22WI_CACC2993-Exploration Project

Please complete the following activity to practice working with aggregates and NumPy arrays. You may work with a partner or individually. 

1.     Create a new Jupyter Notebook (we will be working with earthquake data). Pay attention to where you create the Notebook.

2.     Download the data file from Moodle and store it in the same folder as the Jupyter Notebook.

3.     Create a report on all earthquakes that occurred globally in the last 30 days. Your report should include:

a.     Compute basic stats on all of the magnitude (max, min, median, mean, and 25^th/75^th percentiles)

b.     The average magnitude where the type is ml

c.      Find the date and location of the largest magnitude (can use reverse geocode of latitude/longitude or place)

d.     Plot the significant earthquakes (mag 4.5+) in a histogram

e.     Plot a histogram of only earthquakes that occurred in the US (hint: you know latitude/longitude).

For the submission, please create a Jupyter Notebook downloaded as pdf. You should include text throughout to explain/caption the code. To turn a cell into something you can type words in, with the cell selected, click Cell > Type > markdown.

Mini-Assignment: NumPy and Pandas

Please complete the following questions.  A pdf of your Jupyter Notebook is preferred unless that currently is not working for you. This should be a quick review assignment for you. 

1.     Create a 5x7 NumPy array filled with random numbers. Use fancy indexing to access the following entries (row, col): (1,3),(4,5),(3,2), and (0,6).

2.     Explain what np.newaxis does and provide an example of how one might use it.

3.     What sorting method does np.sort use and what is its runtime complexity (Big-O)? What does this mean?

4.     Given the following NumPy array: [4,1,5,2,7,3], what are the indices of the sorted order?

5.     Using the array from #1, sort the columns.

6.     Using the array from #1, sort the rows.

7.     How is a structured array different from a traditional NumPy array, and why might one use it?

8.     Complete the following exercise – activity created by Guilherme Samora.

a.     Import two datasets: cars1 and cars2.

b.     Print the top five rows of each.

c.      Clean cars1 removing any unnamed columns (hint: loc might be helpful)

d.     How many observations exist in each dataset?

e.     Join cars1 and cars2 into one dataframe.

f.      Let’s add a column called owners consisting of random numbers from 15,000 to 73,000.

g.     Add it to the cars dataframe.

h.     Print the last five rows of the dataset.

i.       Which car has the min mpg?

j.       Which car has the max horsepower?

• You are welcome to work with a partner. Please experiment:
• Grab the dataset called cars1 (on Moodle) and plot mpg, cylinders, and weight to illustrate correlation using your preferred chart
• Simulate data…you’ll need 5 questions, likert scale (1-5) for each question, randomize responses (15 responses per question). Create a plot showing the deviation. 
• Using the earthquake dataset, creatively display the distribution of the magnitude

You have a choice =)

Details about the cars dataset: https://corgis-edu.github.io/corgis/csv/cars/

Please submit your code/analysis as pdf, html, or the py/notebook file. 

Two options are attached. Third option is to create a dandelion image classifier: https://www.kaggle.com/coloradokb/dandelionimages.

Please complete the following activity with a partner or two. The goal is to create a recommendation engine for Netflix movies or TV shows based on title or cast. 

Please complete the attached assignment to the best of your ability. If you run into issues along the way, please reach out.