{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "2fb5f5d8",
   "metadata": {},
   "source": [
    "# Training a custom Yolo model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d0e005ba",
   "metadata": {},
   "outputs": [],
   "source": [
    "#  Imports\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7ee04c1a",
   "metadata": {},
   "source": [
    "## You'll need to decide on a type of task to complete (Classification, Detection, Segmentaion, Pose, OBB, Tracking)\n",
    "- (Write your choice of task type here)\n",
    "### Then you'll need to decide on a Dataset for training\n",
    "- (Copy the link to the dataset here (I suggest lookign at Roboflow, or YOLO Datasets))\n",
    "- For fun think about how you would collect all these images and annotate them \"in the wild\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8c484e25",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load the model that you'll use to perform transfer learning\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a8008e42",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Have the Niave model do some predictions on the data you already have\n",
    "# This is just for comparing before and after training\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2f98b89e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Train the model\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8d803817",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save the model (many MIME types are supported .onnx is commonly used for edge devices (low end computers))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "313f7519",
   "metadata": {},
   "source": [
    "# Discuss the outputs of the training process\n",
    "\n",
    "### Was this a success?\n",
    "### Discuss performance\n",
    "- Before Transfer Learning:\n",
    "- After Transfer Learning:\n",
    "### What did you learn?\n",
    "- \n",
    "### What excites you for future?\n",
    "- \n",
    "\n",
    "### How can you use this process in you life / work?\n",
    "- For instance I've considered training a spooky model to predict when someone rings my doorbell / is on my porch to scare kids on Halloween 🎃🧟‍♂️\n",
    "- "
   ]
  }
 ],
 "metadata": {
  "language_info": {
   "name": "python"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}