PragueML-Gensim-workshop-setup-sheet.md

ML Prague: Gensim workshop setup sheet

Please bring your laptop, the workshop is interactive!

Follow the instructions below before coming, to minimize installation delays and make sure you’re ready for our Friday workshop.

Requirements

• Python 3.5 preferred (but 2.7 should work too)
• Install dependencies, download datasets and models:

  pip install gensim==3.4.0 plotly==2.5.0 scikit-learn==0.19.1 jupyter==1.0.0 pandas==0.22 matplotlib==2.2.0
  python -m gensim.downloader --download text8
  python -m gensim.downloader --download 20-newsgroups
  python -m gensim.downloader --download fasttext-wiki-news-subwords-300
  python -m gensim.downloader --download word2vec-google-news-300

• Fetch the workshop notebooks (will be available on March 22):

  git clone https://github.com/RaRe-Technologies/talks.git
  cd talks/2018-03-23_MLPrague-workshop/

Sanity check

Run the commands below to verify that you have installed all the dependencies successfully:

• Сheck dependencies (shouldn’t raise an exception): python -c "import gensim, sklearn, plotly, jupyter, pandas, matplotlib"
• Check datasets & models (should print True): python -c "import os; from gensim.downloader import base_dir; print(all(os.path.exists(path) for path in [os.path.join(base_dir, '20-newsgroups', '20-newsgroups.gz'), os.path.join(base_dir, 'text8', 'text8.gz'), os.path.join(base_dir, 'fasttext-wiki-news-subwords-300', 'fasttext-wiki-news-subwords-300.gz'), os.path.join(base_dir, 'word2vec-google-news-300', 'word2vec-google-news-300.gz')]))"

If in doubt, feel free to reach out at [email protected].

See you soon! 🙂