mnist_the_forward_forward_algorithm.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100, Loss: 0.6622, Accuracy: 71.03%\n",
      "Epoch 2/100, Loss: 0.3617, Accuracy: 88.36%\n",
      "Epoch 3/100, Loss: 0.3165, Accuracy: 89.63%\n",
      "Epoch 4/100, Loss: 0.3067, Accuracy: 90.66%\n",
      "Epoch 5/100, Loss: 0.2986, Accuracy: 92.27%\n",
      "Epoch 6/100, Loss: 0.2902, Accuracy: 92.23%\n",
      "Epoch 7/100, Loss: 0.2871, Accuracy: 92.42%\n",
      "Epoch 8/100, Loss: 0.2896, Accuracy: 91.72%\n",
      "Epoch 9/100, Loss: 0.2886, Accuracy: 93.10%\n",
      "Epoch 10/100, Loss: 0.2836, Accuracy: 93.03%\n",
      "Epoch 11/100, Loss: 0.2815, Accuracy: 92.99%\n",
      "Epoch 12/100, Loss: 0.2820, Accuracy: 93.33%\n",
      "Epoch 13/100, Loss: 0.2788, Accuracy: 93.43%\n",
      "Epoch 14/100, Loss: 0.2775, Accuracy: 93.11%\n",
      "Epoch 15/100, Loss: 0.2774, Accuracy: 93.08%\n",
      "Epoch 16/100, Loss: 0.2793, Accuracy: 92.27%\n",
      "Epoch 17/100, Loss: 0.2760, Accuracy: 92.81%\n",
      "Epoch 18/100, Loss: 0.2762, Accuracy: 93.48%\n",
      "Epoch 19/100, Loss: 0.2782, Accuracy: 93.14%\n",
      "Epoch 20/100, Loss: 0.2782, Accuracy: 92.89%\n",
      "Epoch 21/100, Loss: 0.2755, Accuracy: 93.50%\n",
      "Epoch 22/100, Loss: 0.2789, Accuracy: 93.73%\n",
      "Epoch 23/100, Loss: 0.2748, Accuracy: 93.76%\n",
      "Epoch 24/100, Loss: 0.2800, Accuracy: 92.71%\n",
      "Epoch 25/100, Loss: 0.2761, Accuracy: 92.95%\n",
      "Epoch 26/100, Loss: 0.2776, Accuracy: 93.35%\n",
      "Epoch 27/100, Loss: 0.2736, Accuracy: 93.43%\n",
      "Epoch 28/100, Loss: 0.2717, Accuracy: 93.22%\n",
      "Epoch 29/100, Loss: 0.2729, Accuracy: 93.33%\n",
      "Epoch 30/100, Loss: 0.2743, Accuracy: 93.76%\n",
      "Epoch 31/100, Loss: 0.2724, Accuracy: 93.38%\n",
      "Epoch 32/100, Loss: 0.2763, Accuracy: 92.74%\n",
      "Epoch 33/100, Loss: 0.2729, Accuracy: 93.36%\n",
      "Epoch 34/100, Loss: 0.2727, Accuracy: 94.19%\n",
      "Epoch 35/100, Loss: 0.2703, Accuracy: 93.90%\n",
      "Epoch 36/100, Loss: 0.2729, Accuracy: 93.95%\n",
      "Epoch 37/100, Loss: 0.2772, Accuracy: 93.78%\n",
      "Epoch 38/100, Loss: 0.2723, Accuracy: 93.81%\n",
      "Epoch 39/100, Loss: 0.2769, Accuracy: 93.62%\n",
      "Epoch 40/100, Loss: 0.2720, Accuracy: 93.95%\n",
      "Epoch 41/100, Loss: 0.2696, Accuracy: 93.83%\n",
      "Epoch 42/100, Loss: 0.2727, Accuracy: 92.30%\n",
      "Epoch 43/100, Loss: 0.2744, Accuracy: 93.83%\n",
      "Epoch 44/100, Loss: 0.2687, Accuracy: 92.63%\n",
      "Epoch 45/100, Loss: 0.2762, Accuracy: 93.76%\n",
      "Epoch 46/100, Loss: 0.2706, Accuracy: 93.83%\n",
      "Epoch 47/100, Loss: 0.2729, Accuracy: 93.40%\n",
      "Epoch 48/100, Loss: 0.2706, Accuracy: 93.79%\n",
      "Epoch 49/100, Loss: 0.2709, Accuracy: 93.35%\n",
      "Epoch 50/100, Loss: 0.2764, Accuracy: 93.97%\n",
      "Epoch 51/100, Loss: 0.2726, Accuracy: 93.41%\n",
      "Epoch 52/100, Loss: 0.2687, Accuracy: 93.11%\n",
      "Epoch 53/100, Loss: 0.2695, Accuracy: 93.56%\n",
      "Epoch 54/100, Loss: 0.2694, Accuracy: 93.84%\n",
      "Epoch 55/100, Loss: 0.2679, Accuracy: 93.03%\n",
      "Epoch 56/100, Loss: 0.2841, Accuracy: 92.59%\n",
      "Epoch 57/100, Loss: 0.3249, Accuracy: 93.20%\n",
      "Epoch 58/100, Loss: 0.2942, Accuracy: 92.84%\n",
      "Epoch 59/100, Loss: 0.2903, Accuracy: 92.83%\n",
      "Epoch 60/100, Loss: 0.2864, Accuracy: 93.30%\n",
      "Epoch 61/100, Loss: 0.2826, Accuracy: 92.22%\n",
      "Epoch 62/100, Loss: 0.2830, Accuracy: 93.26%\n",
      "Epoch 63/100, Loss: 0.2986, Accuracy: 93.21%\n",
      "Epoch 64/100, Loss: 0.2868, Accuracy: 92.98%\n",
      "Epoch 65/100, Loss: 0.2897, Accuracy: 93.39%\n",
      "Epoch 66/100, Loss: 0.2897, Accuracy: 93.62%\n",
      "Epoch 67/100, Loss: 0.2791, Accuracy: 93.20%\n",
      "Epoch 68/100, Loss: 0.2961, Accuracy: 93.49%\n",
      "Epoch 69/100, Loss: 0.2892, Accuracy: 93.87%\n",
      "Epoch 70/100, Loss: 0.2890, Accuracy: 92.69%\n",
      "Epoch 71/100, Loss: 0.3043, Accuracy: 93.31%\n",
      "Epoch 72/100, Loss: 0.2988, Accuracy: 93.39%\n",
      "Epoch 73/100, Loss: 0.2938, Accuracy: 93.04%\n",
      "Epoch 74/100, Loss: 0.2905, Accuracy: 94.05%\n",
      "Epoch 75/100, Loss: 0.2979, Accuracy: 92.72%\n",
      "Epoch 76/100, Loss: 0.3046, Accuracy: 91.90%\n",
      "Epoch 77/100, Loss: 0.2906, Accuracy: 92.60%\n",
      "Epoch 78/100, Loss: 0.2850, Accuracy: 92.75%\n",
      "Epoch 79/100, Loss: 0.3029, Accuracy: 91.13%\n",
      "Epoch 80/100, Loss: 0.2907, Accuracy: 92.54%\n",
      "Epoch 81/100, Loss: 0.2901, Accuracy: 92.88%\n",
      "Epoch 82/100, Loss: 0.2825, Accuracy: 93.05%\n",
      "Epoch 83/100, Loss: 0.2802, Accuracy: 92.32%\n",
      "Epoch 84/100, Loss: 0.2792, Accuracy: 92.16%\n",
      "Epoch 85/100, Loss: 0.2816, Accuracy: 91.32%\n",
      "Epoch 86/100, Loss: 0.2841, Accuracy: 93.01%\n",
      "Epoch 87/100, Loss: 0.2820, Accuracy: 93.43%\n",
      "Epoch 88/100, Loss: 0.2747, Accuracy: 93.43%\n",
      "Epoch 89/100, Loss: 0.3022, Accuracy: 93.06%\n",
      "Epoch 90/100, Loss: 0.3072, Accuracy: 93.37%\n",
      "Epoch 91/100, Loss: 0.2981, Accuracy: 93.01%\n",
      "Epoch 92/100, Loss: 0.2988, Accuracy: 92.97%\n",
      "Epoch 93/100, Loss: 0.2941, Accuracy: 93.20%\n",
      "Epoch 94/100, Loss: 0.2888, Accuracy: 93.49%\n",
      "Epoch 95/100, Loss: 0.2894, Accuracy: 93.01%\n",
      "Epoch 96/100, Loss: 0.3006, Accuracy: 93.42%\n",
      "Epoch 97/100, Loss: 0.2949, Accuracy: 93.33%\n",
      "Epoch 98/100, Loss: 0.2863, Accuracy: 93.42%\n",
      "Epoch 99/100, Loss: 0.2872, Accuracy: 93.13%\n",
      "Epoch 100/100, Loss: 0.2871, Accuracy: 93.36%\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Implements the Forward-Forward Algorithm\n",
    "# https://arxiv.org/abs/2212.13345\n",
    "# The algorithm replaces backpropagation's forward and backward passes with two forward passes on different data with opposite objectives.\n",
    "\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "from torchvision import datasets, transforms\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Hyperparams\n",
    "BATCH_SIZE = 512\n",
    "TEST_BATCH_SIZE = 1000\n",
    "EPOCHS = 100\n",
    "LEARNING_RATE = 0.03\n",
    "THRESHOLD = 2.0\n",
    "WEIGHT_DECAY = 0.000001\n",
    "\n",
    "# Load MNIST\n",
    "transform = transforms.Compose([transforms.ToTensor(), transforms.Normalize((0.1307,), (0.3081,))])\n",
    "train_dataset = datasets.MNIST('./data', train=True, download=True, transform=transform)\n",
    "test_dataset = datasets.MNIST('./data', train=False, transform=transform)\n",
    "train_loader = torch.utils.data.DataLoader(train_dataset, batch_size=BATCH_SIZE, shuffle=True)\n",
    "test_loader = torch.utils.data.DataLoader(test_dataset, batch_size=TEST_BATCH_SIZE, shuffle=False)\n",
    "\n",
    "class FFLayer(nn.Module):\n",
    "    def __init__(self, input_size, output_size):\n",
    "        super(FFLayer, self).__init__()\n",
    "        self.linear = nn.Linear(input_size, output_size)\n",
    "        self.activation = nn.ReLU()\n",
    "\n",
    "    def forward(self, x):\n",
    "        x_direction = x / (x.norm(2, 1, keepdim=True) + 1e-4)\n",
    "        return self.activation(self.linear(x_direction))\n",
    "\n",
    "    def compute_goodness(self, x):\n",
    "        x = self.forward(x)\n",
    "        return x.pow(2).mean(1) # sum of squared activations\n",
    "\n",
    "class FFNetwork(nn.Module):\n",
    "    def __init__(self, layer_sizes):\n",
    "        super(FFNetwork, self).__init__()\n",
    "        self.layers = nn.ModuleList([FFLayer(layer_sizes[i], layer_sizes[i+1]) for i in range(len(layer_sizes)-1)])\n",
    "\n",
    "    def forward(self, x):\n",
    "        goodnesses = []\n",
    "        for layer in self.layers:\n",
    "            goodness = layer.compute_goodness(x)\n",
    "            goodnesses.append(goodness)\n",
    "            x = layer(x)\n",
    "        return torch.stack(goodnesses, dim=1)\n",
    "\n",
    "    def predict(self, x):\n",
    "        goodness_per_label = []\n",
    "        for label in range(10):\n",
    "            h = inject_label_into_input(x, label)\n",
    "            goodness = self.forward(h).sum(1) \n",
    "            goodness_per_label.append(goodness.unsqueeze(1))\n",
    "        goodness_per_label = torch.cat(goodness_per_label, 1)\n",
    "        return goodness_per_label.argmax(1)\n",
    "\n",
    "def inject_label_into_input(x, y, classes=10): \n",
    "    # One-hot encoding\n",
    "    x_ = x.clone()\n",
    "    x_[:, :classes] *= 0.0\n",
    "    if isinstance(y, int):\n",
    "        y = torch.full((x.shape[0],), y, dtype=torch.long, device=x.device)\n",
    "    x_[range(x.shape[0]), y] = x.max()\n",
    "    return x_ \n",
    "\n",
    "def train(model, train_loader, optimizer, threshold, device):\n",
    "    model.train()\n",
    "    total_loss = 0\n",
    "    \n",
    "    for data, target in train_loader:\n",
    "        data, target = data.to(device), target.to(device)\n",
    "        data = data.view(data.size(0), -1)\n",
    "\n",
    "        x_pos = inject_label_into_input(data, target)\n",
    "        g_pos = model(x_pos).mean(1)\n",
    "\n",
    "        y_neg = torch.randint(0, 10, target.shape, device=device)\n",
    "        x_neg = inject_label_into_input(data, y_neg)\n",
    "        g_neg = model(x_neg).mean(1)\n",
    "\n",
    "        loss = torch.log(1 + torch.exp(torch.cat([-g_pos + threshold, g_neg - threshold]))).mean()\n",
    "\n",
    "        optimizer.zero_grad()\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "\n",
    "        total_loss += loss.item()\n",
    "\n",
    "    return total_loss / len(train_loader)\n",
    "\n",
    "def evaluate(model, test_loader, device):\n",
    "    model.eval()\n",
    "    correct = 0\n",
    "    total = 0\n",
    "    with torch.no_grad():\n",
    "        for data, target in test_loader:\n",
    "            data, target = data.to(device), target.to(device)\n",
    "            data = data.view(data.size(0), -1)\n",
    "            pred = model.predict(data)\n",
    "            correct += (pred == target).sum().item()\n",
    "            total += target.size(0)\n",
    "    \n",
    "    return 100. * correct / total\n",
    "\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "model = FFNetwork([784, 500, 500]).to(device) # 784 = 28*28 \n",
    "optimizer = optim.Adam(model.parameters(), lr=LEARNING_RATE, weight_decay=WEIGHT_DECAY)\n",
    "\n",
    "train_losses = []\n",
    "test_accuracies = []\n",
    "\n",
    "for epoch in range(EPOCHS):\n",
    "    train_loss = train(model, train_loader, optimizer, THRESHOLD, device)\n",
    "    accuracy = evaluate(model, test_loader, device)\n",
    "    print(f\"Epoch {epoch+1}/{EPOCHS}, Loss: {train_loss:.4f}, Accuracy: {accuracy:.2f}%\")\n",
    "    \n",
    "    train_losses.append(train_loss)\n",
    "    test_accuracies.append(accuracy)\n",
    "\n",
    "# Visualize\n",
    "fig, ax1 = plt.subplots(figsize=(10, 6))\n",
    "\n",
    "color = 'tab:red'\n",
    "ax1.set_xlabel('Epoch')\n",
    "ax1.set_ylabel('Loss', color=color)\n",
    "ax1.plot(train_losses, color=color)\n",
    "ax1.tick_params(axis='y', labelcolor=color)\n",
    "\n",
    "ax2 = ax1.twinx()\n",
    "color = 'tab:blue'\n",
    "ax2.set_ylabel('Accuracy (%)', color=color)\n",
    "ax2.plot(test_accuracies, color=color)\n",
    "ax2.tick_params(axis='y', labelcolor=color)\n",
    "\n",
    "fig.tight_layout()\n",
    "plt.title('Training Loss & Test Accuracy over Epochs')\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}