創建者 rayed a•
Mar 04, 2019
Great List of Courses for People who are interested
創建者 PRALAY P•
Jun 23, 2018
創建者 Mark H•
Nov 07, 2019
This was a challenging yet relevant capstone course. The assignment really pushes you to understand the course material, and to go back and review it if not.
I recommend using tools other than the Analytical Solver Platform if you know how, e.g. Jupyter. The data set will absolutely not work on the cloud version of the solver (the VM/server hangs and crashes with every operation). The Mac version of the solver is useless for the second week, but does the third week of the assignment just fine (simulation & optimization). The Windows version of solver works flawlessly on Excel - I wasted so much time getting it to run on the Mac, don't even bother.
I've harped on ASP in previous reviews, so I won't score the course down again, since I took it of my own volition knowing I'd have to use the software. This course deserves a full 5 of 5 stars!
創建者 Alex D•
Nov 15, 2019
It's been a great opportunity to practically apply knowledge I got over the specialization. Highly recommended.
創建者 Sally M•
Jul 10, 2017
This is a very poorly structured course. Assignments are not clear. The grading rubrics have multiple errors.
For example, week 1 Assignment: "Data visualization - part A. Create a plot to illustrate the distribution of interest rate (int_rate). Make sure to use a plot appropriate for illustrating distribution."
Distribution plots are histograms, density plots and dot plots. The rubric first asks if one of those plots was used. Then the next question grades THIS SAME PLOT based on relationship between two variables (scatterplot), NOT distribution. I correctly plotted a histogram, but lost a point because it was not a scatterplot?! It's not supposed to be!
Another example from the same assignment: "Are there columns that have rows with missing values or outliers? The missing values are tagged as NA in the dataset. Suggest strategies to handle them."
8 correct columns identified as having outliers (installment, annual_inc, revol_bal,revol_util, total_acc, acc_open_past_24mths, and total_pymnt) and strategies to handle them suggested
At least 6 correct columns identified
There are only 7 columns with outliers!!! You list them and only them in the first answer choice. Yet I lost a point because it's not 8 columns!!!
The ASP is a joke. It is SLOOOOW. Like, the assignment that should have taken me an hour to do turned into 6 because the platform would crash every 30 seconds. I can't believe anyone actually pays to use it. We'd be much better off using a programming language like R or Python to do these assignments.
Also, the assignments that depend on running a model in ASP should specify what random seed to use in partitioning the data, what exact classification tree type to use, so that we would all get same answers, and it would be easier to grade other people's work.
The rest of this specialization was great, but this last course just ruined the whole impression for me.