Share your work here ✅

KarlH · October 26, 2018, 5:36am

I grabbed activations from inside the model and ran them through a t-SNE to see how different dog/cat classes would cluster.

This is from the final conv block:

Also looking at how different activation layers in a conv block extract different features.

From a ReLU layer:

github.com

kheyer/ML-DL-Projects/blob/master/Pets TSNE/pets_tsne.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# CNN Visualizations with Dogs and Cats\n",
    "\n",
    "Taking a peek inside CNNs with fastai 1.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%reload_ext autoreload\n",
    "%autoreload 2\n",
    "%matplotlib inline"

This file has been truncated. show original