From 255a9aa73caacdcf8383daa52a860b9bf3e0037b Mon Sep 17 00:00:00 2001
From: KeshavAnandCode <keshavanand.dev@gmail.com>
Date: Wed, 18 Mar 2026 17:56:49 -0500
Subject: [PATCH] finished CNN class

---
 ...l-training-v1.ipynb => tutorial-cnn.ipynb} | 106 ++++++++++++++++--
 1 file changed, 99 insertions(+), 7 deletions(-)
 rename notebooks/{model-training-v1.ipynb => tutorial-cnn.ipynb} (92%)

diff --git a/notebooks/model-training-v1.ipynb b/notebooks/tutorial-cnn.ipynb
similarity index 92%
rename from notebooks/model-training-v1.ipynb
rename to notebooks/tutorial-cnn.ipynb
index c171d74..0418348 100644
--- a/notebooks/model-training-v1.ipynb
+++ b/notebooks/tutorial-cnn.ipynb
@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 2,
    "id": "7a37220a",
    "metadata": {},
    "outputs": [
@@ -20,7 +20,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 3,
    "id": "d318d1f0",
    "metadata": {},
    "outputs": [
@@ -61,7 +61,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 4,
    "id": "5604ace3",
    "metadata": {},
    "outputs": [
@@ -110,7 +110,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 5,
    "id": "3cedd586",
    "metadata": {},
    "outputs": [
@@ -150,7 +150,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 6,
    "id": "8f556b22",
    "metadata": {},
    "outputs": [
@@ -184,7 +184,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 7,
    "id": "37793c77",
    "metadata": {},
    "outputs": [
@@ -222,7 +222,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 8,
    "id": "f68c1a25",
    "metadata": {},
    "outputs": [
@@ -244,6 +244,98 @@
     "\n",
     "print(f\"Train: {len(train_dataset)}, Test: {len(test_dataset)}\")"
    ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "2eede814",
+   "metadata": {},
+   "source": [
+    "Load the data into batches for faster loading...use 4 threads on CPU\n",
+    "\n",
+    "Shuffling train dataset so as to not overfit on one class"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "e1539eaa",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "8\n"
+     ]
+    }
+   ],
+   "source": [
+    "train_loader = DataLoader(train_dataset, batch_size=32, shuffle=True, num_workers=4)\n",
+    "test_loader = DataLoader(test_dataset, batch_size=32, shuffle=False, num_workers=4)\n",
+    "\n",
+    "classes = ('bicycle', 'bus', 'car', 'motorcycle',\n",
+    "           'nonvehicles', 'taxi', 'truck', 'van')\n",
+    "\n",
+    "print(len(classes))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "acebe98e",
+   "metadata": {},
+   "source": [
+    "Tutorial CNN architecture implmenetation"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "d1b7d9ca",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Net(\n",
+      "  (conv1): Conv2d(3, 6, kernel_size=(5, 5), stride=(1, 1))\n",
+      "  (pool): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
+      "  (conv2): Conv2d(6, 16, kernel_size=(5, 5), stride=(1, 1))\n",
+      "  (fc1): Linear(in_features=2704, out_features=120, bias=True)\n",
+      "  (fc2): Linear(in_features=120, out_features=84, bias=True)\n",
+      "  (fc3): Linear(in_features=84, out_features=8, bias=True)\n",
+      ")\n"
+     ]
+    }
+   ],
+   "source": [
+    "import torch\n",
+    "import torch.nn as nn\n",
+    "import torch.nn.functional as F\n",
+    "\n",
+    "class Net(nn.Module):\n",
+    "    def __init__(self):\n",
+    "        super().__init__()\n",
+    "        self.conv1 = nn.Conv2d(3, 6, 5) # 3 input channels for RGB, 6 filters [arbitrailly chosen ffrom PyTorch], 5x5 Kernel Size\n",
+    "        self.pool = nn.MaxPool2d(2, 2) # Halves spatial dimensions\n",
+    "        self.conv2 = nn.Conv2d(6, 16, 5) # 6 matches output of first conv layer, 16 arbitrary filter (more than earlier), 5x5 kernel\n",
+    "        self.fc1 = nn.Linear(16 * 13 * 13, 120) # 16 channels from conv layer 2; 13 *13 bc shape after 2 layers and 2 pools, 120 is arbitrary output size from tutorial\n",
+    "        self.fc2 = nn.Linear(120, 84) # 120 output from fc1 --> arbitrary lowering to 84\n",
+    "        self.fc3 = nn.Linear(84, len(classes))  # Bring down to number of classes \n",
+    "\n",
+    "    def forward(self, x):\n",
+    "        x = self.pool(F.relu(self.conv1(x))) #Conv layer 1 --> relu --> pool \n",
+    "        x = self.pool(F.relu(self.conv2(x))) #Conv layer 1 --> relu --> pool \n",
+    "        x = torch.flatten(x, 1) # Flattedn to 1d\n",
+    "        x = F.relu(self.fc1(x)) #fully connected layer shriking 1\n",
+    "        x = F.relu(self.fc2(x)) # Shrink again\n",
+    "        x = self.fc3(x) # Shrink to number of classes\n",
+    "        return x\n",
+    "\n",
+    "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
+    "model = Net().to(device)\n",
+    "print(model)"
+   ]
   }
  ],
  "metadata": {