drytuna/cs373-netflix

Just use the probe data, and see if you can meet or beat their RMSE of 0.9474.

cs373-netflix's Issues

Invalid file pathing

Re-check the file path.
Formatting the path to traverse the Training Set in a loop instead of running through it 1 by 1

Big files traverse

2.6 million customers and 17770 movies
Cache generator took hours to process, but did not finish writing average values
Need a better method

Adding blank lines to test files

Output the specified format by the professor

Index out of bound exception in our customer cache

need to fix issue with index out of bound when calling the customer cache

Create a cache folder

Begin the project with calculated average ratings and other statistics cached in files

Git Repository

Set up a private Git repository at GitHub, named cs373-netflix.
Invite the grader to your repository. Commit at least 5 times. Commit once for each bug or feature. If you cannot describe your changes in a sentence, you are not committing often enough. Write meaningful commit messages and identify the corresponding issue in the issue tracker (below). Create a log of the commits. Push frequently. It is your responsibility to protect your code from the rest of the students in the class. If your code gets out, you are as guilty as the recipient of academic dishonesty.

Submission

turnin --submit chpkim cs373pj3 Netflix.zip
Submit a single ZIP file, named Netflix.zip, to the grader's Turnin account, with the following files:

README.txt
caches/*
Netflix.html
Netflix.log
Netflix.py
RunNetflix.in
RunNetflix.out
RunNetflix.py
TestNetflix.out
TestNetflix.py

Date cache

Determine a weighing technique for date created vs date watched.
This might help increase the RMSE

Issue Tracker

The GitHub repository comes with an issue tracker.
Create an issue for each of the requirements in this table. Create an issue for each bug or feature, both open and closed. Label and describe each issue adequately. Create at least 10 more issues in addition to the requirements in this table.

Enhance the prediction algorithm

Enhance the prediction algorithm by adding factors other than just the average. Give each factor a weight and see the impact.

Unit Tests

The grader's GitHub account will have a public Git repository for unit tests and acceptance tests.
Write unit tests before you write the code. When you encounter a bug, write a unit test that fails, fix the bug, and confirm that the unit test passes. Write at least 3 unit tests for each function. Tests corner cases and failure cases. Name tests logically. Push and pull the unit tests to and from the grader's repository. Prepend - to the file names at GitHub (i.e. foo-TestNetflix.py and foo-TestNetflix.out). Reach consensus on the unit tests.

Weight values graph

Create a weight vs. RMSE graph to choose the weight with the best RMSE

Acceptance Tests

The grader's GitHub account will have a public Git repository for unit tests and acceptance tests.
Write acceptance tests before your write the code. When you encounter a bug, write an acceptance test that fails, fix the bug, and confirm that the acceptance test passes. Write an auxiliary program to randomly generate acceptance tests. Create at least 1000 lines of acceptance tests. Tests corner cases and failure cases. Push and pull the acceptance tests to and from the grader's repository. Prepend - to the file names at GitHub (i.e. foo-RunNetflix.in and foo-RunNetflix.out). Reach consensus on the acceptance tests.

Create the dumbest solution

Create the dumbest solution by simply predicting that the customer's review is going to be his/her average. And then calculate the RMSE of that.

Rating exceeded 0.0 - 5.0 boundary

Need to handle cases when predicted rating exceeds the boundaries

Implementation

Use assert to check pre-conditions, post-conditions, argument validity, return-value validity, and invariants. Worry about this last, but your program should run as fast as possible and use as little memory as possible.

Documentation

Use pydoc to document the interfaces.
The above documentation only needs to be generated for Netflix.py. Comment each function meaningfully. Use comments only if you need to explain the why of a particular implementation. Choose a coding convention and be consistent. Use good variable names. Write readable code with good indentation, blank lines, and blank spaces.