nlp-specialization---coursera-deeplearning.ai

Natural Language Processing Specialization - Coursera/Deeplearning.ai

Creating python environment

Install Anaconda from https://www.anaconda.com/products/distribution

conda create -n coursera_nlp python=3.9
conda activate coursera_nlp

In order to find dependencies for Python environment, clone some popular Github repos from others who have taken the course.

# this one seems to have the best original files so can perform exercises on my own
git clone git@github.com:amanjeetsahu/Natural-Language-Processing-Specialization.git
# alternatives. all of these are 4 years old...
# git clone git@github.com:y33-j3T/Coursera-Deep-Learning.git
# git clone git@github.com:ibrahimjelliti/Deeplearning.ai-Natural-Language-Processing-Specialization.git
# git clone git@github.com:amanchadha/coursera-natural-language-processing-specialization.git
# search all notebooks and py files for import statements
find ./ -name *.py -o -name *.ipynb -print0 | xargs -0 egrep -sh '^\s*"import.*' | sed 's:#.*$::g' | sed 's/[[:space:]]*$//g' | sort | uniq

"import ast\n",
"import base64\n",
"import emoji\n",
"import functools\n",
"import gensim\n",
"import gin\n",
"import itertools\n",
"import jax\n",
"import json\n",
"import math
...

# after a bit of manual curation and removal of python provided libraries left with this
conda install gensim jax matplotlib nltk numpy pandas sacrebleu scipy sentencepiece scikit-learn
# now install some libraries that I like
conda install pytest
conda install -c conda-forge seaborn spacy jupytext jupyterlab

Not yet sure if I'll need these:

# nltk setup
python -m nltk.downloader popular
# spacy setup
python -m spacy download en_core_web_trf
python -m spacy download en_core_web_sm

Course 1: Natural Language Processing with Classification and Vector Spaces

Lecture notes available at https://community.deeplearning.ai/t/nlp-course-1-lecture-notes/64244 (required free account through link https://community.deeplearning.ai/invites/YaM2nrk6Xx#!)

Slides from each week are available.

For each week, I have a fiddled folder that I store my changes/edits in. Original labs and assignments are in the week X folders

Week 1: Sentiment Analysis with Logistic Regression

Lab 1 - Natural Language preprocessing
Lab 2 - Visualizing word frequencies
Lab 3 - Visualizing tweets and Logistic Regression models
Assignment 1
Assignment 1 Solution

Week 2: Sentiment Analysis with Naïve Bayes

Week 3: Vector Space Models

Week 4: Machine Translation and Document Search

lab_1_natural_language_preprocessing lab_2_visualizing_word_frequencies lab_3_visualizing_tweets_and_logistic_regression_models

Path to copy to /Users/seth/Projects/nlp-specialization---coursera-deeplearning.ai/course_1_nlp_with_classification_and_vector_spaces/week_1_sentiment_analysis_with_logistic_regression/

magic-lantern / nlp-specialization---coursera-deeplearning.ai