tommct

From an open Jupyter Notebook homework assignment, select "Coursera" to take you to the home page. Make a new notebook and fill it with the following and excute the cell with:

%%bash
tar cvfz hw.tar.gz .

This may take a little while to run depending on the packages. Select "Coursera" again to take you to the Home directory. Check the hw.tar.gz file and then Download. After the file is downloaded, delete it.

tommct

def columns_via_merge(df: pd.DataFrame, df2: pd.DataFrame, oncols: list, assigning: list):
    """
    Add (or replace) columns to df that map via a merge with df2.

    Examples:

        # Add the ord value to a subset of a DataFrame
        ABC = [chr(x) for x in range(ord('A'), ord('Z') + 1)]
        AABBCC = [chr(x)+chr(x) for x in range(ord('A'), ord('Z') + 1)]
        abc = [chr(x) for x in range(ord('a'), ord('z') + 1)]

tommct

This is a recipe for making box plots overlaying histograms in Tableau version 9.3. It largely borrows from http://vizpainter.com/some-tableau-tips-options-for-box-and-whisker/ and http://vizdiff.blogspot.com/2015/11/overlaying-histogram-with-box-and.html.

1. Create a fixed continuous variable for number of objects per dimension. For example, the number of unique assignments per user:

    [Assignments Per User] = {FIXED [Userid] : COUNTD([Assignmentid])}
   

2. Set the variable's Default Aggregation to COUNT.

3. Drag the variable from Measures to the columns shelf.

4. Set it to "Dimension" instead of CNT().

tommct

This is a recipe for using Sklearn to build a cosine similarity matrix and then to build dendrograms from it.

import numpy as np
import matplotlib.pyplot as plt
import scipy.cluster.hierarchy
import scipy.spatial.distance
from scipy.spatial.distance import pdist
from sklearn.metrics.pairwise import cosine_similarity

Make a "feature matrix" of 15 items that will be the binary representation of each index.

tommct

This is Python code for updating the file modification date of a file on MacOSX or Linux. In this example, I had copied .dv files from my camcorder, which encoded the date in the filename, but had as the modification date, the time I transferred the file from the camcorder.

import os
import time
fpath = '/path/to/dv/files'
for root, dirs, files in os.walk(fpath):
    for name in files:

if name[-3:]=='.dv':

tommct

Implements multiple, stacked plots with brushing. This extends the example at http://bl.ocks.org/mbostock/1667367 and allows for multiple panels where each subsequent panel zooms from the previous. Data points are also smoothed, permitting data with over 100,000 points to have an overview with subsequent telescoping while maintaining context.

tommct

This D3 example demonstrates constrained zooming, much like http://bl.ocks.org/tommct/5671250, but also illustrates the use of hierarchical ordinal tick marks. It does this by using the normalized values that one gets when using a hierarchical partition layout.

tommct

Implements constrained zooming of an image constructed from a data-driven ImageData object placed onto an HTML5 Canvas while giving the marginal distributions of the underlying data. Borrows heavily from https://gist.github.com/mbostock/3074470 and https://gist.github.com/tommct/8049508.

tommct

Implements constrained zooming of an image constructed from a data-driven ImageData object placed onto an HTML5 Canvas. Borrows heavily from https://gist.github.com/mbostock/3074470.

tommct

Implements constrained zooming of an image put onto an HTML5 Canvas.