Inspired by the job-interview with Joel Grus, the goal of this challenge is to write a tensorflow (or other deep/machine learning) program that learns Fizzbuzz and correctly prints out the answers to the positive integers less than 1000.

You can assume there are files named train.csv and test.csv and each contain a sorted list of sequential integers and the fizzbuzz answer:

 100, buzz
 101, 101
 102, fizz
 103, 103
 104, 104
 105, buzz
 150000, fizzbuzz

test.csv spans 1-1000 and train.csv spans 1001-150000.


  1. You must not hard-code the rules to Fizzbuzz anywhere in your program. The output must be from a machine learned representation that is learned while running the code.
  2. You must utilize train.csv in the training set and check your output against test.csv. You cannot use test.csv during training.
  3. You must get all outputs correct from test.csv (but as is case with deep-learning, we'll allow your code to fail this rule no more than 5% of the time).
  4. You may use any language and any external module (eg. python/tensorflow) as long they explicitly perform some kind of learning. Reference both the language and the module in the title of your post.
  5. This is a popularity contest, so the submission with the most votes after one week wins.
  • 3
    \$\begingroup\$ Hi Hooked, welcome to the site! I'd encourage you to try using the sandbox before posting a challenge next time, since you probably would have been advised to avoid a popularity contest. Right now, we're in an awkward stage where popularity contests are still technically allowed, but frowned upon, and hard to do right (especially for new users). One thing that might attract downvotes on this question is that it's asking to solve a task in a very specific way, which is cautioned against in the tag wiki. \$\endgroup\$ – James May 24 '16 at 4:41
  • 3
    \$\begingroup\$ Thanks for the info @DrGreenEggsandHamDJ. I'm a long time lurker and I tried to model this off of previous questions (without knowledge of the transition). Re restrictions: there are many creative ways to solve this problem (the type of NN isn't enforced). If it gets downvoted I'll understand and pack this one up, but I'm really looking forward to see what comes out of this! (I don't think there are any deep learning codegolf.SE questions at all) \$\endgroup\$ – Hooked May 24 '16 at 5:13
  • \$\begingroup\$ Not all machine learning algorithms are "Deep Learning", so I don't really see what it has to do with the challenge. \$\endgroup\$ – Fatalize May 24 '16 at 6:28
  • 3
    \$\begingroup\$ Where can we find train.csv and test.csv? What is the validity criterion? \$\endgroup\$ – flawr May 24 '16 at 8:28
  • 1
    \$\begingroup\$ Here's an example of a good machine learning question: codegolf.stackexchange.com/questions/44707/… \$\endgroup\$ – durron597 May 24 '16 at 17:45