Developing Intuitions about parameter tuning

vishak_bharadwaj · March 10, 2020, 9:43am

Hello People,

I just built a facial keypoints recognition dataset that feeds in a greyscale image into the network and outputs 72 keypoints on the face.
I loaded up a pretrained resnet34 and trained the model on this data.

github.com

VishakBharadwaj94/facial_keypoints/blob/master/facial_keypoints.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "facial_keypoints.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "kxnZs6gIzknB",

This file has been truncated. show original

The loss was all over the place. So what am I doing wrong here? How do I make the model work better?

ptrblck · March 10, 2020, 11:48pm

Try to scale down the problem and try to overfit a small data sample (e.g. just 10 samples) by playing around with the hyperparameters.
Once your model can overfit it, you could try to scale it up again.