SeaWiDS Datathon Workshop Day 1

  • Interactive computing
  • Notebooks
  • Collaborating

Interactive Computing

  • You've probably already done interactive computing in your life.

Calculator

calculator

Graphing Calculator

graphing

Spreadsheet

spreadsheet

Notebook

In [22]:
import numpy as np
import pandas as pd

df = pd.DataFrame({'x': np.arange(0.0, 2.0, 0.01)})
df['y = sin(x)'] = np.sin(2 * np.pi * df['x'])
df = df.set_index('x')
df.plot(grid=True)
Out[22]:
<matplotlib.axes._subplots.AxesSubplot at 0x124686950>

Why Interactive Computing?

  • Build up intuition for your data as you work with it.

Facilitates the following process:

In [ ]:
raw_data = load_some_data()
clean_data = clean_up_the_data(raw_data)

satisfied = False
while not satisfied:
    sliced_data = slice_the_data(data)
    visualize(sliced_data)

    if new_ideas_or_intuitions_gained():
        update_cleaup_method()
        update_how_we_slice_data()

    if objective_achieved() or sleepy() or now() > deadline():
        satisfied = True

Notebook

What is it?

A document that combines:

  • markdown cells (for communicating with humans)
  • input code cells (for communicating with humans and computers)
  • output cells (for communicating with humans)

I Am A Markdown Title Cell

$ i = a_m * A + \frac{math}{cell} $

In [29]:
# I am an input cell
print(''.join(reversed('.llec tuptuo na ma I')))

I am an output cell.

Why is it good?

A single tool to enable the communication of technical ideas.

At a highlevel:

  • Write markdown and use equations to explain your ideas.
  • Write code to test your ideas.
  • Execute code that validates your ideas.
  • Share your ideas with others by uploading to GitHub.

How do I get started?

Note: Colab and Jupyter produce .ipynb files that are compatible with each other.

Ask mentors for help!

Collaborating

SeaWiDS recommends the following workflow for datathon teams:

  • Create a directory in Google Drive to store your Notebooks (.ipynb files).
  • Create a "master" notebook file.
  • Each team member maintains their own personal notebook file.
  • During datathon meetings, teams can get together and merge the most valuable things produced by each of the team's members.

Resources

Mentors

  • Charlene Yan
  • Jasmine Wilkerson
  • Katlyn Edwards
  • Minmin Jin
  • Riley Porter
  • Shad Sharma