to a feature
- y (numpy.ndarray): A 1D-array of length N, where each element corresponds to a sample label
- Returns:
- None
- Raises:
- None
- """
- self.model.fit(X, y)
-
- def predict(self, X):
- """Make a prediction on the data using the trained model
-
- Args:
- X (numpy.ndarray):A NxM 2D-array where each row corresponds to a sample and each column to a feature
- Returns:
- numpy.ndarray: A 1D array (with a dtype of int) containing the predicted
- label for each sample
-
- Raises:
- None
- """
- prediction = self.model.predict(X)
-
- return prediction
diff --git a/scripts/config.py b/scripts/config.py
deleted file mode 100644
index 0ffb049..0000000
--- a/scripts/config.py
+++ /dev/null
@@ -1,3 +0,0 @@
-data_path = "../data/csh101/csh101.ann.features.csv"
-nb_trees_experiment = {"nb_trees": (1, 2), "tree_depth": 10}
-tree_depth_experiment = {"nb_trees": 10, "tree_depth": (1, 2)}
diff --git a/setup.py b/setup.py
new file mode 100644
index 0000000..a4cb444
--- /dev/null
+++ b/setup.py
@@ -0,0 +1,36 @@
+#!/usr/bin/env python
+# coding=utf-8
+
+from setuptools import setup, find_packages
+
+
+def load_requirements(f):
+ retval = [str(k.strip()) for k in open(f, "rt")]
+ return [k for k in retval if k and k[0] not in ("#", "-")]
+
+
+setup(
+ name="rr",
+ version="1.0.0",
+ description="Basic example of a Reproducible Research Project in Python",
+ url="https://github.com/sdevenes/M05_MiniProject",
+ license="MIT",
+ license_file="LICENSE",
+ author="Steve Devènes, Amara Spano",
+ author_email="steve.devenes@etu.unidistance.ch, amara.spano@etu.unidistance.ch",
+ long_description=open("README.rst").read(),
+ long_description_content_type="text/x-rst",
+ packages=find_packages(),
+ include_package_data=True,
+ install_requires=load_requirements("requirements.txt"),
+ entry_points={"console_scripts": ["rr-paper = rr.experiment.__main__:main"]},
+ classifiers=[
+ "Development Status :: 1 - Beta",
+ "Intended Audience :: Developers",
+ "License :: OSI Approved :: MIT License",
+ "Natural Language :: English",
+ "Programming Language :: Python",
+ "Programming Language :: Python :: 3",
+ "Topic :: Software Development :: Libraries :: Python Modules",
+ ],
+)