update README

alvinthai · Mar 4, 2018 · 1320f71 · 1320f71
1 parent 186c02b
commit 1320f71
Show file tree

Hide file tree

Showing 3 changed files with 37 additions and 8 deletions.
diff --git a/README.md b/README.md
@@ -1,2 +1,31 @@
-# histoviewer
-Dynamic histograms for pandas DataFrames
+## Introduction
+Histoviewer is an interactive bokeh application for dynamically viewing histograms and bar charts for a pandas DataFrame input.
+
+Try out the application here (it may take up to a minute to start the heroku app):  
+https://histoviewer.herokuapp.com/sample_app
+
+Play around with the sliders, text inputs, dropdowns, and button widgets to update the data dynamically!
+
+## Features
+Histoviewer uses bokeh widgets to interact with the displayed plots. The column selection dropdown changes the data source to **df[column]**.
+
+- Continuous variables
+  - dynamically rebin histogram from 10-100 bins
+  - dynamically rescale minimum/maximum bounds of histogram
+- Categorical variables
+  - resort by y-axis (descending) or x-axis (alphabetical)
+  - reindex x-axis to filter categories
+
+## Getting Started
+To use Histoviewer, simply install the required dependencies:  
+```
+pip install -r requirements.txt
+```
+
+Then run the following code (in Jupyter notebook) to interact with a pandas DataFrame:
+```
+from histoviewer import histoviewer
+histoviewer(df)
+```
+
+For an example of how to run histoviewer from a python 2.7 script, see [sample_app.py](https://github.com/alvinthai/histoviewer/blob/master/sample_app.py).
diff --git a/histoviewer.py b/histoviewer.py
@@ -20,11 +20,11 @@ def histoviewer(df, default_col=None, serve='notebook'):
     pandas DataFrame input.
 
     Histograms (continuous variables) have interactive inputs for:
-        - rebinning from 10-100 Bins
-        - rescaling minimum and maximum bounds of axis
+        - rebinning histogram from 10-100 bins
+        - rescaling minimum/maximum bounds of histogram
     Bar Charts (categorical variables) have interactive inputs for:
-        - sorting by y-axis (descending) or x-axis (ascending)
-        - reindexing x-axis for hiding categories
+        - resorting by y-axis (descending) or x-axis (alphabetical)
+        - reindexing x-axis to filter categories
 
     Default serve mode assumes jupyter notebook. Histoviewer application could
     also be served from a python script via:

diff --git a/requirements.txt b/requirements.txt
@@ -1,3 +1,3 @@
 bokeh==0.12.14
-numpy==1.14.1
-pandas==0.22.0
+numpy>=1.14.1
+pandas>=0.22.0