Merge pull request #139 from AlexandrovLab/container

Add Docker Container

Author: mdbarnesUCSD

CHANGELOG.md

@@ -1,4 +1,3 @@
-
 # Changelog
 
 All notable changes to this project will be documented in this file.
@@ -7,6 +6,14 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 
 ## [Unreleased]
 
+## [0.1.8] - 2024-08-20
+
+### Added
+- Added a Dockerfile to the repository for containerization. Documentation on how to use the Dockerfile needs to be added to the README.
+
+### Changed
+- Removed unnecessary imports from `setup.py` to clean up the codebase.
+
 ## [0.1.7] - 2024-06-21
 
 ### Added
@@ -18,5 +25,3 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 ### Changed
 - Improved input_type value check mechanism to use .lower() before checking against values 'vcf' or 'matrix'.
 - Updated dependency from PyPDF2 to pypdf to increase compatibility and resolve installation issues on bioconda.
-
-

Dockerfile

@@ -0,0 +1,31 @@
+# Start with a base Ubuntu image and install Python
+FROM ubuntu:22.04
+
+# Avoid prompts from apt
+ARG DEBIAN_FRONTEND=noninteractive
+
+# Install Python and other dependencies, and apply updates
+RUN apt-get update && apt-get upgrade -y && \
+    apt-get install -y python3-pip python3-dev git && \
+    apt-get clean && rm -rf /var/lib/apt/lists/*
+
+# Set the working directory in the container
+WORKDIR /usr/src/app
+
+# Clone the specific branch of the repository
+RUN git clone https://github.com/AlexandrovLab/SigProfilerAssignment.git .
+
+# Install the dependencies from the requirements.txt in the cloned repository
+RUN pip3 install --no-cache-dir -r requirements.txt
+
+# Install the SigProfilerAssignment package from PyPI
+RUN pip3 install SigProfilerAssignment==0.1.8
+
+# Create a non-root user named 'spm_user'
+RUN useradd -m -s /bin/bash spm_user
+
+# Change the ownership of the /usr/src/app directory and its contents to the new non-root user
+RUN chown -R spm_user:spm_user /usr/src/app
+
+# Switch to the non-root user for subsequent commands and when running the container
+USER spm_user

SigProfilerAssignment/decompose_subroutines.py

@@ -9,19 +9,12 @@
 import numpy as np
 import os, sys
 
-# from matplot,pdblib.backends.backend_pdf import PdfPages
 import pandas as pd
 import matplotlib.pyplot as plt
 
 plt.switch_backend("agg")
 from matplotlib.backends.backend_pdf import PdfPages
 
-# from sklearn import metrics
-# import time
-# import multiprocessing
-# from multiprocessing import current_process
-# from functools import partial
-# from numpy import linalg as LA
 import sigProfilerPlotting as plot
 from SigProfilerAssignment.DecompositionPlots import PlotDecomposition as sp
 from sigProfilerPlotting import plotActivity as plot_ac
@@ -30,7 +23,6 @@
 import pypdf
 import scipy
 
-# import SigProfilerAssignment as sspro
 from pypdf import PdfMerger
 import SigProfilerAssignment as spa
 from SigProfilerAssignment import single_sample as ss

SigProfilerAssignment/decomposition.py

@@ -6,13 +6,10 @@
 @author: mishugeb
 """
 
-# from SigProfilerExtractor import subroutines as sub
-
 from cmath import cos
 import datetime
 import platform
 
-# from torch import sign
 from SigProfilerAssignment import decompose_subroutines as sub
 from SigProfilerAssignment.DecompositionPlots import PlotDecomposition as plot_decomp
 import SigProfilerAssignment

requirements.txt

@@ -1,12 +1,94 @@
-scipy
-numpy
-pandas 
-SigProfilerMatrixGenerator 
-sigProfilerPlotting 
-pillow
-statsmodels
-scikit-learn
-psutil
-reportlab
-pypdf
-alive_progress
+#
+# This file is autogenerated by pip-compile with Python 3.11
+# by the following command:
+#
+#    pip-compile setup.py
+#
+about-time==4.2.1
+    # via alive-progress
+alive-progress==3.1.5
+    # via SigProfilerAssignment (setup.py)
+chardet==5.2.0
+    # via reportlab
+contourpy==1.2.1
+    # via matplotlib
+cycler==0.12.1
+    # via matplotlib
+fonttools==4.53.1
+    # via matplotlib
+grapheme==0.6.0
+    # via alive-progress
+joblib==1.4.2
+    # via scikit-learn
+kiwisolver==1.4.5
+    # via matplotlib
+matplotlib==3.9.2
+    # via
+    #   sigprofilermatrixgenerator
+    #   sigprofilerplotting
+numpy==1.26.4
+    # via
+    #   SigProfilerAssignment (setup.py)
+    #   contourpy
+    #   matplotlib
+    #   pandas
+    #   patsy
+    #   scikit-learn
+    #   scipy
+    #   sigprofilermatrixgenerator
+    #   statsmodels
+packaging==24.1
+    # via
+    #   matplotlib
+    #   statsmodels
+pandas==1.5.3
+    # via
+    #   SigProfilerAssignment (setup.py)
+    #   sigprofilermatrixgenerator
+    #   sigprofilerplotting
+    #   statsmodels
+patsy==0.5.6
+    # via statsmodels
+pillow==10.4.0
+    # via
+    #   matplotlib
+    #   reportlab
+    #   sigprofilerplotting
+pymupdf==1.24.9
+    # via SigProfilerAssignment (setup.py)
+pymupdfb==1.24.9
+    # via pymupdf
+pyparsing==3.1.2
+    # via matplotlib
+pypdf==4.3.1
+    # via SigProfilerAssignment (setup.py)
+python-dateutil==2.9.0.post0
+    # via
+    #   matplotlib
+    #   pandas
+pytz==2024.1
+    # via pandas
+reportlab==4.2.2
+    # via SigProfilerAssignment (setup.py)
+scikit-learn==1.5.1
+    # via sigprofilerplotting
+scipy==1.14.0
+    # via
+    #   SigProfilerAssignment (setup.py)
+    #   scikit-learn
+    #   sigprofilermatrixgenerator
+    #   statsmodels
+sigprofilermatrixgenerator==1.2.28
+    # via SigProfilerAssignment (setup.py)
+sigprofilerplotting==1.3.24
+    # via
+    #   SigProfilerAssignment (setup.py)
+    #   sigprofilermatrixgenerator
+six==1.16.0
+    # via
+    #   patsy
+    #   python-dateutil
+statsmodels==0.14.2
+    # via sigprofilermatrixgenerator
+threadpoolctl==3.5.0
+    # via scikit-learn

setup.py

@@ -6,7 +6,7 @@
 if os.path.exists("dist"):
     shutil.rmtree("dist")
 
-VERSION = "0.1.7"
+VERSION = "0.1.8"
 
 
 def write_version_py(filename="SigProfilerAssignment/version.py"):
@@ -15,7 +15,7 @@ def write_version_py(filename="SigProfilerAssignment/version.py"):
 # THIS FILE IS GENERATED FROM SigProfilerAssignment SETUP.PY
 short_version = '%(version)s'
 version = '%(version)s'
-Update = 'v0.1.7: Update CLI, add pytest, and update pypdf dependency.'
+Update = 'v0.1.8: Add Dockerfile and remove uncessary dependencies'
 
     
     """
@@ -34,18 +34,14 @@ def write_version_py(filename="SigProfilerAssignment/version.py"):
 
 requirements = [
     "scipy>=1.6.3",
-    "numpy>=1.21.2",
+    "numpy>=1.21.2,<2.0.0",
     "pandas>=1.2.4,<2.0.0",
-    "SigProfilerMatrixGenerator>=1.2.17",
-    "sigProfilerPlotting>=1.3.23",
-    "statsmodels>=0.9.0",
-    "scikit-learn>=0.24.2",
-    "psutil>=5.6.1",
+    "SigProfilerMatrixGenerator>=1.2.28",
+    "sigProfilerPlotting>=1.3.24",
     "reportlab>=3.5.42",
     "pypdf>=3.1.0",
     "alive_progress>=2.4.1",
-    "pdf2image>=1.16.0",
-    "PyMuPDF>=1.21.0",
+    "PyMuPDF>=1.21.0",  # required for package "fitz"
 ]
 
 write_version_py()