Skip to main content

Explabox demo for the UCI drug reviews dataset

Project description

explabox-logo

explabox-demo-drugreview

Welcome to the demo of the explabox on the UCI Drug Reviews dataset. To speed up the demo, we made a smaller subset of the train and test dataset. The demo also includes a pretrained black-box classifier, which aims to predict whether a review in the text got a rating of negative (1-5), neutral (5-6) or positive (6-10).

To start the demo, you require:

Install the demo via:

  • pip3 install explabox
  • pip3 install explabox-demo-drugreview

Importing the necessary files

To start the demo, open your Jupyter Notebook and run the following line:

from explabox_demo_drugreview import model, dataset_file 

The dataset_file is the location of the dataset (drugsCom.zip), containing a train split (drugsComTrain.tsv) and test split (drugsComTest.tsv). You can import this dataset with the explabox with the data in the column review and labels in the column rating:

from explabox import import_data
data = import_data(dataset_file, data_cols='review', label_cols='rating')  

The model can directly be imported as-is. Make sure you explicitly include that drugsComTrain.tsv includes the train split and drugsComTest.tsv the test split of the data:

from explabox import Explabox

box = Explabox(data=data,
               model=model,
               splits={'train': 'drugsComTrain.tsv', 'test': 'drugsComTest.tsv'})

Now you are ready to .explore, .examine, .expose and .explain with the explabox!

Documentation

Having trouble? Want to know which functionalities the explabox includes? Check out the documentation at https://explabox.rtfd.io.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

explabox-demo-drugreview-0.9b10.tar.gz (420.1 kB view hashes)

Uploaded Source

Built Distribution

explabox_demo_drugreview-0.9b10-py3-none-any.whl (412.4 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page