{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "# Quantum-Classical Hybrid Neural Network Comparison\n",
    "\n",
    "This notebook compares different neural network architectures including:\n",
    "- Classical Multi-Layer Perceptrons (MLP)\n",
    "- Quantum circuits with LINEAR output mapping\n",
    "- Quantum circuits with various GROUPING strategies"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "\n",
    "## 1. Import Required Libraries\n",
    "\n",
    " First, we'll import all necessary libraries for our experiment:\n",
    " - **PyTorch**: For neural network implementation and training\n",
    " - **Perceval**: For quantum circuit simulation\n",
    " - **Matplotlib/Seaborn**: For visualization\n",
    " - **Custom modules**: For quantum layers and MLP models\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-11-10T09:09:17.656122600Z",
     "start_time": "2025-11-10T09:09:17.418122Z"
    },
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from collections import defaultdict\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import perceval as pcvl\n",
    "import seaborn as sns\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "from sklearn.metrics import accuracy_score, classification_report, confusion_matrix\n",
    "from tqdm import tqdm\n",
    "\n",
    "%matplotlib inline\n",
    "\n",
    "from merlin import (\n",
    "    ComputationSpace,\n",
    "    LexGrouping,\n",
    "    MeasurementStrategy,\n",
    "    ModGrouping,\n",
    "    QuantumLayer,\n",
    ")\n",
    "from merlin.datasets import iris"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "p\n",
    " ## 2. Load and Prepare the Iris Dataset\n",
    "\n",
    " We'll use the classic Iris dataset for multi-class classification. This dataset is ideal for comparing different architectures as it provides a simple but non-trivial classification task.\n",
    "\n",
    " The dataset contains:\n",
    " - **4 features**: sepal length, sepal width, petal length, petal width\n",
    " - **3 classes**: Setosa, Versicolor, Virginica\n",
    " - **150 samples** total (split into train/test sets)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-11-10T09:09:17.672033200Z",
     "start_time": "2025-11-10T09:09:17.537523200Z"
    },
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training samples: 120\n",
      "Test samples: 30\n",
      "Features: 4\n",
      "Classes: 3\n"
     ]
    }
   ],
   "source": [
    "train_features, train_labels, train_metadata = iris.get_data_train()\n",
    "test_features, test_labels, test_metadata = iris.get_data_test()\n",
    "\n",
    "# Convert data to PyTorch tensors\n",
    "X_train = torch.FloatTensor(train_features)\n",
    "y_train = torch.LongTensor(train_labels)\n",
    "X_test = torch.FloatTensor(test_features)\n",
    "y_test = torch.LongTensor(test_labels)\n",
    "\n",
    "print(f\"Training samples: {X_train.shape[0]}\")\n",
    "print(f\"Test samples: {X_test.shape[0]}\")\n",
    "print(f\"Features: {X_train.shape[1]}\")\n",
    "print(f\"Classes: {len(torch.unique(y_train))}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "\n",
    " ## 3. Define Model Variants\n",
    "\n",
    " We'll test several model architectures to compare classical and quantum approaches:\n",
    "\n",
    " **Classical Models:**\n",
    " - `MLP`: Multi-layer perceptron with 8 hidden units, ReLU activation, and 0.1 dropout\n",
    "\n",
    " **Quantum Models:**\n",
    " - `LINEAR-7modes-nobunching`: Uses linear output mapping with 7 quantum modes and no-bunching constraint\n",
    " - `LEXGROUPING-7modes`: Uses lexicographic grouping strategy for output mapping\n",
    " - `MODGROUPING-7modes`: Uses modular grouping strategy for output mapping\n",
    "\n",
    " Each model type has a distinct color and line style for visualization.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-11-10T09:09:17.687882200Z",
     "start_time": "2025-11-10T09:09:17.607913800Z"
    },
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from dataclasses import dataclass\n",
    "from typing import Literal\n",
    "\n",
    "import torch\n",
    "\n",
    "\n",
    "@dataclass\n",
    "class MLPConfig:\n",
    "    \"\"\"Configuration for Multi-Layer Perceptron\"\"\"\n",
    "\n",
    "    hidden_sizes: list[int] = None\n",
    "    dropout: float = 0.0\n",
    "    activation: Literal[\n",
    "        \"relu\", \"tanh\", \"sigmoid\", \"leaky_relu\", \"elu\", \"gelu\", \"selu\"\n",
    "    ] = \"relu\"\n",
    "    normalization: Literal[\"batch\", \"layer\"] | None = None\n",
    "\n",
    "\n",
    "class MLP(nn.Module):\n",
    "    \"\"\"\n",
    "    An enhanced Multi-Layer Perceptron (MLP) implementation with customizable options.\n",
    "\n",
    "    Features:\n",
    "    - Multiple activation functions\n",
    "    - Batch/Layer normalization options\n",
    "    - Customizable weight initialization\n",
    "\n",
    "    Args:\n",
    "        input_size (int): Dimension of the input features\n",
    "        output_size (int): Dimension of the output\n",
    "        config (MLPConfig): Configuration for the network architecture\n",
    "    \"\"\"\n",
    "\n",
    "    # Class-level mapping of available activation functions\n",
    "    ACTIVATIONS = {\n",
    "        \"relu\": nn.ReLU(),\n",
    "        \"tanh\": nn.Tanh(),\n",
    "        \"sigmoid\": nn.Sigmoid(),\n",
    "        \"leaky_relu\": nn.LeakyReLU(),\n",
    "        \"elu\": nn.ELU(),\n",
    "        \"gelu\": nn.GELU(),\n",
    "        \"selu\": nn.SELU(),\n",
    "    }\n",
    "\n",
    "    def __init__(self, input_size: int, output_size: int, config: MLPConfig):\n",
    "        super().__init__()\n",
    "        last_layer_size = input_size\n",
    "        layers = []\n",
    "\n",
    "        self.config = config\n",
    "\n",
    "        # Validate activation function\n",
    "        if config.activation not in self.ACTIVATIONS:\n",
    "            raise ValueError(f\"Unsupported activation function: {config.activation}\")\n",
    "\n",
    "        # Build network architecture\n",
    "        for layer_size in config.hidden_sizes:\n",
    "            # Add linear layer\n",
    "            layers.append(nn.Linear(last_layer_size, layer_size))\n",
    "\n",
    "            # Add normalization if specified\n",
    "            if config.normalization == \"batch\":\n",
    "                layers.append(nn.BatchNorm1d(layer_size))\n",
    "            elif config.normalization == \"layer\":\n",
    "                layers.append(nn.LayerNorm(layer_size))\n",
    "\n",
    "            # Add activation function\n",
    "            layers.append(self.ACTIVATIONS[config.activation])\n",
    "\n",
    "            # Add dropout if specified\n",
    "            if config.dropout > 0:\n",
    "                layers.append(nn.Dropout(config.dropout))\n",
    "\n",
    "            last_layer_size = layer_size\n",
    "\n",
    "        # Add final output layer\n",
    "        layers.append(nn.Linear(last_layer_size, output_size))\n",
    "\n",
    "        # Create sequential model from layers\n",
    "        self.network = nn.Sequential(*layers)\n",
    "\n",
    "    def initialize_weights(self, method=\"xavier\"):\n",
    "        \"\"\"\n",
    "        Initialize network weights using the specified method.\n",
    "\n",
    "        Args:\n",
    "            method (str): Initialization method ('xavier' or 'kaiming') (default: 'xavier')\n",
    "        \"\"\"\n",
    "\n",
    "        for layer in self.network:\n",
    "            if isinstance(layer, nn.Linear):\n",
    "                if method == \"xavier\":\n",
    "                    nn.init.xavier_uniform_(layer.weight)\n",
    "                    nn.init.zeros_(layer.bias)\n",
    "                elif method == \"kaiming\":\n",
    "                    nn.init.kaiming_normal_(layer.weight, nonlinearity=\"relu\")\n",
    "                    nn.init.zeros_(layer.bias)\n",
    "                else:\n",
    "                    raise ValueError(f\"Unsupported initialization method: {method}\")\n",
    "\n",
    "    def forward(self, x):\n",
    "        \"\"\"\n",
    "        Forward pass of the network.\n",
    "\n",
    "        Args:\n",
    "            x (torch.Tensor): Input tensor with shape (batch_size, input_size)\n",
    "\n",
    "        Returns:\n",
    "            torch.Tensor: Output tensor with shape (batch_size, output_size)\n",
    "        \"\"\"\n",
    "        return self.network(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-11-10T09:09:17.689136600Z",
     "start_time": "2025-11-10T09:09:17.664857200Z"
    },
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def get_model_variants():\n",
    "    \"\"\"Define different variants for each model type\"\"\"\n",
    "    # Define consistent colors for each model type\n",
    "    MODEL_COLORS = {\n",
    "        \"MLP\": \"#1f77b4\",  # Blue\n",
    "        \"LINEAR\": \"#2ca02c\",  # Green\n",
    "        \"GROUPING\": \"#ff7f0e\",  # Orange\n",
    "    }\n",
    "\n",
    "    # Define line styles for variants\n",
    "    LINE_STYLES = [\"--\", \"-\", \":\", \"-.\"]\n",
    "\n",
    "    variants = {\n",
    "        \"MLP\": [\n",
    "            {\n",
    "                \"name\": \"MLP\",\n",
    "                \"config\": MLPConfig(hidden_sizes=[8], dropout=0.1, activation=\"relu\"),\n",
    "                \"color\": MODEL_COLORS[\"MLP\"],\n",
    "                \"linestyle\": LINE_STYLES[0],\n",
    "            }\n",
    "        ],\n",
    "        \"LINEAR\": [\n",
    "            {\n",
    "                \"name\": \"LINEAR-7modes-nobunching\",\n",
    "                \"config\": {\n",
    "                    \"m\": 7,\n",
    "                    \"post_processing\": \"linear\",\n",
    "                    \"no_bunching\": True,\n",
    "                },\n",
    "                \"color\": MODEL_COLORS[\"LINEAR\"],\n",
    "                \"linestyle\": LINE_STYLES[1],\n",
    "            }\n",
    "        ],\n",
    "        \"GROUPING\": [\n",
    "            {\n",
    "                \"name\": \"LEXGROUPING-7modes\",\n",
    "                \"config\": {\n",
    "                    \"m\": 6,\n",
    "                    \"post_processing\": \"lex\",\n",
    "                },\n",
    "                \"color\": MODEL_COLORS[\"GROUPING\"],\n",
    "                \"linestyle\": LINE_STYLES[2],\n",
    "            },\n",
    "            {\n",
    "                \"name\": \"MODGROUPING-7modes\",\n",
    "                \"config\": {\n",
    "                    \"m\": 6,\n",
    "                    \"post_processing\": \"mod\",\n",
    "                },\n",
    "                \"color\": MODEL_COLORS[\"GROUPING\"],\n",
    "                \"linestyle\": LINE_STYLES[1],\n",
    "            },\n",
    "        ],\n",
    "    }\n",
    "    return variants"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    " ## 4. Quantum Circuit Architecture\n",
    "\n",
    " The quantum circuit is composed of three main sections:\n",
    "\n",
    " 1. **Left interferometer (WL)**: Performs initial quantum state transformation using beam splitters and phase shifters\n",
    " 2. **Variable phase shifters**: Encodes the 4 input features into quantum phases\n",
    " 3. **Right interferometer (WR)**: Performs final quantum state transformation\n",
    "\n",
    " The interferometers use a rectangular arrangement of optical elements, where each layer consists of beam splitters (BS) and programmable phase shifters (PS). The phase shifters contain trainable parameters that are optimized during training.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-11-10T09:09:17.719481900Z",
     "start_time": "2025-11-10T09:09:17.687882200Z"
    },
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def create_quantum_circuit(m):\n",
    "    \"\"\"Create quantum circuit with specified number of modes\n",
    "\n",
    "    Parameters:\n",
    "    -----------\n",
    "    m : int\n",
    "        Number of quantum modes in the circuit\n",
    "\n",
    "    Returns:\n",
    "    --------\n",
    "    pcvl.Circuit\n",
    "        Complete quantum circuit with trainable parameters\n",
    "    \"\"\"\n",
    "    # Left interferometer with trainable parameters\n",
    "    wl = pcvl.GenericInterferometer(\n",
    "        m,\n",
    "        lambda i: (\n",
    "            pcvl.BS()\n",
    "            // pcvl.PS(pcvl.P(f\"theta_li{i}\"))\n",
    "            // pcvl.BS()\n",
    "            // pcvl.PS(pcvl.P(f\"theta_lo{i}\"))\n",
    "        ),\n",
    "        shape=pcvl.InterferometerShape.RECTANGLE,\n",
    "    )\n",
    "\n",
    "    # Variable phase shifters for input encoding\n",
    "    c_var = pcvl.Circuit(m)\n",
    "    for i in range(4):  # 4 input features\n",
    "        px = pcvl.P(f\"px{i + 1}\")\n",
    "        c_var.add(i + (m - 4) // 2, pcvl.PS(px))\n",
    "\n",
    "    # Right interferometer with trainable parameters\n",
    "    wr = pcvl.GenericInterferometer(\n",
    "        m,\n",
    "        lambda i: (\n",
    "            pcvl.BS()\n",
    "            // pcvl.PS(pcvl.P(f\"theta_ri{i}\"))\n",
    "            // pcvl.BS()\n",
    "            // pcvl.PS(pcvl.P(f\"theta_ro{i}\"))\n",
    "        ),\n",
    "        shape=pcvl.InterferometerShape.RECTANGLE,\n",
    "    )\n",
    "\n",
    "    # Combine all components\n",
    "    c = pcvl.Circuit(m)\n",
    "    c.add(0, wl, merge=True)\n",
    "    c.add(0, c_var, merge=True)\n",
    "    c.add(0, wr, merge=True)\n",
    "\n",
    "    return c"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "\n",
    " ## 5. Model Factory Functions\n",
    "\n",
    " These utility functions handle model creation and parameter counting. The `create_model` function serves as a factory that instantiates either classical MLP models or quantum models based on the specified configuration.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-11-10T09:09:17.719481900Z",
     "start_time": "2025-11-10T09:09:17.703670Z"
    },
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def create_model(model_type, variant):\n",
    "    \"\"\"Create model instance based on type and variant\n",
    "\n",
    "    Parameters:\n",
    "    -----------\n",
    "    model_type : str\n",
    "        Type of model ('MLP' or quantum model types)\n",
    "    variant : dict\n",
    "        Configuration dictionary for the model variant\n",
    "\n",
    "    Returns:\n",
    "    --------\n",
    "    nn.Module\n",
    "        PyTorch model instance\n",
    "    \"\"\"\n",
    "    if model_type == \"MLP\":\n",
    "        return MLP(input_size=4, output_size=3, config=variant[\"config\"])\n",
    "    else:\n",
    "        m = variant[\"config\"][\"m\"]\n",
    "        no_bunching = variant[\"config\"].get(\"no_bunching\", False)\n",
    "        post_processing = variant[\"config\"].get(\"post_processing\")\n",
    "        c = create_quantum_circuit(m)\n",
    "        quantum_layer = QuantumLayer(\n",
    "            input_size=4,\n",
    "            circuit=c,\n",
    "            trainable_parameters=[\"theta\"],\n",
    "            input_parameters=[\"px\"],\n",
    "            input_state=[1, 0] * (m // 2) + [0] * (m % 2),\n",
    "            measurement_strategy=MeasurementStrategy.probs(\n",
    "                computation_space=ComputationSpace.UNBUNCHED\n",
    "                if no_bunching\n",
    "                else ComputationSpace.FOCK\n",
    "            ),\n",
    "        )\n",
    "\n",
    "        if post_processing == \"linear\":\n",
    "            head = nn.Linear(quantum_layer.output_size, 3)\n",
    "            return nn.Sequential(quantum_layer, head)\n",
    "        if post_processing == \"lex\":\n",
    "            return nn.Sequential(\n",
    "                quantum_layer,\n",
    "                LexGrouping(quantum_layer.output_size, 3),\n",
    "            )\n",
    "        if post_processing == \"mod\":\n",
    "            return nn.Sequential(\n",
    "                quantum_layer,\n",
    "                ModGrouping(quantum_layer.output_size, 3),\n",
    "            )\n",
    "        return quantum_layer\n",
    "\n",
    "\n",
    "def count_parameters(model):\n",
    "    \"\"\"Count trainable parameters in a PyTorch model\"\"\"\n",
    "    return sum(p.numel() for p in model.parameters() if p.requires_grad)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "\n",
    "\n",
    " ## 6. Training Function\n",
    "\n",
    " The training function implements a standard supervised learning loop with the following characteristics:\n",
    "\n",
    " - **Optimizer**: Adam with learning rate 0.02\n",
    " - **Loss function**: Cross-entropy for multi-class classification\n",
    " - **Batch size**: 32 samples\n",
    " - **Epochs**: 50 (default)\n",
    "\n",
    " The function tracks training loss, training accuracy, and test accuracy at each epoch. After training completes, it generates a detailed classification report.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-11-10T09:09:17.743348600Z",
     "start_time": "2025-11-10T09:09:17.735217100Z"
    },
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def train_model(\n",
    "    model,\n",
    "    X_train,\n",
    "    y_train,\n",
    "    X_test,\n",
    "    y_test,\n",
    "    model_name,\n",
    "    n_epochs=50,\n",
    "    batch_size=32,\n",
    "    lr=0.02,\n",
    "):\n",
    "    \"\"\"Train a model and track metrics\n",
    "\n",
    "    Parameters:\n",
    "    -----------\n",
    "    model : nn.Module\n",
    "        Model to train\n",
    "    X_train, y_train : torch.Tensor\n",
    "        Training data and labels\n",
    "    X_test, y_test : torch.Tensor\n",
    "        Test data and labels\n",
    "    model_name : str\n",
    "        Name for progress bar display\n",
    "    n_epochs : int\n",
    "        Number of training epochs\n",
    "    batch_size : int\n",
    "        Batch size for mini-batch training\n",
    "    lr : float\n",
    "        Learning rate\n",
    "\n",
    "    Returns:\n",
    "    --------\n",
    "    dict\n",
    "        Training results including losses, accuracies, and final report\n",
    "    \"\"\"\n",
    "    optimizer = torch.optim.Adam(model.parameters(), lr=lr)\n",
    "    criterion = nn.CrossEntropyLoss()\n",
    "\n",
    "    losses = []\n",
    "    train_accuracies = []\n",
    "    test_accuracies = []\n",
    "\n",
    "    model.train()\n",
    "\n",
    "    pbar = tqdm(range(n_epochs), leave=False, desc=f\"Training {model_name}\")\n",
    "    for _epoch in pbar:\n",
    "        # Shuffle training data\n",
    "        permutation = torch.randperm(X_train.size()[0])\n",
    "        total_loss = 0\n",
    "\n",
    "        # Mini-batch training\n",
    "        for i in range(0, X_train.size()[0], batch_size):\n",
    "            indices = permutation[i : i + batch_size]\n",
    "            batch_x, batch_y = X_train[indices], y_train[indices]\n",
    "\n",
    "            outputs = model(batch_x)\n",
    "            loss = criterion(outputs, batch_y)\n",
    "\n",
    "            optimizer.zero_grad()\n",
    "            loss.backward()\n",
    "            optimizer.step()\n",
    "\n",
    "            total_loss += loss.item()\n",
    "\n",
    "        avg_loss = total_loss / (X_train.size()[0] // batch_size)\n",
    "        losses.append(avg_loss)\n",
    "        pbar.set_description(f\"Training {model_name} - Loss: {avg_loss:.4f}\")\n",
    "\n",
    "        # Evaluation\n",
    "        model.eval()\n",
    "        with torch.no_grad():\n",
    "            # Training accuracy\n",
    "            train_outputs = model(X_train)\n",
    "            train_preds = torch.argmax(train_outputs, dim=1).numpy()\n",
    "            train_acc = accuracy_score(y_train.numpy(), train_preds)\n",
    "            train_accuracies.append(train_acc)\n",
    "\n",
    "            # Test accuracy\n",
    "            test_outputs = model(X_test)\n",
    "            test_preds = torch.argmax(test_outputs, dim=1).numpy()\n",
    "            test_acc = accuracy_score(y_test.numpy(), test_preds)\n",
    "            test_accuracies.append(test_acc)\n",
    "\n",
    "        model.train()\n",
    "\n",
    "    # Generate final classification report\n",
    "    model.eval()\n",
    "    with torch.no_grad():\n",
    "        final_test_outputs = model(X_test)\n",
    "        final_test_preds = torch.argmax(final_test_outputs, dim=1).numpy()\n",
    "        final_report = classification_report(y_test.numpy(), final_test_preds)\n",
    "\n",
    "    return {\n",
    "        \"losses\": losses,\n",
    "        \"train_accuracies\": train_accuracies,\n",
    "        \"test_accuracies\": test_accuracies,\n",
    "        \"final_test_acc\": test_accuracies[-1],\n",
    "        \"classification_report\": final_report,\n",
    "    }"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "\n",
    " ## 7. Multi-Run Training Function\n",
    "\n",
    " To ensure robust and statistically meaningful results, we train each model variant multiple times with different random initializations. This approach helps us:\n",
    "\n",
    " - Understand the variance in model performance\n",
    " - Identify which architectures are more stable\n",
    " - Avoid drawing conclusions from lucky/unlucky single runs\n",
    " - Calculate confidence intervals for performance metrics\n",
    "\n",
    " The function tracks both individual run results and aggregate statistics across all runs.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-11-10T09:09:17.775170100Z",
     "start_time": "2025-11-10T09:09:17.756740300Z"
    },
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def train_all_variants(X_train, y_train, X_test, y_test, num_runs=5):\n",
    "    \"\"\"Train all model variants multiple times and return results\n",
    "\n",
    "    Parameters:\n",
    "    -----------\n",
    "    X_train, y_train : torch.Tensor\n",
    "        Training data and labels\n",
    "    X_test, y_test : torch.Tensor\n",
    "        Test data and labels\n",
    "    num_runs : int\n",
    "        Number of independent training runs per variant\n",
    "\n",
    "    Returns:\n",
    "    --------\n",
    "    tuple\n",
    "        (all_results, best_models) - Complete results and best performing models\n",
    "    \"\"\"\n",
    "    variants = get_model_variants()\n",
    "    all_results = defaultdict(dict)\n",
    "    best_models = {}\n",
    "\n",
    "    for model_type, model_variants in variants.items():\n",
    "        print(f\"\\n\\nTraining {model_type} variants:\")\n",
    "        best_acc = 0\n",
    "\n",
    "        for variant in model_variants:\n",
    "            print(\n",
    "                f\"\\nTraining {variant['name']}... ({count_parameters(create_model(model_type, variant))} parameters)\"\n",
    "            )\n",
    "\n",
    "            # Store results from multiple runs\n",
    "            variant_runs = []\n",
    "\n",
    "            for run in range(num_runs):\n",
    "                model = create_model(model_type, variant)\n",
    "                print(f\"  Run {run + 1}/{num_runs}...\")\n",
    "\n",
    "                results = train_model(\n",
    "                    model,\n",
    "                    X_train,\n",
    "                    y_train,\n",
    "                    X_test,\n",
    "                    y_test,\n",
    "                    f\"{variant['name']}-run{run + 1}\",\n",
    "                )\n",
    "                results[\"model\"] = model\n",
    "                results[\"color\"] = variant[\"color\"]\n",
    "                results[\"linestyle\"] = variant[\"linestyle\"]\n",
    "                variant_runs.append(results)\n",
    "\n",
    "                # Track best model for each type\n",
    "                if results[\"final_test_acc\"] > best_acc:\n",
    "                    best_acc = results[\"final_test_acc\"]\n",
    "                    best_models[model_type] = {\n",
    "                        \"name\": variant[\"name\"],\n",
    "                        \"model\": model,\n",
    "                        \"results\": results,\n",
    "                    }\n",
    "\n",
    "            # Store all runs for this variant\n",
    "            all_results[model_type][variant[\"name\"]] = {\n",
    "                \"runs\": variant_runs,\n",
    "                \"color\": variant[\"color\"],\n",
    "                \"linestyle\": variant[\"linestyle\"],\n",
    "                \"avg_final_test_acc\": sum(run[\"final_test_acc\"] for run in variant_runs)\n",
    "                / num_runs,\n",
    "            }\n",
    "\n",
    "    return all_results, best_models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "\n",
    " ## 8. Visualization Functions\n",
    "\n",
    " ### Training Curves Visualization\n",
    "\n",
    " This function creates a comprehensive view of the training process by plotting three key metrics:\n",
    "\n",
    " 1. **Training Loss**: Shows how well the model fits the training data\n",
    " 2. **Training Accuracy**: Indicates the model's performance on training samples\n",
    " 3. **Test Accuracy**: Reveals the model's generalization capability\n",
    "\n",
    " For each model variant, we display:\n",
    " - **Solid line**: Mean performance across all runs\n",
    " - **Shaded area**: Performance envelope (min to max values)\n",
    "\n",
    " This visualization helps identify overfitting, convergence speed, and stability.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-11-10T09:09:17.782612700Z",
     "start_time": "2025-11-10T09:09:17.759907600Z"
    },
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def plot_training_curves(all_results):\n",
    "    \"\"\"Plot training curves for all model variants with average and envelope\n",
    "\n",
    "    Shows three subplots:\n",
    "    1. Training loss over epochs\n",
    "    2. Training accuracy over epochs\n",
    "    3. Test accuracy over epochs\n",
    "\n",
    "    Each line represents the mean across runs, with shaded areas showing min/max envelope.\n",
    "    \"\"\"\n",
    "    fig, (ax1, ax2, ax3) = plt.subplots(1, 3, figsize=(20, 5))\n",
    "\n",
    "    # Plot each metric\n",
    "    for _model_type, variants in all_results.items():\n",
    "        for variant_name, variant_data in variants.items():\n",
    "            color = variant_data[\"color\"]\n",
    "            linestyle = variant_data[\"linestyle\"]\n",
    "            label = f\"{variant_name}\"\n",
    "\n",
    "            # Get data from all runs\n",
    "            losses_runs = [run[\"losses\"] for run in variant_data[\"runs\"]]\n",
    "            train_acc_runs = [run[\"train_accuracies\"] for run in variant_data[\"runs\"]]\n",
    "            test_acc_runs = [run[\"test_accuracies\"] for run in variant_data[\"runs\"]]\n",
    "\n",
    "            # Calculate mean values across all runs\n",
    "            epochs = len(losses_runs[0])\n",
    "            mean_losses = [\n",
    "                sum(run[i] for run in losses_runs) / len(losses_runs)\n",
    "                for i in range(epochs)\n",
    "            ]\n",
    "            mean_train_acc = [\n",
    "                sum(run[i] for run in train_acc_runs) / len(train_acc_runs)\n",
    "                for i in range(epochs)\n",
    "            ]\n",
    "            mean_test_acc = [\n",
    "                sum(run[i] for run in test_acc_runs) / len(test_acc_runs)\n",
    "                for i in range(epochs)\n",
    "            ]\n",
    "\n",
    "            # Calculate min and max values for the envelope\n",
    "            min_losses = [min(run[i] for run in losses_runs) for i in range(epochs)]\n",
    "            max_losses = [max(run[i] for run in losses_runs) for i in range(epochs)]\n",
    "\n",
    "            min_train_acc = [\n",
    "                min(run[i] for run in train_acc_runs) for i in range(epochs)\n",
    "            ]\n",
    "            max_train_acc = [\n",
    "                max(run[i] for run in train_acc_runs) for i in range(epochs)\n",
    "            ]\n",
    "\n",
    "            min_test_acc = [min(run[i] for run in test_acc_runs) for i in range(epochs)]\n",
    "            max_test_acc = [max(run[i] for run in test_acc_runs) for i in range(epochs)]\n",
    "\n",
    "            # Plot mean lines\n",
    "            ax1.plot(\n",
    "                mean_losses, label=label, color=color, linestyle=linestyle, linewidth=2\n",
    "            )\n",
    "            ax2.plot(\n",
    "                mean_train_acc,\n",
    "                label=label,\n",
    "                color=color,\n",
    "                linestyle=linestyle,\n",
    "                linewidth=2,\n",
    "            )\n",
    "            ax3.plot(\n",
    "                mean_test_acc,\n",
    "                label=label,\n",
    "                color=color,\n",
    "                linestyle=linestyle,\n",
    "                linewidth=2,\n",
    "            )\n",
    "\n",
    "            # Plot envelopes (filled area between min and max)\n",
    "            epochs_range = range(epochs)\n",
    "            ax1.fill_between(\n",
    "                epochs_range, min_losses, max_losses, color=color, alpha=0.2\n",
    "            )\n",
    "            ax2.fill_between(\n",
    "                epochs_range, min_train_acc, max_train_acc, color=color, alpha=0.2\n",
    "            )\n",
    "            ax3.fill_between(\n",
    "                epochs_range, min_test_acc, max_test_acc, color=color, alpha=0.2\n",
    "            )\n",
    "\n",
    "    # Customize plots\n",
    "    for ax, title in zip(\n",
    "        [ax1, ax2, ax3],\n",
    "        [\"Training Loss\", \"Training Accuracy\", \"Test Accuracy\"],\n",
    "        strict=False,\n",
    "    ):\n",
    "        ax.set_title(title, fontsize=12, pad=10)\n",
    "        ax.set_xlabel(\"Epoch\", fontsize=10)\n",
    "        ax.set_ylabel(title.split()[-1], fontsize=10)\n",
    "        ax.legend(fontsize=8, bbox_to_anchor=(1.05, 1), loc=\"upper left\")\n",
    "        ax.grid(True, linestyle=\"--\", alpha=0.7)\n",
    "        ax.spines[\"top\"].set_visible(False)\n",
    "        ax.spines[\"right\"].set_visible(False)\n",
    "\n",
    "    plt.tight_layout()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "\n",
    " ### Confusion Matrix Visualization\n",
    "\n",
    " Confusion matrices provide detailed insights into classification performance by showing:\n",
    " - Which classes are correctly classified (diagonal elements)\n",
    " - Which classes are confused with each other (off-diagonal elements)\n",
    " - Overall performance patterns for each model architecture\n",
    "\n",
    " We display the confusion matrix for the best-performing model of each type.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-11-10T09:09:17.798960200Z",
     "start_time": "2025-11-10T09:09:17.793299500Z"
    },
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def plot_best_confusion_matrices(best_models, X_test, y_test):\n",
    "    \"\"\"Plot confusion matrices for the best model of each type\"\"\"\n",
    "    fig, axes = plt.subplots(1, 3, figsize=(20, 5))\n",
    "    class_names = [\"setosa\", \"versicolor\", \"virginica\"]\n",
    "\n",
    "    for idx, (model_type, best) in enumerate(best_models.items()):\n",
    "        model = best[\"model\"]\n",
    "        model.eval()\n",
    "        with torch.no_grad():\n",
    "            outputs = model(X_test)\n",
    "            predictions = torch.argmax(outputs, dim=1).numpy()\n",
    "\n",
    "        cm = confusion_matrix(y_test.numpy(), predictions)\n",
    "        sns.heatmap(\n",
    "            cm,\n",
    "            annot=True,\n",
    "            fmt=\"d\",\n",
    "            cmap=\"Blues\",\n",
    "            xticklabels=class_names,\n",
    "            yticklabels=class_names,\n",
    "            ax=axes[idx],\n",
    "        )\n",
    "        axes[idx].set_title(f\"Best {model_type}\\n({best['name']})\")\n",
    "        axes[idx].set_xlabel(\"Predicted\")\n",
    "        axes[idx].set_ylabel(\"True\")\n",
    "        plt.setp(axes[idx].get_xticklabels(), rotation=45)\n",
    "        plt.setp(axes[idx].get_yticklabels(), rotation=45)\n",
    "\n",
    "    plt.tight_layout()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "\n",
    " ## 9. Results Summary Function\n",
    "\n",
    " This function provides a comprehensive statistical analysis of the experimental results, including:\n",
    "\n",
    " - **Parameter efficiency**: Number of trainable parameters for each model\n",
    " - **Performance statistics**: Mean accuracy, standard deviation, min/max values\n",
    " - **Detailed classification reports**: Precision, recall, and F1-score for each class\n",
    "\n",
    " These metrics help in making informed decisions about model selection based on the specific requirements of your application.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-11-10T09:09:17.846457400Z",
     "start_time": "2025-11-10T09:09:17.819929900Z"
    },
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def print_comparison_results(all_results, best_models):\n",
    "    \"\"\"Print detailed comparison of all models and variants with multi-run statistics\"\"\"\n",
    "    print(\"\\n----- Model Comparison Results -----\")\n",
    "\n",
    "    # Print results for all variants\n",
    "    print(\"\\nAll Model Variants Results (averaged over multiple runs):\")\n",
    "    for model_type, variants in all_results.items():\n",
    "        print(f\"\\n{model_type} Variants:\")\n",
    "        for variant_name, variant_data in variants.items():\n",
    "            print(f\"\\n{variant_name}:\")\n",
    "            print(f\"Parameters: {count_parameters(variant_data['runs'][0]['model'])}\")\n",
    "\n",
    "            # Calculate statistics across runs\n",
    "            final_accs = [run[\"final_test_acc\"] for run in variant_data[\"runs\"]]\n",
    "            avg_acc = sum(final_accs) / len(final_accs)\n",
    "            min_acc = min(final_accs)\n",
    "            max_acc = max(final_accs)\n",
    "            std_acc = (\n",
    "                sum((acc - avg_acc) ** 2 for acc in final_accs) / len(final_accs)\n",
    "            ) ** 0.5\n",
    "\n",
    "            print(\n",
    "                f\"Final Test Accuracy: {avg_acc:.4f} ± {std_acc:.4f} (min: {min_acc:.4f}, max: {max_acc:.4f})\"\n",
    "            )\n",
    "\n",
    "    # Print best model results\n",
    "    print(\"\\nBest Models:\")\n",
    "    for model_type, best in best_models.items():\n",
    "        print(f\"\\nBest {model_type} Model: {best['name']}\")\n",
    "        print(f\"Final Test Accuracy: {best['results']['final_test_acc']:.4f}\")\n",
    "        print(f\"Classification Report:\\n{best['results']['classification_report']}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "\n",
    " ## 10. Run the Complete Experiment\n",
    "\n",
    " Now we'll execute the full experimental pipeline. The experiment consists of:\n",
    "\n",
    " 1. **Multiple training runs**: Each model variant is trained 5 times with different random seeds\n",
    " 2. **Performance tracking**: All metrics are recorded for statistical analysis\n",
    " 3. **Visualization**: Training curves and confusion matrices are generated\n",
    " 4. **Statistical summary**: Detailed performance statistics are computed\n",
    "\n",
    " This comprehensive approach ensures reliable and reproducible results.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-11-10T09:10:55.159331300Z",
     "start_time": "2025-11-10T09:09:17.830714100Z"
    },
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Starting training process...\n",
      "\n",
      "\n",
      "Training MLP variants:\n",
      "\n",
      "Training MLP... (67 parameters)\n",
      "  Run 1/5...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Run 2/5...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                        \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Run 3/5...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                        \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Run 4/5...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                        \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Run 5/5...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                        \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "Training LINEAR variants:\n",
      "\n",
      "Training LINEAR-7modes-nobunching... (192 parameters)\n",
      "  Run 1/5...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                                      \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Run 2/5...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                                      \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Run 3/5...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                                      \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Run 4/5...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                                      \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Run 5/5...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                                      \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "Training GROUPING variants:\n",
      "\n",
      "Training LEXGROUPING-7modes... (60 parameters)\n",
      "  Run 1/5...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/lfvigneux/miniconda3/envs/MerLin_dev/lib/python3.12/site-packages/sklearn/metrics/_classification.py:1833: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", result.shape[0])\n",
      "/Users/lfvigneux/miniconda3/envs/MerLin_dev/lib/python3.12/site-packages/sklearn/metrics/_classification.py:1833: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", result.shape[0])\n",
      "/Users/lfvigneux/miniconda3/envs/MerLin_dev/lib/python3.12/site-packages/sklearn/metrics/_classification.py:1833: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", result.shape[0])\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Run 2/5...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                                \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Run 3/5...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                                \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Run 4/5...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                                \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Run 5/5...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                                \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Training MODGROUPING-7modes... (60 parameters)\n",
      "  Run 1/5...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/lfvigneux/miniconda3/envs/MerLin_dev/lib/python3.12/site-packages/sklearn/metrics/_classification.py:1833: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", result.shape[0])\n",
      "/Users/lfvigneux/miniconda3/envs/MerLin_dev/lib/python3.12/site-packages/sklearn/metrics/_classification.py:1833: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", result.shape[0])\n",
      "/Users/lfvigneux/miniconda3/envs/MerLin_dev/lib/python3.12/site-packages/sklearn/metrics/_classification.py:1833: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", result.shape[0])\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Run 2/5...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/lfvigneux/miniconda3/envs/MerLin_dev/lib/python3.12/site-packages/sklearn/metrics/_classification.py:1833: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", result.shape[0])\n",
      "/Users/lfvigneux/miniconda3/envs/MerLin_dev/lib/python3.12/site-packages/sklearn/metrics/_classification.py:1833: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", result.shape[0])\n",
      "/Users/lfvigneux/miniconda3/envs/MerLin_dev/lib/python3.12/site-packages/sklearn/metrics/_classification.py:1833: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", result.shape[0])\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Run 3/5...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                                                                \r"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Run 4/5...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/lfvigneux/miniconda3/envs/MerLin_dev/lib/python3.12/site-packages/sklearn/metrics/_classification.py:1833: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", result.shape[0])\n",
      "/Users/lfvigneux/miniconda3/envs/MerLin_dev/lib/python3.12/site-packages/sklearn/metrics/_classification.py:1833: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", result.shape[0])\n",
      "/Users/lfvigneux/miniconda3/envs/MerLin_dev/lib/python3.12/site-packages/sklearn/metrics/_classification.py:1833: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", result.shape[0])\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  Run 5/5...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/lfvigneux/miniconda3/envs/MerLin_dev/lib/python3.12/site-packages/sklearn/metrics/_classification.py:1833: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", result.shape[0])\n",
      "/Users/lfvigneux/miniconda3/envs/MerLin_dev/lib/python3.12/site-packages/sklearn/metrics/_classification.py:1833: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", result.shape[0])\n",
      "/Users/lfvigneux/miniconda3/envs/MerLin_dev/lib/python3.12/site-packages/sklearn/metrics/_classification.py:1833: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, f\"{metric.capitalize()} is\", result.shape[0])\n"
     ]
    }
   ],
   "source": [
    "# Set number of independent runs per model variant\n",
    "num_runs = 5\n",
    "\n",
    "# Train all variants\n",
    "print(\"Starting training process...\")\n",
    "all_results, best_models = train_all_variants(\n",
    "    X_train, y_train, X_test, y_test, num_runs=num_runs\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    " ### Visualize Training Progress\n",
    "\n",
    " The following plots show how each model's performance evolves during training. Pay attention to:\n",
    " - **Convergence speed**: How quickly each model reaches its best performance\n",
    " - **Stability**: Width of the shaded area indicates variance across runs\n",
    " - **Final performance**: Where each model plateaus"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-11-10T09:10:56.751500900Z",
     "start_time": "2025-11-10T09:10:55.159331300Z"
    },
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x500 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_training_curves(all_results)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "\n",
    " ### Confusion Matrices for Best Models\n",
    "\n",
    " These confusion matrices reveal the classification patterns of the best-performing model from each architecture type. Look for:\n",
    " - **Perfect classification**: Bright blue diagonal with zeros elsewhere\n",
    " - **Common confusions**: Off-diagonal elements show which classes are hard to distinguish\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-11-10T09:10:56.752504500Z",
     "start_time": "2025-11-10T09:10:56.664051200Z"
    },
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "----- Model Comparison Results -----\n",
      "\n",
      "All Model Variants Results (averaged over multiple runs):\n",
      "\n",
      "MLP Variants:\n",
      "\n",
      "MLP:\n",
      "Parameters: 67\n",
      "Final Test Accuracy: 0.9267 ± 0.0249 (min: 0.9000, max: 0.9667)\n",
      "\n",
      "LINEAR Variants:\n",
      "\n",
      "LINEAR-7modes-nobunching:\n",
      "Parameters: 192\n",
      "Final Test Accuracy: 0.9200 ± 0.0452 (min: 0.8667, max: 0.9667)\n",
      "\n",
      "GROUPING Variants:\n",
      "\n",
      "LEXGROUPING-7modes:\n",
      "Parameters: 60\n",
      "Final Test Accuracy: 0.8333 ± 0.1116 (min: 0.6333, max: 0.9667)\n",
      "\n",
      "MODGROUPING-7modes:\n",
      "Parameters: 60\n",
      "Final Test Accuracy: 0.6733 ± 0.0800 (min: 0.6333, max: 0.8333)\n",
      "\n",
      "Best Models:\n",
      "\n",
      "Best MLP Model: MLP\n",
      "Final Test Accuracy: 0.9667\n",
      "Classification Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00        13\n",
      "           1       0.86      1.00      0.92         6\n",
      "           2       1.00      0.91      0.95        11\n",
      "\n",
      "    accuracy                           0.97        30\n",
      "   macro avg       0.95      0.97      0.96        30\n",
      "weighted avg       0.97      0.97      0.97        30\n",
      "\n",
      "\n",
      "Best LINEAR Model: LINEAR-7modes-nobunching\n",
      "Final Test Accuracy: 0.9667\n",
      "Classification Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00        13\n",
      "           1       0.86      1.00      0.92         6\n",
      "           2       1.00      0.91      0.95        11\n",
      "\n",
      "    accuracy                           0.97        30\n",
      "   macro avg       0.95      0.97      0.96        30\n",
      "weighted avg       0.97      0.97      0.97        30\n",
      "\n",
      "\n",
      "Best GROUPING Model: LEXGROUPING-7modes\n",
      "Final Test Accuracy: 0.9667\n",
      "Classification Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00        13\n",
      "           1       0.86      1.00      0.92         6\n",
      "           2       1.00      0.91      0.95        11\n",
      "\n",
      "    accuracy                           0.97        30\n",
      "   macro avg       0.95      0.97      0.96        30\n",
      "weighted avg       0.97      0.97      0.97        30\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print_comparison_results(all_results, best_models)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    " ## 11. Key Findings and Conclusions\n",
    "\n",
    " Based on the experimental results, we can draw several important conclusions:\n",
    "\n",
    " **Model Complexity vs Performance:**\n",
    " - Compare the number of parameters required by each architecture\n",
    " - Assess whether quantum models achieve similar performance with fewer parameters\n",
    "\n",
    " **Training Stability:**\n",
    " - The variance across multiple runs indicates how sensitive each model is to initialization\n",
    " - Lower variance suggests more reliable training\n",
    "\n",
    " **Generalization Capability:**\n",
    " - The gap between training and test accuracy reveals overfitting tendencies\n",
    " - Smaller gaps indicate better generalization\n",
    "\n",
    " **Classification Patterns:**\n",
    " - Confusion matrices show which flower species are most difficult to distinguish\n",
    " - This can guide feature engineering or model selection\n",
    "\n",
    " **Practical Considerations:**\n",
    "\n",
    " For deployment, consider:\n",
    " - **Classical MLPs**: Well-understood, fast inference, established deployment pipelines\n",
    " - **Quantum models**: Potentially more parameter-efficient, may offer advantages on quantum hardware\n",
    "\n",
    " **Future Research Directions:**\n",
    "\n",
    " 1. **Scaling to larger datasets**: Test on more complex classification tasks\n",
    " 2. **Noise modeling**: Investigate performance under realistic quantum noise conditions\n",
    " 3. **Hybrid architectures**: Combine classical and quantum layers\n",
    " 4. **Hardware implementation**: Evaluate on actual quantum photonic devices\n",
    " 5. **Feature encoding strategies**: Explore different ways to encode classical data into quantum states"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "MerLin_dev",
   "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": "3.12.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}