Ideal Prediction reporting framework

Idealreport translates data from pandas DataFrames to HTML presentations, including plots, tables, and text. The HTML is rendered using javascript libraries based on D3.js so the resulting graphs are aesthetically pleasing.

For your python install, we recommend miniconda: https://conda.io/miniconda.html which enables setting up segregated environments.

For example, for mac:

1. Install miniconda https://conda.io/docs/user-guide/install/index.html
2. Follow the instructions to configure the conda script https://conda.io/docs/user-guide/install/macos.html
3. Create an environment https://conda.io/docs/user-guide/getting-started.html conda create --name py2
4. Activate the environment

source activate py2  # linux, mac
activate py2         # window
pip install sphinx, htmltag, pandas
python

Option 1: Pip is the easiest install method. The module is built as a python wheel so it easily works across various OSs. This method will only get the most recent tagged release, which is not necessarily the most recent version of the code.

pip install idealreport

Option 2: For the latest code, clone the repository.

git clone https://github.com/idealprediction/idealreport

• required: htmltag (requires sphinx), pandas
• important note: install sphinx before installing htmltag
• recommended: phantomjs to generate PDFs from HTML

pip install sphinx, htmltag, pandas 

See sample_plots.py for examples of how to create plots using the Reporter class (inputs pandas DataFrames and outputs python dictionaries of plot specifications) and create_html (inputs python dictionaries of plot specifications).

This example uses the Reporter class to create an HTML file with a single plot:

from datetime import datetime
import os
import htmltag
import idealreport
import pandas as pd

# example P+L data - note: plots will use the index as the x axis 
df = pd.DataFrame(data={'time': [datetime(2016, 2, 3, 10, 0, 0), datetime(2016, 2, 3, 11, 0, 0), datetime(2016, 2, 3, 12, 0, 0)],
                    'P+L': [80, -20, 45]})
df = df.set_index('time')

# instantiate the Reporter class and specify the title and output location
output_path = '.'
r = idealreport.Reporter(title='Report', output_file=os.path.join(output_path, 'report.html'))

# title
r.h += htmltag.h4('Example report')

# bar plot 
r.h += r.plot.bar(df=df, title='Hourly P+L', y_label='$k')

# render to HTML
r.generate()