erfanMhi/cs224n_Assignments

Stanford University cs224n Assignments solutions

CS224N Assignments

Solutions for cs224n(Deep Learning For Natural Language Processing) course by stanford university

Prerequisites

1. Install Anaconda
2. go to assignmentX where X is either 1, 2, 3 using a Terminal:

   $ cd \path\to\assignment1

3. create a python 2.7 environnment using

   $ conda env -n cs224n python=2.7 anaconda

4. activate your environment using (add source before activate if you're working with Linux/Mac)

   $ activate cs224n

5. install the dependencies using requirements.txt

   $ pip install -r requirements.txt

6. Don't forget to deactivate your environment when you're done (add source before deactivate if you're on Linux/Mac)

   $ deactivate cs224n

Deployment

Assignment 1

Word2vec

Steps for training Word2vec in assignment 1 :

1. Go to folder /assignment 1
2. Run get_datasets.sh(By running this file you can download datasets)
3. Open a terminal and write(Be careful, for this assignment you should have installed python2.7) :

   $ python q3_run.py

Sentiment Analysis(before you running this section you should download datasets as i mentioned in Word2vec section)

1. If you want to use logistic regression classifier on word vectors that we trained in Word2vec section you can use this code :

   $ cd path/to/assignment1
   $ python q4_sentiment.py --yourvectors

2. If you want to use pretrained GLOVE model you can use this code :

   $ cd path/to/assignment1
   $ python q4_sentiment.py --pretrained

Results

Assignment 1

Word2vec

Sentiment analysis train/dev accuracy

Sentiment analysis truth table

Built With

• Numpy - A library for doing optimized matrix operations
• Matplotlib - A library for visulizing our outputs as graphs
• scipy - A Library for doing mathematic, science and engineering stuff.
• sklearn - Simple and efficient Library for data mining and data analysis

License

This project is licensed under the MIT License - see the LICENSE.md file for details