{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "header",
   "metadata": {},
   "source": [
    "# Training Recurrent Neural Networks: Integrator Task\n",
    "\n",
    "This tutorial demonstrates how to train a recurrent neural network (RNN) to perform an **integration task** - a fundamental computation in neuroscience where networks must accumulate evidence over time.\n",
    "\n",
    "## Learning Objectives\n",
    "\n",
    "By the end of this tutorial, you will:\n",
    "- Understand the integration task and its importance\n",
    "- Build custom RNN cells in BrainState\n",
    "- Train RNNs on temporal tasks\n",
    "- Use trainable initial states\n",
    "- Apply L2 regularization to prevent overfitting\n",
    "- Visualize RNN predictions on time-series data\n",
    "\n",
    "## The Integration Task\n",
    "\n",
    "**Goal**: Given a noisy input signal, the network must compute the cumulative sum (integral) over time.\n",
    "\n",
    "```\n",
    "Input:  [x₁, x₂, x₃, ...]\n",
    "Output: [x₁, x₁+x₂, x₁+x₂+x₃, ...]\n",
    "```\n",
    "\n",
    "This task requires:\n",
    "- **Memory**: Remember past inputs\n",
    "- **Accumulation**: Continuously integrate information\n",
    "- **Robustness**: Handle noise in inputs\n",
    "\n",
    "**Applications**:\n",
    "- Evidence accumulation in decision-making\n",
    "- Position estimation from velocity\n",
    "- Financial modeling (cumulative returns)\n",
    "- Signal processing"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "setup",
   "metadata": {},
   "source": [
    "## Setup and Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "imports",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-10-11T09:58:07.603916Z",
     "start_time": "2025-10-11T09:58:04.111801Z"
    }
   },
   "outputs": [],
   "source": [
    "from typing import Callable\n",
    "\n",
    "import jax\n",
    "import jax.numpy as jnp\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "import brainstate\n",
    "import braintools\n",
    "\n",
    "# Set random seeds\n",
    "np.random.seed(42)\n",
    "brainstate.random.seed(42)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "config",
   "metadata": {},
   "source": [
    "## Configuration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "hyperparams",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-10-11T09:58:07.614207Z",
     "start_time": "2025-10-11T09:58:07.608297Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Configuration:\n",
      "  Sequence length: 25 steps\n",
      "  Batch size: 512\n",
      "  Hidden units: 100\n",
      "  Training batches: 2500\n"
     ]
    }
   ],
   "source": [
    "# Task parameters\n",
    "dt = 0.04              # Time step\n",
    "num_step = int(1.0 / dt)  # Steps per trial (25 steps for 1.0 time unit)\n",
    "num_batch = 512        # Batch size\n",
    "\n",
    "# Network parameters\n",
    "num_hidden = 100       # Hidden units in RNN\n",
    "\n",
    "# Training parameters\n",
    "learning_rate = 0.025\n",
    "lr_decay_rate = 0.99975\n",
    "l2_reg = 2e-4\n",
    "num_epochs = 5\n",
    "batches_per_epoch = 500\n",
    "\n",
    "print(f\"Configuration:\")\n",
    "print(f\"  Sequence length: {num_step} steps\")\n",
    "print(f\"  Batch size: {num_batch}\")\n",
    "print(f\"  Hidden units: {num_hidden}\")\n",
    "print(f\"  Training batches: {num_epochs * batches_per_epoch}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "data_section",
   "metadata": {},
   "source": [
    "## Generate Data\n",
    "\n",
    "### Data Generation Function\n",
    "\n",
    "We'll create random walk inputs and compute their cumulative sums as targets:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "data_gen",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-10-11T09:58:07.623573Z",
     "start_time": "2025-10-11T09:58:07.619211Z"
    }
   },
   "outputs": [],
   "source": [
    "@brainstate.transform.jit(static_argnums=2)\n",
    "def build_inputs_and_targets(mean=0.025, scale=0.01, batch_size=10):\n",
    "    \"\"\"Generate integration task data.\n",
    "    \n",
    "    Args:\n",
    "        mean: Mean of the random walk bias\n",
    "        scale: Standard deviation of noise\n",
    "        batch_size: Number of sequences\n",
    "        \n",
    "    Returns:\n",
    "        inputs: [num_step, batch_size, 1] - Input sequences\n",
    "        targets: [num_step, batch_size, 1] - Target cumulative sums\n",
    "    \"\"\"\n",
    "    # Create initial bias\n",
    "    sample = brainstate.random.normal(size=(1, batch_size, 1))\n",
    "    bias = mean * 2.0 * (sample - 0.5)\n",
    "    \n",
    "    # Generate white noise\n",
    "    samples = brainstate.random.normal(size=(num_step, batch_size, 1))\n",
    "    noise_t = scale / dt ** 0.5 * samples\n",
    "    \n",
    "    # Inputs = bias + noise\n",
    "    inputs = bias + noise_t\n",
    "    \n",
    "    # Targets = cumulative sum of inputs\n",
    "    targets = jnp.cumsum(inputs, axis=0)\n",
    "    \n",
    "    return inputs, targets\n",
    "\n",
    "\n",
    "def train_data():\n",
    "    \"\"\"Generator for training data.\"\"\"\n",
    "    for _ in range(batches_per_epoch * num_epochs):\n",
    "        yield build_inputs_and_targets(0.025, 0.01, num_batch)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "visualize_data",
   "metadata": {},
   "source": [
    "### Visualize Sample Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "plot_data",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-10-11T09:58:08.124430Z",
     "start_time": "2025-10-11T09:58:07.634037Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1400x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Generate one batch for visualization\n",
    "sample_inputs, sample_targets = build_inputs_and_targets(0.025, 0.01, 3)\n",
    "\n",
    "fig, axes = plt.subplots(1, 2, figsize=(14, 4))\n",
    "\n",
    "# Plot inputs\n",
    "for i in range(3):\n",
    "    axes[0].plot(sample_inputs[:, i, 0], alpha=0.7, label=f'Sample {i+1}')\n",
    "axes[0].set_xlabel('Time Step')\n",
    "axes[0].set_ylabel('Input Value')\n",
    "axes[0].set_title('Input Sequences (Random Walks)', fontweight='bold')\n",
    "axes[0].legend()\n",
    "axes[0].grid(True, alpha=0.3)\n",
    "\n",
    "# Plot targets\n",
    "for i in range(3):\n",
    "    axes[1].plot(sample_targets[:, i, 0], alpha=0.7, label=f'Sample {i+1}')\n",
    "axes[1].set_xlabel('Time Step')\n",
    "axes[1].set_ylabel('Cumulative Sum')\n",
    "axes[1].set_title('Target Sequences (Integrals)', fontweight='bold')\n",
    "axes[1].legend()\n",
    "axes[1].grid(True, alpha=0.3)\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "model_section",
   "metadata": {},
   "source": [
    "## Building the RNN\n",
    "\n",
    "### Custom RNN Cell\n",
    "\n",
    "We'll create a vanilla RNN cell with optional trainable initial state:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "rnn_cell",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-10-11T09:58:08.146616Z",
     "start_time": "2025-10-11T09:58:08.139758Z"
    }
   },
   "outputs": [],
   "source": [
    "class RNNCell(brainstate.nn.Module):\n",
    "    \"\"\"Vanilla RNN cell with trainable weights.\n",
    "    \n",
    "    h_t = activation(W_combined @ [x_t; h_{t-1}] + b)\n",
    "    \"\"\"\n",
    "    \n",
    "    def __init__(\n",
    "        self,\n",
    "        num_in: int,\n",
    "        num_out: int,\n",
    "        state_initializer: Callable = braintools.init.ZeroInit(),\n",
    "        w_initializer: Callable = braintools.init.XavierNormal(),\n",
    "        b_initializer: Callable = braintools.init.ZeroInit(),\n",
    "        activation: Callable = brainstate.nn.relu,\n",
    "        train_state: bool = False,  # Whether to train initial state\n",
    "    ):\n",
    "        super().__init__()\n",
    "        \n",
    "        self.num_out = num_out\n",
    "        self.train_state = train_state\n",
    "        \n",
    "        # Activation function\n",
    "        self.activation = activation\n",
    "        \n",
    "        # Combined weight matrix [input; hidden] -> hidden\n",
    "        W = braintools.init.param(\n",
    "            w_initializer, \n",
    "            (num_in + num_out, num_out)\n",
    "        )\n",
    "        b = braintools.init.param(b_initializer, (num_out,))\n",
    "        \n",
    "        self.W = brainstate.ParamState(W)\n",
    "        self.b = brainstate.ParamState(b) if b is not None else None\n",
    "        \n",
    "        # Trainable initial state (optional)\n",
    "        if train_state:\n",
    "            self.state2train = brainstate.ParamState(\n",
    "                braintools.init.ZeroInit()(num_out)\n",
    "            )\n",
    "        \n",
    "        self._state_initializer = state_initializer\n",
    "    \n",
    "    def init_state(self, batch_size=None, **kwargs):\n",
    "        \"\"\"Initialize hidden state.\"\"\"\n",
    "        self.state = brainstate.HiddenState(\n",
    "            braintools.init.param(\n",
    "                self._state_initializer, \n",
    "                (self.num_out,), \n",
    "                batch_size\n",
    "            )\n",
    "        )\n",
    "        \n",
    "        # Use trainable initial state if specified\n",
    "        if self.train_state:\n",
    "            self.state.value = jnp.repeat(\n",
    "                jnp.expand_dims(self.state2train.value, axis=0), \n",
    "                batch_size, \n",
    "                axis=0\n",
    "            )\n",
    "    \n",
    "    def update(self, x):\n",
    "        \"\"\"Update RNN cell for one time step.\n",
    "        \n",
    "        Args:\n",
    "            x: Input [batch, input_dim]\n",
    "            \n",
    "        Returns:\n",
    "            h: New hidden state [batch, hidden_dim]\n",
    "        \"\"\"\n",
    "        # Concatenate input and previous hidden state\n",
    "        x_combined = jnp.concatenate([x, self.state.value], axis=-1)\n",
    "        \n",
    "        # Linear transformation\n",
    "        h = x_combined @ self.W.value\n",
    "        if self.b is not None:\n",
    "            h += self.b.value\n",
    "        \n",
    "        # Apply activation\n",
    "        h = self.activation(h)\n",
    "        \n",
    "        # Update state\n",
    "        self.state.value = h\n",
    "        \n",
    "        return h"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "network_section",
   "metadata": {},
   "source": [
    "### Complete RNN Network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "rnn_network",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-10-11T09:58:08.157911Z",
     "start_time": "2025-10-11T09:58:08.153554Z"
    }
   },
   "outputs": [],
   "source": [
    "class RNN(brainstate.nn.Module):\n",
    "    \"\"\"RNN with recurrent layer and linear output.\"\"\"\n",
    "    \n",
    "    def __init__(self, num_in, num_hidden):\n",
    "        super().__init__()\n",
    "        \n",
    "        # RNN layer with trainable initial state\n",
    "        self.rnn = RNNCell(num_in, num_hidden, train_state=True)\n",
    "        \n",
    "        # Output projection\n",
    "        self.out = brainstate.nn.Linear(num_hidden, 1)\n",
    "    \n",
    "    def update(self, x):\n",
    "        \"\"\"Process one time step.\n",
    "        \n",
    "        Args:\n",
    "            x: Input at current time step\n",
    "            \n",
    "        Returns:\n",
    "            output: Prediction at current time step\n",
    "        \"\"\"\n",
    "        # RNN forward pass using >> operator (pipe)\n",
    "        return x >> self.rnn >> self.out"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "create_model",
   "metadata": {},
   "source": [
    "### Create Model and Optimizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "instantiate",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-10-11T09:58:41.775272Z",
     "start_time": "2025-10-11T09:58:41.742833Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+------------------------+------------+\n",
      "|        Modules         | Parameters |\n",
      "+------------------------+------------+\n",
      "|   ('out', 'weight')    |    101     |\n",
      "|      ('rnn', 'W')      |   10.10K   |\n",
      "|      ('rnn', 'b')      |    100     |\n",
      "| ('rnn', 'state2train') |    100     |\n",
      "|         Total          |   10.40K   |\n",
      "+------------------------+------------+\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "10401"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create RNN model\n",
    "model = RNN(num_in=1, num_hidden=num_hidden)\n",
    "\n",
    "# Get trainable parameters\n",
    "weights = model.states(brainstate.ParamState)\n",
    "\n",
    "# Create optimizer with learning rate decay\n",
    "lr_schedule = braintools.optim.ExponentialDecayLR(\n",
    "    learning_rate,\n",
    "    decay_steps=1, \n",
    "    decay_rate=lr_decay_rate\n",
    ")\n",
    "optimizer = braintools.optim.Adam(lr=lr_schedule, eps=1e-1)\n",
    "optimizer.register_trainable_weights(weights)\n",
    "\n",
    "brainstate.nn.count_parameters(model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "training_section",
   "metadata": {},
   "source": [
    "## Training the RNN\n",
    "\n",
    "### Define Prediction and Loss Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "functions",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-10-11T09:58:46.850437Z",
     "start_time": "2025-10-11T09:58:46.846157Z"
    }
   },
   "outputs": [],
   "source": [
    "@brainstate.transform.jit\n",
    "def f_predict(inputs):\n",
    "    \"\"\"Make predictions for a sequence.\n",
    "    \n",
    "    Args:\n",
    "        inputs: [num_steps, batch_size, 1]\n",
    "        \n",
    "    Returns:\n",
    "        predictions: [num_steps, batch_size, 1]\n",
    "    \"\"\"\n",
    "    # Initialize RNN state\n",
    "    brainstate.nn.init_all_states(model, batch_size=inputs.shape[1])\n",
    "    \n",
    "    # Process sequence\n",
    "    return brainstate.transform.for_loop(model.update, inputs)\n",
    "\n",
    "\n",
    "def f_loss(inputs, targets, l2_reg=2e-4):\n",
    "    \"\"\"Compute loss with L2 regularization.\n",
    "    \n",
    "    Args:\n",
    "        inputs: Input sequences\n",
    "        targets: Target sequences\n",
    "        l2_reg: L2 regularization coefficient\n",
    "        \n",
    "    Returns:\n",
    "        loss: Total loss value\n",
    "    \"\"\"\n",
    "    # Get predictions\n",
    "    predictions = f_predict(inputs)\n",
    "    \n",
    "    # Mean squared error\n",
    "    mse = braintools.metric.squared_error(predictions, targets).mean()\n",
    "    \n",
    "    # L2 regularization on weights\n",
    "    l2 = 0.0\n",
    "    for weight in weights.values():\n",
    "        for leaf in jax.tree.leaves(weight.value):\n",
    "            l2 += jnp.sum(leaf ** 2)\n",
    "    \n",
    "    return mse + l2_reg * l2"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "train_step_section",
   "metadata": {},
   "source": [
    "### Define Training Step"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "train_step",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-10-11T09:58:50.380060Z",
     "start_time": "2025-10-11T09:58:50.376128Z"
    }
   },
   "outputs": [],
   "source": [
    "@brainstate.transform.jit\n",
    "def f_train(inputs, targets):\n",
    "    \"\"\"Perform one training step.\n",
    "    \n",
    "    Args:\n",
    "        inputs: Input sequences\n",
    "        targets: Target sequences\n",
    "        \n",
    "    Returns:\n",
    "        loss: Loss value\n",
    "    \"\"\"\n",
    "    # Compute gradients\n",
    "    grads, loss = brainstate.transform.grad(\n",
    "        f_loss, \n",
    "        weights, \n",
    "        return_value=True\n",
    "    )(inputs, targets)\n",
    "    \n",
    "    # Update parameters\n",
    "    optimizer.update(grads)\n",
    "    \n",
    "    return loss"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "run_training",
   "metadata": {},
   "source": [
    "### Run Training Loop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "training_loop",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-10-11T10:00:05.708106Z",
     "start_time": "2025-10-11T09:58:51.844330Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Starting training...\n",
      "\n",
      "Epoch 0, Batch 100, Loss 0.19316\n",
      "Epoch 0, Batch 200, Loss 0.02372\n",
      "Epoch 0, Batch 300, Loss 0.02119\n",
      "Epoch 0, Batch 400, Loss 0.02136\n",
      "Epoch 0, Batch 500, Loss 0.04400\n",
      "\n",
      "Epoch 0 completed: Avg Loss = 0.06069\n",
      "\n",
      "Epoch 1, Batch 100, Loss 0.02979\n",
      "Epoch 1, Batch 200, Loss 0.01970\n",
      "Epoch 1, Batch 300, Loss 0.01925\n",
      "Epoch 1, Batch 400, Loss 0.01887\n",
      "Epoch 1, Batch 500, Loss 0.01854\n",
      "\n",
      "Epoch 1 completed: Avg Loss = 0.02123\n",
      "\n",
      "Epoch 2, Batch 100, Loss 0.01823\n",
      "Epoch 2, Batch 200, Loss 0.01819\n",
      "Epoch 2, Batch 300, Loss 0.01765\n",
      "Epoch 2, Batch 400, Loss 0.01752\n",
      "Epoch 2, Batch 500, Loss 0.01741\n",
      "\n",
      "Epoch 2 completed: Avg Loss = 0.01780\n",
      "\n",
      "Epoch 3, Batch 100, Loss 0.01673\n",
      "Epoch 3, Batch 200, Loss 0.01644\n",
      "Epoch 3, Batch 300, Loss 0.01648\n",
      "Epoch 3, Batch 400, Loss 0.01619\n",
      "Epoch 3, Batch 500, Loss 0.01601\n",
      "\n",
      "Epoch 3 completed: Avg Loss = 0.01637\n",
      "\n",
      "Epoch 4, Batch 100, Loss 0.01526\n",
      "Epoch 4, Batch 200, Loss 0.09099\n",
      "Epoch 4, Batch 300, Loss 0.03909\n",
      "Epoch 4, Batch 400, Loss 0.01513\n",
      "Epoch 4, Batch 500, Loss 0.01464\n",
      "\n",
      "Epoch 4 completed: Avg Loss = 0.03502\n",
      "\n",
      "Training complete!\n"
     ]
    }
   ],
   "source": [
    "print(\"Starting training...\\n\")\n",
    "\n",
    "for i_epoch in range(num_epochs):\n",
    "    epoch_losses = []\n",
    "    \n",
    "    for i_batch, (inputs, targets) in enumerate(train_data()):\n",
    "        if i_batch >= batches_per_epoch:\n",
    "            break\n",
    "        \n",
    "        loss = f_train(inputs, targets)\n",
    "        epoch_losses.append(float(loss))\n",
    "        \n",
    "        if (i_batch + 1) % 100 == 0:\n",
    "            avg_loss = np.mean(epoch_losses[-100:])\n",
    "            print(f'Epoch {i_epoch}, Batch {i_batch + 1:3d}, Loss {avg_loss:.5f}')\n",
    "    \n",
    "    avg_epoch_loss = np.mean(epoch_losses)\n",
    "    print(f'\\nEpoch {i_epoch} completed: Avg Loss = {avg_epoch_loss:.5f}\\n')\n",
    "\n",
    "print(\"Training complete!\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eval_section",
   "metadata": {},
   "source": [
    "## Evaluation and Visualization\n",
    "\n",
    "### Test on New Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "evaluation",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-10-11T10:00:06.068883Z",
     "start_time": "2025-10-11T10:00:05.724537Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test data generated:\n",
      "  Input shape: (25, 1, 1)\n",
      "  Target shape: (25, 1, 1)\n",
      "  Prediction shape: (25, 1, 1)\n"
     ]
    }
   ],
   "source": [
    "# Generate test data\n",
    "brainstate.nn.init_all_states(model, 1)\n",
    "x_test, y_test = build_inputs_and_targets(0.025, 0.01, 1)\n",
    "predictions = f_predict(x_test)\n",
    "\n",
    "print(f\"Test data generated:\")\n",
    "print(f\"  Input shape: {x_test.shape}\")\n",
    "print(f\"  Target shape: {y_test.shape}\")\n",
    "print(f\"  Prediction shape: {predictions.shape}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "plot_results",
   "metadata": {},
   "source": [
    "### Plot Predictions vs. Ground Truth"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "visualization",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-10-11T10:00:06.213455Z",
     "start_time": "2025-10-11T10:00:06.073378Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Test MSE: 0.000113\n"
     ]
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 5))\n",
    "\n",
    "# Convert to numpy for plotting\n",
    "y_true = np.asarray(y_test[:, 0]).flatten()\n",
    "y_pred = np.asarray(predictions[:, 0]).flatten()\n",
    "time_steps = np.arange(len(y_true))\n",
    "\n",
    "# Plot ground truth and predictions\n",
    "plt.plot(time_steps, y_true, 'b-', linewidth=2, label='Ground Truth', alpha=0.7)\n",
    "plt.plot(time_steps, y_pred, 'r--', linewidth=2, label='RNN Prediction', alpha=0.7)\n",
    "\n",
    "plt.xlabel('Time Step', fontsize=12)\n",
    "plt.ylabel('Cumulative Value', fontsize=12)\n",
    "plt.title('RNN Integration Task Performance', fontsize=14, fontweight='bold')\n",
    "plt.legend(fontsize=11)\n",
    "plt.grid(True, alpha=0.3)\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "\n",
    "# Compute final error\n",
    "mse = np.mean((y_true - y_pred) ** 2)\n",
    "print(f\"\\nTest MSE: {mse:.6f}\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Ecosystem-py",
   "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": "3.11.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}
