Day7

Rohit
Dec 31, 2018

Yo! That’s NEWYEAR'EVE guys. Yesterday, The thought of 1-D convolution was left in my mind. I was not sure how it can be used and proper working of them. So, i decided to do a project on time-sequence data with 1-D convolutions. Here it is:-

Pledge

I, Rohit Sethi swear to complete #100DaysOfCode to the best of my ability and with true spirit.


Task Completed

On Day 7, I did the following :-

Trained a 1-D CNN for classifying data from an accelerometer into various human activities with accuracy of 92%


Learnings

On Day 7, I learnt the following :-

  1. confusion matrix and it’s importance.

  2. Scoring the model as precision , recall, f1 - score.

  3. callbacks such as keras.EarlyStopping.


link to the work on Day 7 here