Hola! @mansiarora1009 has created a ZenHub account for the egovernments organization. ZenHub is the only project management tool integrated natively in GitHub – created specifically for fast-moving, software-driven teams.

How do I use ZenHub?

To get set up with ZenHub, all you have to do is download the browser extension and log in with your GitHub account. Once you do, you’ll get access to ZenHub’s complete feature-set immediately.

What can ZenHub do?

ZenHub adds a series of enhancements directly inside the GitHub UI:

• Real-time, customizable task boards for GitHub issues;
• Multi-Repository burndown charts, estimates, and velocity tracking based on GitHub Milestones;
• Personal to-do lists and task prioritization;
• Time-saving shortcuts – like a quick repo switcher, a “Move issue” button, and much more.

Add ZenHub to GitHub

Still curious? See more ZenHub features or read user reviews. This issue was written by your friendly ZenHub bot, posted by request from @mansiarora1009.

Need to come up with a defn for an anomaly in the context of complaints.
Is it a point anomaly, or is it a 'trend' that needs to be detected

Need to have data for the entire year. Consolidate the data Ankit fetched, and the one given to us in the sprint into a single file for the analysis.

This will be used in the validation phase to see how well our forecasts work

Ref:

• [1] - http://www.jmlr.org/proceedings/papers/v39/oliveira14.pdf
• [2] - http://stats.stackexchange.com/questions/47950/ensemble-time-series-model

Right now, the ARIMA notebook requires manual selection of the (p, d, q) parameters. It should be automated, in the following way:

• Build a grid of possible (p, d, q) params
• Split the data into a train, test, and validation set
• Select [1] one param set, fit a model, compute accuracy[2].
• Repeat until desired.
• Pick the param set and model which scores highest on the validation set
• Compute final accuracy on the test set

[1]:
The sampling can be done in two ways, and both needs to be explored. This entirely depends on the size of the sample space.

1. Iterate through the entire space: Possible only if it is small, and if modeling is fast
2. Perform Random Sampling of the space: Might not give us the best model

[2]:
As of today, the accuracy metric is still not finalized. Refer #39

Check whether seasonal models outperform non-seasonal variants, and compare it with the excel sheet provided by the NGO

Its bit confusing when you call "Admin.create()".

Right now the script takes up an enormous amount of memory and time.
The script shouldn't generate data for each complaint type x date x ward, which is what it is doing.

Need to have:

• README: How to run the script
• README: Example run
• Documentation in scripts

Right now the Periodicity value is ignored, and only monthly aggregated plots are shown