The entire pipeline is now complete. The program is now able to collect data for all routes and train against all stops. Currently it only considers lat and lng which is not enough to make a good prediction. Will probably continue on improving the prediction focusing on route 21.
Currently route 21 and stop 14907 is trained -
Had a super great time at code for philly tonight, met lots of smart people with so many awesome ideas. A group of us came up with an MVP for this that would basically be a web page that would pump out a PDF.
Basically you would choose your schools and click 'go' and then the page would output a PDF with your dates.
To get to this point there will be a few steps:
I will create a excel sheet that basically does the same thing. Enter in one date and all the rest of them populate. Got good advice tonight to include fafsa, school calendar and SAT test dates in the date algorithms.
I'll then take the excel out to guidance counselors and verify that they would like it and use it and that the data is correct.
Once that's done, I'll head back to code for philly for help in building the webpage.
Future Visioning was great too. And include:
Use Phonegap to easily wrap the webpage
With the app, encourage a sharing buddy, someone who can help you with the process.
Be able to link to facebook and post pictures at key points like 'here's me at the SAT!!!'
Give Guidance - not just short summaries but in-depth on the WHY you need to do each of the steps.
So the update…i'm working on the excel sheet. See ya soon!
Last Code for Philly we coded the shared library. This time we're working on reading the temperature sensor and displaying on the LCD screen. We've also posted tons of photos on our Climate Tracker Facebook page! facebook.com/CACclimatetracker
Cyclephilly is an online map and a mobile application that let the users record their daily routes in the city of Philadelphia. Although it has been around only for a couple of months, it can be already called a success story.
The code now stores data in mysql, prepares the training data, trains with the training data, and parses the result. The front end API is already done. The last remaining step is to store the results into mysql and modify the php front end to use it.
Now that posts have been successfully publishing to the website via email for one solid month, it's time to add two more variables: synchronizing website posts to facebook and twitter. Simple, right?
After 3hrs of unsuccessful testing configurations in wordpress plugins for facebook and twitter, I finally took a suggestion to use If This Then That (http://ifttt.com). I followed instructions for both facebook and twitter plugins and authorization was not established for either one. I feel defeated by what should be a simple OAuth linkage with an API token but the effectiveness of IFTTT is very appealing after hours of frustration.
Project update: Team is continuing to work on Python web app and general html assets. Going on hiatus for a few weeks while main programmer is on vacation. Housekeeping note: We have not seen and do not know the person "Xiangwei Meng" who lists themself as 'CEO' (this is not a business, it is an open source project.) But we'd love any and all help people would like to provide. :) Current team is: Jack Amoratis, main organizer and programmer; Abdus-Salaam Muwwakkil researching web app; Alex Olson helping with web design; Ian Nieves willing to learn Python to help with web app. Looking forward to resuming work in a few weeks!