Classifier with multiple images as input and multiple labels as output

ste · November 13, 2018, 7:00pm

Here is the updated model with the option to use transfer learning with multiple channel images (any number of channels).
It duplicates RGB weights of first convolution on top of missing channels (IE: RGBY image use RGBR weights, RGBXYZ use RGBRGB).

github.com

artste/fastai-samples/blob/master/kaggle/lesson2-protein-human-protein-atlas-v1_256-resnet50.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Lesson 2 - Contribute to human protein atlas"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Try to address the uman protein atlas competition using concepts from lesson2/v3\n",
    "\n",
    "https://www.kaggle.com/c/human-protein-atlas-image-classification/"
   ]
  },
  {
   "cell_type": "code",

This file has been truncated. show original

TODO: Data Block Api support…