Machine Learning Demo

Fizz Buzz by Deep Learning

Transcript

1. Machine Learning Demo
2. Fizz Buzz by Deep Learning Just a funny joke…
3. interviewer: OK, so are you familiar with "fizz buzz"? me: ... interviewer: Is that a yes or a no? me: It's more of a "I can't believe you're asking me that." interviewer: OK, so I need you to print the numbers from 1 to 100, except that if the number is divisible by 3 print "fizz", if it's divisible by 5 print "buzz", and if it's divisible by 15 print "fizzbuzz". me: I'm familiar with it. interviewer: Great, we find that candidates who can't get this right don't do well here. me: ... interviewer: Here's a marker and an eraser. me: [thinks for a couple of minutes] interviewer: Do you need help getting started? me: No, no, I'm good. So let's start with some standard imports: import numpy as np import tensorflow as tf IUUQKPFMHSVTDPNp[[CV[[JOUFOTPSqPX
4. Test Data and Training Data Y Z      'J[[ ʜ ʜ      'J[[ ʜ ʜ   5FTU
   %BUB 5SBJOJOH
   %BUB 5IJT
   JT
   B
   CBE
   QSBDUJDF
   %BUB
   NVTU
   CF
   TIV⒐FE
   CFGPSF
   TQMJUUJOH
5. Prepare Data for Model Y Z    'J[[      #V[[ Y` Z` <
   
   
   
   
   
   
   
   
   > <
   
   
   > <
   
   
   
   
   
   
   
   
   > <
   
   
   > <
   
   
   
   
   
   
   
   
   > <
   
   
   > <
   
   
   
   
   
   
   
   
   > <
   
   
   > <
   
   
   
   
   
   
   
   
   > <
   
   
   > 
   'FBUVSFT 
   $MBTTFT Y
   
   d
    Z
   /VNCFS
   'J[[
   #V[[
   PS
   'J[[#V[[
6. Define and Train Model
7. 1, [9.97905254e-01 4.31590917e-04 1.65658456e-03 6.49713138e-06] 2, [9.98885453e-01 4.59088405e-05 1.06858183e-03 4.36434107e-08] 3, [2.83584304e-05 9.99966264e-01 7.07421236e-07 4.65488665e-06] 4, [9.99127209e-01 4.70238447e-05 8.25370138e-04 3.38930732e-07] 5, [9.57046461e-04 2.11936946e-04 9.98373508e-01 4.57487913e-04] … 100, [3.64703596e-01 1.86683770e-04 6.35105133e-01 4.65522362e-06] Predict
8. Learning Curve
9. Experimentations • Try increasing or decreasing the size of data. • Try adding or removing hidden layer. • Try changing badge size. • Try including Test data into Training data.