Updated installation instructions

pythoncourse/installation/create_conda_env.bat

@@ -13,7 +13,7 @@ call conda activate py36class
 REM Install Tensorflow, PyTorch and Anaconda (lots of pacakged)
 REM Pandas 0.24.2 is needed for some packages and scikit-learn 0.20.2
 REM only if you have GPU below 2 lines instead of CPU versions
-REM conda install tensorflow-gpu=2.1.0 anaconda pandas=0.24.2 scikit-learn=0.20.2 --yes
+REM conda install anaconda tensorflow-gpu=2.1.0 anaconda pandas=0.24.2 scikit-learn=0.20.2 --yes
 REM conda install pytorch torchvision cudatoolkit=10.1 -c pytorch --yes
 call conda install anaconda tensorflow=2.1.0 anaconda pandas=0.24.2 scikit-learn=0.20.2 --yes
 call conda install pytorch torchvision cpuonly -c pytorch --yes

pythoncourse/installation/create_conda_env.sh

@@ -18,7 +18,7 @@ source activate py36class
 # Install Tensorflow, PyTorch and Anaconda (lots of pacakged)
 # Pandas 0.24.2 is needed for some packages and scikit-learn 0.20.2
 # only if you have GPU below 2 lines instead of CPU versions
-# conda install tensorflow-gpu=2.1.0 anaconda pandas=0.24.2 scikit-learn=0.20.2 --yes
+# conda install anaconda tensorflow-gpu=2.1.0 anaconda pandas=0.24.2 scikit-learn=0.20.2 --yes
 # conda install pytorch torchvision cudatoolkit=10.1 -c pytorch --yes
 conda install anaconda tensorflow=2.1.0 anaconda pandas=0.24.2 scikit-learn=0.20.2 --yes
 conda install pytorch torchvision cpuonly -c pytorch --yes

pythoncourse/installation/installing_anaconda_and_pycharm.html

@@ -13183,7 +13183,7 @@ <h2 id="Installing-tabula-py-and-pytesseract">Installing tabula-py and pytessera
 <div class="cell border-box-sizing text_cell rendered"><div class="prompt input_prompt">
 </div><div class="inner_cell">
 <div class="text_cell_render border-box-sizing rendered_html">
-<h2 id="Installing-a-conda-environment-for-Windows-from-YML-file-(quickest-method---recommended)">Installing a conda environment for Windows from YML file (quickest method - recommended)<a class="anchor-link" href="#Installing-a-conda-environment-for-Windows-from-YML-file-(quickest-method---recommended)">&#182;</a></h2><p>We can download <a href="https://raw.githubusercontent.com/cuemacro/teaching/master/pythoncourse/installation/environment_windows.yml">environment_windows.yml (click to download)</a> and use that to create our conda environment with all the necessary packages. This also installs the exact same libraries versions that I have.</p>
+<h2 id="Installing-a-conda-environment-for-Windows-from-YML-file-(quickest-method---recommended)">Installing a conda environment for Windows from YML file (quickest method - recommended)<a class="anchor-link" href="#Installing-a-conda-environment-for-Windows-from-YML-file-(quickest-method---recommended)">&#182;</a></h2><p>We can download <a href="https://raw.githubusercontent.com/cuemacro/teaching/master/pythoncourse/installation/environment_windows.yml">environment_windows.yml (click to download)</a> and use that to create our conda environment with all the necessary packages. This also installs the exact same libraries versions that I have (and reduces the likelihood of version conflicts between libraries).</p>
 <ul>
 <li>Open up the Anaconda Prompt (should be in the Start Menu)</li>
 <li><p>In this prompt, your Anaconda folder will be on the path (ie. it will recognise where <code>conda</code> is installed etc.)</p>
@@ -13244,28 +13244,31 @@ <h3 id="Copy-the-below-to-create_conda_env.bat-for-Windows-(or-download-it-from-
 
 REM Remove any existing environment called py36class, and create a py36class with anaconda packages
 call conda remove -n py36class --all --yes
-call conda create -n py36class python=3.6 anaconda
+call conda create -n py36class python=3.6
 call conda activate py36class
 
-REM Install scikit-learn (downgrade for some later dependencies) - xlwings for Windows
-call conda install anaconda scikit-learn=0.20.2 --yes
-call conda install -c conda-forge xlwings=0.19.4 --yes
-
-REM Install PyTorch &amp; Tensorflow
+REM Install Tensorflow, PyTorch and Anaconda (lots of pacakged)
+REM Pandas 0.24.2 is needed for some packages and scikit-learn 0.20.2
 REM only if you have GPU below 2 lines instead of CPU versions
-REM call conda install tensorflow-gpu=2.1.0
-REM call conda install pytorch torchvision cudatoolkit=10.1 -c pytorch --yes
-call conda install anaconda tensorflow=2.1.0 --yes
+REM conda install anaconda tensorflow-gpu=2.1.0 anaconda pandas=0.24.2 scikit-learn=0.20.2 --yes
+REM conda install pytorch torchvision cudatoolkit=10.1 -c pytorch --yes
+call conda install anaconda tensorflow=2.1.0 anaconda pandas=0.24.2 scikit-learn=0.20.2 --yes
 call conda install pytorch torchvision cpuonly -c pytorch --yes
 
+REM OpenCV and PyArrow need to be installed earlier
+call conda install -c conda-forge pyarrow=0.17.1 opencv --yes
+
+REM xlwings only works for Mac or Windows
+call conda install -c conda-forge xlwings=0.19.4 --yes
+
 REM Install from conda-forge (pyarrow is a newer version!)
 call conda install -c conda-forge ^
  jupyter_contrib_nbextensions jupyter_nbextensions_configurator ^
  redis-py python-blosc pathos graphviz python-graphviz textblob ^
  vaex-core vaex-viz vaex-jupyter vaex-arrow vaex-server vaex-hdf5 vaex-astro vaex-distributed ^
  pyspark=2.4.0 koalas ^
  spacy fastparquet python-snappy nodejs pystan fbprophet setuptools-git ^
- vispy datashader pyproj holoviews streamz quandl bqplot opencv s3fs blpapi ^
+ vispy datashader pyproj holoviews streamz quandl bqplot blpapi ^
  --yes
 
 REM Install database
@@ -13274,35 +13277,32 @@ <h3 id="Copy-the-below-to-create_conda_env.bat-for-Windows-(or-download-it-from-
 REM Install transformers/huggingface and table libraries
 REM Install graphics/plotting libraries
 REM Install findatapy, chartpy and findatapy
+REM opencv-contrib-python==4.2.0.34 \
 call pip install arctic==1.79.2 ^
  celery==4.4.0 celery[redis] celery[msgpack] msgpack-python ^
- fxcmpy alpha_vantage yfinance twython newspaper3k seasonal pdfminer.six vaderSentiment gensim wordcloud rise requests_html ^
+ fxcmpy alpha_vantage yfinance twython newspaper3k seasonal pdfminer.six vaderSentiment gensim wordcloud RISE requests_html ^
  textacy==0.8.0 ^
  transformers pytesseract cmdstanpy==0.4 tabula-py==1.4.3 ^
  cvlib ^
  cufflinks==0.17.3 plotly==4.8.0 ^
- dash==1.12.0 dash-html-components==1.0.3 dash-core-components==1.10.0 dash-table==4.7.0 jupyter-dash==0.2.1 dtale==1.8.1 progressbar2==3.38.0 ^
+ dash==1.12.0 dash-html-components==1.0.3 dash-core-components==1.10.0 dash-table==4.7.0 jupyter-dash==0.2.1 ^
+ dtale==1.8.1 progressbar2==3.38.0 ^
  finmarketpy chartpy findatapy
 
 REM To be able to plot Plotly into PNG or JPG
 call conda install -c plotly plotly-orca --yes
 
-REM In case has been installed elsewhere by pip
-call pip uninstall pandas
-
-REM Newer versions of Pandas aren't supported as well by other libraries eg. Arctic so stick to 0.24.2
-call conda install -c anaconda pandas=0.24.2 --yes
-call conda install -c conda-forge pyarrow=0.17.1 --yes
-
 REM Jupyter libraries
 call jupyter contrib nbextension install --user
 call jupyter nbextension enable execute_time/ExecuteTime
 call jupyter-nbextension install rise --py --sys-prefix
-call jupyter labextension install @jupyter-widgets/jupyterlab-manager --no-build
-call jupyter labextension install plotlywidget --no-build
-call jupyter labextension install jupyterlab-plotly --no-build
-call jupyter labextension install bqplot
-call jupyter lab build</code></pre>
+
+REM some of these extensions are not compatible with latest Jupyterlab
+REM call jupyter labextension install @jupyter-widgets/jupyterlab-manager --no-build
+REM call jupyter labextension install plotlywidget --no-build
+REM call jupyter labextension install jupyterlab-plotly --no-build
+REM call jupyter labextension install bqplot --no-build
+REM call jupyter lab build</code></pre>
 
 </div>
 </div>
@@ -13321,7 +13321,7 @@ <h2 id="Installing-xlwings-addin-in-Excel-on-Windows-(or-Mac)">Installing xlwing
 <div class="cell border-box-sizing text_cell rendered"><div class="prompt input_prompt">
 </div><div class="inner_cell">
 <div class="text_cell_render border-box-sizing rendered_html">
-<h2 id="Installing-a-conda-environment-for-Linux-or-Mac-from-YAML-file-(quickest-method---recommended)">Installing a conda environment for Linux or Mac from YAML file (quickest method - recommended)<a class="anchor-link" href="#Installing-a-conda-environment-for-Linux-or-Mac-from-YAML-file-(quickest-method---recommended)">&#182;</a></h2><p>We can download <a href="https://raw.githubusercontent.com/cuemacro/teaching/master/pythoncourse/installation/environment_linux.yml">environment_linux.yml (click to download)</a> and use that to create our conda environment with all the necessary packages. This should be a lot faster to run. This also installs the exact same libraries versions that I have.</p>
+<h2 id="Installing-a-conda-environment-for-Linux-or-Mac-from-YAML-file-(quickest-method---recommended)">Installing a conda environment for Linux or Mac from YAML file (quickest method - recommended)<a class="anchor-link" href="#Installing-a-conda-environment-for-Linux-or-Mac-from-YAML-file-(quickest-method---recommended)">&#182;</a></h2><p>We can download <a href="https://raw.githubusercontent.com/cuemacro/teaching/master/pythoncourse/installation/environment_linux.yml">environment_linux.yml (click to download)</a> and use that to create our conda environment with all the necessary packages. This should be a lot faster to run. This also installs the exact same libraries versions that I have (and reduces the likelihood of version conflicts between libraries).</p>
 <ul>
 <li>Open up a terminal window (making sure that Anaconda is on your path)</li>
 <li><p>In this prompt, your Anaconda folder will be on the path (ie. it will recognise where <code>conda</code> is installed etc.)</p>
@@ -13405,28 +13405,31 @@ <h3 id="Copy-the-below-to-create_conda_env.sh-(or-download-it-from-GitHub)">Copy
 
 # Remove any existing environment called py36class, and create a py36class with anaconda packages
 conda remove -n py36class --all --yes
-conda create -n py36class python=3.6 anaconda
+conda create -n py36class python=3.6
 source activate py36class
 
-# Install scikit-learn (downgrade for some later dependencies) - xlwings only works for Mac (won't install on Linux)
-conda install anaconda scikit-learn=0.20.2 --yes
-conda install conda-forge xlwings=0.19.4 --yes
-
-# Install PyTorch &amp; Tensorflow
+# Install Tensorflow, PyTorch and Anaconda (lots of pacakged)
+# Pandas 0.24.2 is needed for some packages and scikit-learn 0.20.2
 # only if you have GPU below 2 lines instead of CPU versions
-# conda install tensorflow-gpu=2.1.0
+# conda install anaconda tensorflow-gpu=2.1.0 anaconda pandas=0.24.2 scikit-learn=0.20.2 --yes
 # conda install pytorch torchvision cudatoolkit=10.1 -c pytorch --yes
-conda install anaconda tensorflow=2.1.0 --yes
+conda install anaconda tensorflow=2.1.0 anaconda pandas=0.24.2 scikit-learn=0.20.2 --yes
 conda install pytorch torchvision cpuonly -c pytorch --yes
 
+# OpenCV and PyArrow need to be installed earlier
+conda install -c conda-forge pyarrow=0.17.1 opencv --yes
+
+# xlwings only works for Mac
+conda install -c conda-forge xlwings=0.19.4 --yes
+
 # Install from conda-forge (pyarrow is a newer version!)
 conda install -c conda-forge \
  jupyter_contrib_nbextensions jupyter_nbextensions_configurator \
  redis-py python-blosc pathos graphviz python-graphviz textblob \
  vaex-core vaex-viz vaex-jupyter vaex-arrow vaex-server vaex-hdf5 vaex-astro vaex-distributed \
- pyspark=2.4.0 koalas \
  spacy fastparquet python-snappy nodejs pystan fbprophet setuptools-git \
- vispy datashader pyproj holoviews streamz quandl bqplot opencv s3fs \
+ pyspark=2.4.0 koalas \
+ vispy datashader pyproj holoviews streamz quandl bqplot \
  --yes
 
 # Install database
@@ -13436,26 +13439,26 @@ <h3 id="Copy-the-below-to-create_conda_env.sh-(or-download-it-from-GitHub)">Copy
 # Install graphics/plotting libraries
 # Install findatapy, chartpy and findatapy
 # Install modin (later versions need later versions of pandas, may downgrade to pandas to 0.24.2
+# opencv-contrib-python==4.2.0.34 \
 pip install arctic==1.79.2 \
  celery==4.4.0 celery[redis] celery[msgpack] msgpack-python \
- fxcmpy alpha_vantage yfinance twython newspaper3k seasonal pdfminer.six vaderSentiment gensim wordcloud rise requests_html \
+ fxcmpy alpha_vantage yfinance twython newspaper3k seasonal pdfminer.six vaderSentiment gensim wordcloud RISE requests_html \
  textacy==0.8.0 \
  transformers pytesseract cmdstanpy==0.4 tabula-py==1.4.3 \
  cvlib \
  cufflinks==0.17.3 plotly==4.8.0 \
- dash==1.12.0 dash-html-components==1.0.3 dash-core-components==1.10.0 dash-table==4.7.0 jupyter-dash==0.2.1 dtale==1.8.1 progressbar2==3.38.0 \
- finmarketpy chartpy findatapy \
- modin==0.5.4
+ dash==1.12.0 dash-html-components==1.0.3 dash-core-components==1.10.0 dash-table==4.7.0 jupyter-dash==0.2.1 \
+  dtale==1.8.1 progressbar2==3.38.0 \
+ finmarketpy chartpy findatapy
 
 # To be able to plot Plotly into PNG or JPG
 conda install -c plotly plotly-orca --yes
 
-# In case has been installed elsewhere by pip
-pip uninstall pandas
+# Note that modin may not be compatible with other libraries, so it is recommended
+# you install it in its own environment if you do want to use it
+# pip install modin==0.5.4
 
-# Newer versions of Pandas aren't supported as well by other libraries (eg. Arctic), so stick to 0.24.2
-conda install -c anaconda pandas=0.24.2 --yes
-conda install -c conda-forge pyarrow=0.17.1 --yes
+# conda install -c conda-forge pyarrow=0.17.1 --yes
 
 # GPU libraries for working with DataFrames
 # conda install -c rapidsai -c nvidia -c conda-forge -c defaults rapids=0.10 python=3.6
@@ -13464,11 +13467,13 @@ <h3 id="Copy-the-below-to-create_conda_env.sh-(or-download-it-from-GitHub)">Copy
 jupyter contrib nbextension install --user # to activate js on Jupyter
 jupyter nbextension enable execute_time/ExecuteTime
 jupyter-nbextension install rise --py --sys-prefix
-jupyter labextension install @jupyter-widgets/jupyterlab-manager --no-build
-jupyter labextension install plotlywidget --no-build
-jupyter labextension install jupyterlab-plotly --no-build
-jupyter labextension install bqplot
-jupyter lab build</code></pre>
+
+# Some of these extensions are not compatible with latest Jupyterlab
+# jupyter labextension install @jupyter-widgets/jupyterlab-manager --no-build
+# jupyter labextension install plotlywidget --no-build
+# jupyter labextension install jupyterlab-plotly --no-build
+# jupyter labextension install bqplot
+# jupyter lab build</code></pre>
 
 </div>
 </div>