What A Sad Moment Welp, the mystery of a NN and CNN has definitely been demystified for now… at least the surface level of it. I had to start somewhere! I wanted to take an easily digestible example and break it down as a CNN problem. I’d like to think I did that, or at […]
Read More Chi / Larissa Face Detection #18 – ConclusionOkay, so I’ve downloaded all the test images from post #13 and I’ve saved them onto S3. Let’s take a look at all of them Download Test Set From S3 In [1]: # Install tflearn import os os.system(“sudo pip install tflearn tqdm boto3 opencv-python”) Out[1]: 0 In [2]: import cv2 import numpy as np import pandas as […]
Read More Chi / Larissa Face Detection #17 – Testing CNN With Different Test SetPreface To explore the CNN inner workings a bit more, I want to use this post to tweak some parameters and see how that impacts the filters that were learned by the TFlearn model. Off the top of my naive head, I can think of a few parameters in our CNN that can be optimized: […]
Read More Chi / Larissa Face Detection #16 – Tweaking CNN ParametersReview At this point, I’ve tested my CNN a little bit and learned that the hair really matters. If the CNN sees a lighter object representing a head with dark textures on either side of the head, it will think it is Lars. Fair enough, that might be one of the insights I would make […]
Read More Chi / Larissa Face Detection #15 – Viewing CNN FiltersPreface This notebook is another copy of post #11, however, in this post, I’m modifying my CNN to have dedicated “names” at each layer. The names will be important for my next post because it will allow me to access the filters within each convolutional layer and view them. Nothing really new in this post […]
Read More Chi / Larissa Face Detection #14 – Naming CNN LayersReview At this point, we have a trained and working model that has performed literally flawlessly on our test set. When I left off, I talked a bit about how controlled my experiment was, and how I would like to test the model out on a few more variations of photos to understand our NN […]
Read More Chi / Larissa Face Detection #13 – Testing CNN With Various Test CasesFeature Building In the last post, I looked at loading and resizing images within cv2, an image processing library in python. In this post, I’m going to aggregate all the photos of me and my girlfriend and create an input feature array to pass into our TFlearn model. Let’s get it. All my photos are […]
Read More Chi / Larissa Face Detection #11 – Building & Training Convolutional Neural NetworkI said in the last post that I was ready to try this on our face. I’m totally not ready – I’m just missing one last thing I wanted to try. TFlearn gives us the option of easily including drop out in our fully connected layers of the NN, and I wanted to explore that […]
Read More Chi / Larissa Face Detection #9 – Cutting Cloud Costs with Infrastructure Automation (Part IV – Preparing and Training Code on AWS with Drop Out – MNIST)Review At this point, we have a working EC2 instance loaded with the AWS Deep Learning AMI, with the jupyter notebook tested and ready to go. The last step I want to take to automate this analysis as much as I can is to just get a notebook ready so I can just import it […]
Read More Chi / Larissa Face Detection #8 – Cutting Cloud Costs with Infrastructure Automation (Part III – Preparing and Training Code on AWS – MNIST)Review In the last post, to put it short, I spun up and environment that could’ve taken up to a half a year to procure, set up, and configure in my past life. We worked with networking, security, computing, and storage components all in one script! At the end of the day, I have an […]
Read More Chi / Larissa Face Detection #7 – Cutting Cloud Costs with Infrastructure Automation (Part II – EC2 Configuration Automation)
