{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Neural networks and Reservoir Computing\n", "\n", "<img src=\"media/cover.png\" style=\"width: 40%; display: block; margin: auto;\">" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## Introduction\n", "\n", "In this lecture we will cover the following topics:\n", "\n", "- Windowed approaches and nonlinear models for time series forecasting.\n", "- Neural networks and the Multi-Layer Perceptron.\n", "- A brief overview of Recurrent Neural Networks.\n", "- The ESN, a randomized RNN from the family of Reservoir Computing.\n", "- An introduction to dimensionality reduction with Principal Component Analysis.\n", "- Examples of forecasting with MLP and ESN on real-world electricity data." ] }, { "cell_type": "code", "execution_count": 39, "metadata": { "tags": [ "hide-input" ] }, "outputs": [], "source": [ "import sys\n", "\n", "# Install dependencies if the notebook is running in Colab\n", "if 'google.colab' in sys.modules:\n", " !pip install -U -qq reservoir_computing" ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "slideshow": { "slide_type": "skip" }, "tags": [ "hide-input" ] }, "outputs": [], "source": [ "# Imports\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", "import time\n", "import plotly.graph_objects as go\n", "import statsmodels.api as sm\n", "from sklearn.datasets import make_circles\n", "from sklearn.neural_network import MLPRegressor\n", "from sklearn.preprocessing import StandardScaler\n", "from sklearn.metrics import mean_squared_error\n", "from sklearn.datasets import make_blobs\n", "from sklearn.decomposition import PCA\n", "from sklearn.linear_model import Ridge\n", "from sklearn.ensemble import HistGradientBoostingRegressor\n", "from reservoir_computing.reservoir import Reservoir\n", "from reservoir_computing.utils import make_forecasting_dataset\n", "from reservoir_computing.datasets import PredLoader" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "---" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## Windowed approaches for prediction" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- These methods consider a fixed window of size $P$.\n", "- Use the elements of the time series within the window $x(t-P), \\dots, x(t-1), x(t)$ to compute the prediction $x(t+\\tau)$, where $\\tau$ is the forecasting horizon.\n", "- Different algorithms combine the elements of the window in different ways to make predictions.\n", "\n", "<img src=\"media/windowed.png\" style=\"width: 40%; display: block; margin: auto;\">" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- As the window slides along the time series, new predictions are made.\n", "- An example of windowed approach is the linear AR model of order $P$: \n", "\n", "$$\\hat y(t+1) = \\beta_0 + \\beta_1 x(t-1) + \\beta_2 x(t-2) + \\dots + \\beta_P x(t-P)+\\epsilon_t$$\n", "\n", "<img src=\"media/sin.gif\" style=\"width: 40%; display: block; margin: auto;\">" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Limitation of linear models\n", "\n", "- Linear models are widely used for their simplicity and interpretiveness. \n", "- However, they assume a constant relationship between input and target variables.\n", "- As such, they cannot capture these nonlinearities and interactions effectively. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "- Consider the following examples:\n", " 1. Compute the *position* $y$ of a robotic arm given the positions $x_1, x_2, \\dots$ of the joints.\n", " 2. Compute a trajectory $\\dot{y}$ given the angular velocities $\\omega_1, \\omega_2, \\dots$ of the joints. \n", "\n", "<img src=\"media/robot.png\" style=\"width: 40%; display: block; margin: auto;\">\n", "\n", "- A linear model will fail in the second task." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "#### Example: forecasting electricity demand\n", "\n", "- This is a classic example where linear models fail.\n", "- Electricity demand is often governed by complex and nonlinear interactions with other factors." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "**Temperature and Demand**\n", "- The relationship between temperature and electricity demand is typically nonlinear and can exhibit a U-shaped curve. \n", "- Demand is low at moderate temperatures but increases sharply at high or low temperatures due to heating and cooling needs." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "**Time and Demand**\n", "- Demand patterns vary significantly throughout:\n", " - the day (peak hours in the morning and evening),\n", " - week (workdays vs. weekends),\n", " - year (summer vs. winter)." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Advantages of Nonlinear Models\n", "\n", "- Nonlinear models, such as neural networks, SVM, tree-based methods, etc..., can:\n", " - model the nonlinear relationships between demand and factors like temperature, capturing the U-shaped curve accurately,\n", " - take into account the interactions between different variables, such as the combined effect of time, holidays, and temperature on demand,\n", " - adapt to various patterns and changes in trends over time, making them more robust in dynamic environments.\n", "\n", "---" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## Neural networks" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Neural Nets are powerful nonlinear models that are *universal approximators*, i.e., they can learn to approximate any function.\n", "- Basic idea:\n", " 1. Map the data into a high dimensional space (bless of dimensionality).\n", " 2. Apply a nonlinear transformation.\n", " 3. Data become linearly separable." ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "slideshow": { "slide_type": "skip" }, "tags": [ "hide-input" ] }, "outputs": [], "source": [ "def plot_3D(elev=20, azim=30):\n", " X, y = make_circles(n_samples=200, factor=.01, noise=.15)\n", " fig = plt.figure(figsize=(10, 5))\n", " r = np.exp(-(X ** 2).sum(1))\n", " ax = fig.add_subplot(1, 2, 2, projection='3d')\n", " ax.scatter3D(X[:, 0], X[:, 1], r, c=y, s=50, cmap='bwr', alpha=0.5)\n", " ax.view_init(elev=elev, azim=azim)\n", " xx, yy = np.meshgrid(np.linspace(-1.5, 1.5, 50), np.linspace(-1.5, 1.5, 50))\n", " zz = np.full(xx.shape, 0.6)\n", " ax.plot_surface(xx, yy, zz, color='grey', alpha=0.7)\n", " ax.set_xlabel(\"$z_1$\")\n", " ax.set_ylabel(\"$z_2$\")\n", " ax.set_zlabel(\"$z_3$\")\n", " ax.set_xticks(())\n", " ax.set_yticks(())\n", " ax.set_zticks(())\n", " ax2 = fig.add_subplot(1, 2, 1)\n", " ax2.scatter(X[:, 0], X[:, 1], c=y, s=50, cmap='bwr', alpha=0.5)\n", " ax2.set_xlabel(\"$x_1$\")\n", " ax2.set_ylabel(\"$x_2$\")\n", " ax2.set_xticks(())\n", " ax2.set_yticks(())\n", " plt.tight_layout()\n", " plt.show()" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- This figure illustrates how mapping the data from a lower dimension (2D) to a higher dimension (3D) and applying a nonlinear function makes data linearly separable by an hyperplane." ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "<Figure size 1000x500 with 2 Axes>" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plot_3D()" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Multi-Layer Perceptron\n", "\n", "- A Multi-Layer Perceptron (MLP) is a simple Neural Network that consists of at least three layers of nodes: \n", " - An input layer.\n", " - One or more hidden layers.\n", " - An output layer. \n", "- The nodes in the a hidden layers apply a nonlinear activation function $\\sigma$. \n", "- MLP are generally trained with a supervised learing technique called backpropagation." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- The MLP's layers are fully connected, meaning each node in one layer connects with a certain weight to every node in the following layer. \n", "- The first layer receives the input.\n", "- The output of each layer is the input for the next layer until the final layer produces the output of the MLP.\n", "- The weights are stored in matrices $W$ and are the trainable parameters of the model.\n", "- The MLP can learn complex mappings from inputs to outputs to perform a wide range of data modeling and prediction tasks." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "#### Example: a simple MLP\n", "\n", "- Let's consider a concrete example of an MLP with one input layer, two hidden layers (with $h_1$ and $h_2$ units respectively), and one output layer. \n", "- The input vectors are of size $d_\\text{in}$, and the output vectors are of size $d_\\text{out}$.\n", "\n", "<img src=\"media/mlp.png\" style=\"width: 40%; display: block; margin: auto;\">" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "**Input layer**\n", "- The size of the input layer corresponds to the size of the input vectors, $d_\\text{in}$. \n", "- This layer simply passes the input to the first hidden layer without applying any transformation." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "**First hidden layer**\n", "- This layer has $h_1$ nodes. \n", "- Each node connects to all the $d_{\\text{in}}$ inputs.\n", "- It applies a weight matrix $W_\\text{in} \\in \\mathbb{R}^{d_\\text{in} \\times h_1}$ that transforms the input from vectors in $\\mathbb{R}^{d_\\text{in}}$ to vectors in $\\mathbb{R}^{h_1}$.\n", "- Then, it applies a nonlinear activation function $\\sigma$, and outputs $h_1$ intermediate results." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "**Second hidden layer**\n", "\n", "- This layer has $h_2$ nodes. \n", "- It takes the $h_1$ outputs from the first hidden layer. \n", "- It applies another set of weights and a nonlinear activation function to produce $h_2$ intermediate results." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "**Output Layer**\n", "- The output layer has $d_\\text{out}$ units. \n", "- It takes the $h_2$ outputs from the second hidden layer, applies weights and possibly a different activation function that depends on the task (usually, softmax for classification or no activation for regression).\n", "- The final output vector is of size $d_\\text{out}$." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### MLP for time series forecasting\n", "\n", "- The MLP can be used as a windowed technique to perform time series forecasting.\n", "- The sequence of past values $x(t-P), \\dots, x(t-1), x(t)$ will represent our input vector ($d_\\text{in} = P$).\n", "- The future value $x(t+\\tau)$ will be the output that the model will learn to predict ($d_\\text{out} = 1$).\n", "- We can also train the MLP to predict a sequence of $d_\\text{out} = H$ future predictions, e.g., $x(t+\\tau), x(t+\\tau+1), \\dots, x(t+\\tau+H)$.\n", "\n", "<img src=\"media/mlp_windowed.png\" style=\"width: 40%; display: block; margin: auto;\">" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Key for the training of the MLP is the creation of input-output pairs $\\{\\mathbf{x}_i, \\mathbf{y}_i\\}$ used for training.\n", "- Given the input $\\mathbf{x}_i$ the model must learn to predict the output $\\mathbf{y}_i$.\n", "- This is done by adapting the model weights to minimize the discrepancy between the prediction $\\mathbf{\\hat y}_i$ and the desired output $\\mathbf{y}_i$.\n", "- The weights are modified according to the gradient taken with respect to a loss function $\\mathcal{L}(\\mathbf{\\hat y}_i, \\mathbf{y}_i)$ such as the MSE, e.g., $W \\leftarrow W + \\delta \\frac{\\partial \\mathcal{L}}{\\partial W}$.\n", "- $\\delta$ is a small constant defining the gradient step (learning rate)." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Given the chunk of time series used for training, the training samples are generated according to the scheme in the picture.\n", "- Each input-output pair consists of:\n", " - a window of past time series values $x(t-P), \\dots, x(t-1), x(t)$.\n", " - a window of future values $x(t+\\tau), x(t+\\tau+1) \\dots, x(t+\\tau+H)$.\n", " \n", "<img src=\"media/data_split.png\" style=\"width: 100%; display: block; margin: auto;\">" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Python example\n", "\n", "- In this example we will consider a time series of electricity consumption registered on a backbone of the energy distribution network of Rome.\n", "- The original time series has 10min resolution.\n", "- For this example, we will resample it to 1h resolution as it will become more smooth (and easier to predict).\n", "- To train our models faster, we will consider only the first `3000` samples." ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Loaded ElecRome dataset.\n", "Data shape:\n", " X: (137376, 1)\n" ] } ], "source": [ "ts_full = PredLoader().get_data('ElecRome')\n", "\n", "# Resample the time series to hourly frequency\n", "ts_hourly = np.mean(ts_full.reshape(-1, 6), axis=1)\n", "\n", "# Use only the first 3000 time steps\n", "time_series = ts_hourly[0:3000]\n", "time_steps = np.arange(0, len(time_series))" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Next we split the data in train and test set.\n", "- We simply use the first 90% as training and the rest as test." ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "slideshow": { "slide_type": "-" } }, "outputs": [], "source": [ "# Split the time series into training and test sets\n", "train_size = int(0.9*len(time_series))\n", "tr = time_series[:train_size]\n", "te = time_series[train_size:]" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Next, we define the function to create input-output pairs.\n", "- For example, we create as input windows of size $P=12$ by specifying `window_size=12`.\n", "- Similarly, we create as ouput windows of size `forecast_horizon=12`.\n", "- If we are interested in predicting only one sample, e.g., $\\tau=12$ and $H=1$, we simply take the last element of the output window." ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [], "source": [ "def create_windows(data, window_size, forecast_horizon):\n", " X, y = [], []\n", " for i in range(len(data) - window_size - forecast_horizon + 1):\n", " X.append(data[i:(i + window_size)])\n", " y.append(data[i + window_size:i + window_size + forecast_horizon])\n", " return np.array(X), np.array(y)\n", "\n", "# Define window size and forecast horizon\n", "window_size = 12\n", "forecast_horizon = 12\n", "\n", "# Create input-output pairs\n", "X_train, y_train = create_windows(tr, window_size, forecast_horizon)\n", "X_test, y_test = create_windows(te, window_size, forecast_horizon)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- We can see how the input-output pairs look like in the following plot.\n", "- The blue dots will represent the input $x_{t-P}, \\dots x_{t-1}, x_t$ with $P=12$.\n", "- The orange dot is the output $x_{t+\\tau}$ with $\\tau=12$." ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "<Figure size 1000x800 with 5 Axes>" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Plot some pairs of input-output\n", "fig, axes = plt.subplots(5,1,figsize=(10,8))\n", "for i in range(5):\n", " axes[i].scatter(range(i, window_size+i), X_train[i], color='tab:blue', edgecolor='k') # Input\n", " axes[i].scatter(range(window_size+i, i+window_size+forecast_horizon-1), y_train[i,:-1], color='lightgray', edgecolor='k') \n", " axes[i].scatter(i+window_size+forecast_horizon-1, y_train[i,-1], color='tab:orange', edgecolor='k')\n", " axes[i].set_xlim(-1,28)\n", "plt.xlabel('Time step')\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Let say we instead want to predict the energy consumption 1 day ahead by looking at the consumption of the previous 12 hours, which is a more realisitc scenario.\n", "- We define `forecast_horizon = 24` to do that." ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "slideshow": { "slide_type": "-" } }, "outputs": [], "source": [ "# Define window size and forecast horizon\n", "window_size = 12\n", "forecast_horizon = 24\n", "\n", "# Create input-output pairs\n", "X_train, y_train = create_windows(tr, window_size, forecast_horizon)\n", "X_test, y_test = create_windows(te, window_size, forecast_horizon)\n", "\n", "# We are only interested in the last time step of the horizon\n", "y_train = y_train[:, -1] \n", "y_test = y_test[:, -1] " ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Most neural nets, including the MLP, want the data to be normalized in a small range.\n", "- We can do that by applying the `StandardScaler()` from the sklearn library.\n", "- `scaler.fit_transform(X_train)` will subtract from `X_train` its mean and divide by its variance.\n", "- `scaler.transform(X_test)` will subtract from `X_test` the mean of `X_train` and divide by the variance of `X_train`. \n", "- Why not normalizing with the test set statistics?\n", " - Because we do not have access to the test data beforehand, so we cannot compute statistics such as its mean and variance." ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "slideshow": { "slide_type": "-" } }, "outputs": [], "source": [ "# Normalize the data\n", "scaler = StandardScaler()\n", "X_train_scaled = scaler.fit_transform(X_train)\n", "X_test_scaled = scaler.transform(X_test)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Next, we define the MLP.\n", "- We specify the size of the hidden layers as $h_1=16$ and $h_2=8$. Using larger values increases the model capacity, but can lead to overfit.\n", "- As the activation function we use a [ReLU](https://en.wikipedia.org/wiki/Rectifier_(neural_networks)).\n", "- As the algorithms to compute the gradients and update the values of the parameters $W$ we use [Adam](https://en.wikipedia.org/wiki/Stochastic_gradient_descent#Adam).\n", "- Finally, we train the model for `1000` iterations." ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "slideshow": { "slide_type": "-" } }, "outputs": [ { "data": { "text/html": [ "<style>#sk-container-id-1 {\n", " /* Definition of color scheme common for light and dark mode */\n", " --sklearn-color-text: #000;\n", " --sklearn-color-text-muted: #666;\n", " --sklearn-color-line: gray;\n", " /* Definition of color scheme for unfitted estimators */\n", " --sklearn-color-unfitted-level-0: #fff5e6;\n", " --sklearn-color-unfitted-level-1: #f6e4d2;\n", " --sklearn-color-unfitted-level-2: #ffe0b3;\n", " --sklearn-color-unfitted-level-3: chocolate;\n", " /* Definition of color scheme for fitted estimators */\n", " --sklearn-color-fitted-level-0: #f0f8ff;\n", " --sklearn-color-fitted-level-1: #d4ebff;\n", " --sklearn-color-fitted-level-2: #b3dbfd;\n", " --sklearn-color-fitted-level-3: cornflowerblue;\n", "\n", " /* Specific color for light theme */\n", " --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n", " --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));\n", " --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n", " --sklearn-color-icon: #696969;\n", "\n", " @media (prefers-color-scheme: dark) {\n", " /* Redefinition of color scheme for dark theme */\n", " --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n", " --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));\n", " --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n", " --sklearn-color-icon: #878787;\n", " }\n", "}\n", "\n", "#sk-container-id-1 {\n", " color: var(--sklearn-color-text);\n", "}\n", "\n", "#sk-container-id-1 pre {\n", " padding: 0;\n", "}\n", "\n", "#sk-container-id-1 input.sk-hidden--visually {\n", " border: 0;\n", " clip: rect(1px 1px 1px 1px);\n", " clip: rect(1px, 1px, 1px, 1px);\n", " height: 1px;\n", " margin: -1px;\n", " overflow: hidden;\n", " padding: 0;\n", " position: absolute;\n", " width: 1px;\n", "}\n", "\n", "#sk-container-id-1 div.sk-dashed-wrapped {\n", " border: 1px dashed var(--sklearn-color-line);\n", " margin: 0 0.4em 0.5em 0.4em;\n", " box-sizing: border-box;\n", " padding-bottom: 0.4em;\n", " background-color: var(--sklearn-color-background);\n", "}\n", "\n", "#sk-container-id-1 div.sk-container {\n", " /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n", " but bootstrap.min.css set `[hidden] { display: none !important; }`\n", " so we also need the `!important` here to be able to override the\n", " default hidden behavior on the sphinx rendered scikit-learn.org.\n", " See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n", " display: inline-block !important;\n", " position: relative;\n", "}\n", "\n", "#sk-container-id-1 div.sk-text-repr-fallback {\n", " display: none;\n", "}\n", "\n", "div.sk-parallel-item,\n", "div.sk-serial,\n", "div.sk-item {\n", " /* draw centered vertical line to link estimators */\n", " background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n", " background-size: 2px 100%;\n", " background-repeat: no-repeat;\n", " background-position: center center;\n", "}\n", "\n", "/* Parallel-specific style estimator block */\n", "\n", "#sk-container-id-1 div.sk-parallel-item::after {\n", " content: \"\";\n", " width: 100%;\n", " border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n", " flex-grow: 1;\n", "}\n", "\n", "#sk-container-id-1 div.sk-parallel {\n", " display: flex;\n", " align-items: stretch;\n", " justify-content: center;\n", " background-color: var(--sklearn-color-background);\n", " position: relative;\n", "}\n", "\n", "#sk-container-id-1 div.sk-parallel-item {\n", " display: flex;\n", " flex-direction: column;\n", "}\n", "\n", "#sk-container-id-1 div.sk-parallel-item:first-child::after {\n", " align-self: flex-end;\n", " width: 50%;\n", "}\n", "\n", "#sk-container-id-1 div.sk-parallel-item:last-child::after {\n", " align-self: flex-start;\n", " width: 50%;\n", "}\n", "\n", "#sk-container-id-1 div.sk-parallel-item:only-child::after {\n", " width: 0;\n", "}\n", "\n", "/* Serial-specific style estimator block */\n", "\n", "#sk-container-id-1 div.sk-serial {\n", " display: flex;\n", " flex-direction: column;\n", " align-items: center;\n", " background-color: var(--sklearn-color-background);\n", " padding-right: 1em;\n", " padding-left: 1em;\n", "}\n", "\n", "\n", "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n", "clickable and can be expanded/collapsed.\n", "- Pipeline and ColumnTransformer use this feature and define the default style\n", "- Estimators will overwrite some part of the style using the `sk-estimator` class\n", "*/\n", "\n", "/* Pipeline and ColumnTransformer style (default) */\n", "\n", "#sk-container-id-1 div.sk-toggleable {\n", " /* Default theme specific background. It is overwritten whether we have a\n", " specific estimator or a Pipeline/ColumnTransformer */\n", " background-color: var(--sklearn-color-background);\n", "}\n", "\n", "/* Toggleable label */\n", "#sk-container-id-1 label.sk-toggleable__label {\n", " cursor: pointer;\n", " display: flex;\n", " width: 100%;\n", " margin-bottom: 0;\n", " padding: 0.5em;\n", " box-sizing: border-box;\n", " text-align: center;\n", " align-items: start;\n", " justify-content: space-between;\n", " gap: 0.5em;\n", "}\n", "\n", "#sk-container-id-1 label.sk-toggleable__label .caption {\n", " font-size: 0.6rem;\n", " font-weight: lighter;\n", " color: var(--sklearn-color-text-muted);\n", "}\n", "\n", "#sk-container-id-1 label.sk-toggleable__label-arrow:before {\n", " /* Arrow on the left of the label */\n", " content: \"▸\";\n", " float: left;\n", " margin-right: 0.25em;\n", " color: var(--sklearn-color-icon);\n", "}\n", "\n", "#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {\n", " color: var(--sklearn-color-text);\n", "}\n", "\n", "/* Toggleable content - dropdown */\n", "\n", "#sk-container-id-1 div.sk-toggleable__content {\n", " max-height: 0;\n", " max-width: 0;\n", " overflow: hidden;\n", " text-align: left;\n", " /* unfitted */\n", " background-color: var(--sklearn-color-unfitted-level-0);\n", "}\n", "\n", "#sk-container-id-1 div.sk-toggleable__content.fitted {\n", " /* fitted */\n", " background-color: var(--sklearn-color-fitted-level-0);\n", "}\n", "\n", "#sk-container-id-1 div.sk-toggleable__content pre {\n", " margin: 0.2em;\n", " border-radius: 0.25em;\n", " color: var(--sklearn-color-text);\n", " /* unfitted */\n", " background-color: var(--sklearn-color-unfitted-level-0);\n", "}\n", "\n", "#sk-container-id-1 div.sk-toggleable__content.fitted pre {\n", " /* unfitted */\n", " background-color: var(--sklearn-color-fitted-level-0);\n", "}\n", "\n", "#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n", " /* Expand drop-down */\n", " max-height: 200px;\n", " max-width: 100%;\n", " overflow: auto;\n", "}\n", "\n", "#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n", " content: \"▾\";\n", "}\n", "\n", "/* Pipeline/ColumnTransformer-specific style */\n", "\n", "#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n", " color: var(--sklearn-color-text);\n", " background-color: var(--sklearn-color-unfitted-level-2);\n", "}\n", "\n", "#sk-container-id-1 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n", " background-color: var(--sklearn-color-fitted-level-2);\n", "}\n", "\n", "/* Estimator-specific style */\n", "\n", "/* Colorize estimator box */\n", "#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n", " /* unfitted */\n", " background-color: var(--sklearn-color-unfitted-level-2);\n", "}\n", "\n", "#sk-container-id-1 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n", " /* fitted */\n", " background-color: var(--sklearn-color-fitted-level-2);\n", "}\n", "\n", "#sk-container-id-1 div.sk-label label.sk-toggleable__label,\n", "#sk-container-id-1 div.sk-label label {\n", " /* The background is the default theme color */\n", " color: var(--sklearn-color-text-on-default-background);\n", "}\n", "\n", "/* On hover, darken the color of the background */\n", "#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {\n", " color: var(--sklearn-color-text);\n", " background-color: var(--sklearn-color-unfitted-level-2);\n", "}\n", "\n", "/* Label box, darken color on hover, fitted */\n", "#sk-container-id-1 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n", " color: var(--sklearn-color-text);\n", " background-color: var(--sklearn-color-fitted-level-2);\n", "}\n", "\n", "/* Estimator label */\n", "\n", "#sk-container-id-1 div.sk-label label {\n", " font-family: monospace;\n", " font-weight: bold;\n", " display: inline-block;\n", " line-height: 1.2em;\n", "}\n", "\n", "#sk-container-id-1 div.sk-label-container {\n", " text-align: center;\n", "}\n", "\n", "/* Estimator-specific */\n", "#sk-container-id-1 div.sk-estimator {\n", " font-family: monospace;\n", " border: 1px dotted var(--sklearn-color-border-box);\n", " border-radius: 0.25em;\n", " box-sizing: border-box;\n", " margin-bottom: 0.5em;\n", " /* unfitted */\n", " background-color: var(--sklearn-color-unfitted-level-0);\n", "}\n", "\n", "#sk-container-id-1 div.sk-estimator.fitted {\n", " /* fitted */\n", " background-color: var(--sklearn-color-fitted-level-0);\n", "}\n", "\n", "/* on hover */\n", "#sk-container-id-1 div.sk-estimator:hover {\n", " /* unfitted */\n", " background-color: var(--sklearn-color-unfitted-level-2);\n", "}\n", "\n", "#sk-container-id-1 div.sk-estimator.fitted:hover {\n", " /* fitted */\n", " background-color: var(--sklearn-color-fitted-level-2);\n", "}\n", "\n", "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n", "\n", "/* Common style for \"i\" and \"?\" */\n", "\n", ".sk-estimator-doc-link,\n", "a:link.sk-estimator-doc-link,\n", "a:visited.sk-estimator-doc-link {\n", " float: right;\n", " font-size: smaller;\n", " line-height: 1em;\n", " font-family: monospace;\n", " background-color: var(--sklearn-color-background);\n", " border-radius: 1em;\n", " height: 1em;\n", " width: 1em;\n", " text-decoration: none !important;\n", " margin-left: 0.5em;\n", " text-align: center;\n", " /* unfitted */\n", " border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n", " color: var(--sklearn-color-unfitted-level-1);\n", "}\n", "\n", ".sk-estimator-doc-link.fitted,\n", "a:link.sk-estimator-doc-link.fitted,\n", "a:visited.sk-estimator-doc-link.fitted {\n", " /* fitted */\n", " border: var(--sklearn-color-fitted-level-1) 1pt solid;\n", " color: var(--sklearn-color-fitted-level-1);\n", "}\n", "\n", "/* On hover */\n", "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n", ".sk-estimator-doc-link:hover,\n", "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n", ".sk-estimator-doc-link:hover {\n", " /* unfitted */\n", " background-color: var(--sklearn-color-unfitted-level-3);\n", " color: var(--sklearn-color-background);\n", " text-decoration: none;\n", "}\n", "\n", "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n", ".sk-estimator-doc-link.fitted:hover,\n", "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n", ".sk-estimator-doc-link.fitted:hover {\n", " /* fitted */\n", " background-color: var(--sklearn-color-fitted-level-3);\n", " color: var(--sklearn-color-background);\n", " text-decoration: none;\n", "}\n", "\n", "/* Span, style for the box shown on hovering the info icon */\n", ".sk-estimator-doc-link span {\n", " display: none;\n", " z-index: 9999;\n", " position: relative;\n", " font-weight: normal;\n", " right: .2ex;\n", " padding: .5ex;\n", " margin: .5ex;\n", " width: min-content;\n", " min-width: 20ex;\n", " max-width: 50ex;\n", " color: var(--sklearn-color-text);\n", " box-shadow: 2pt 2pt 4pt #999;\n", " /* unfitted */\n", " background: var(--sklearn-color-unfitted-level-0);\n", " border: .5pt solid var(--sklearn-color-unfitted-level-3);\n", "}\n", "\n", ".sk-estimator-doc-link.fitted span {\n", " /* fitted */\n", " background: var(--sklearn-color-fitted-level-0);\n", " border: var(--sklearn-color-fitted-level-3);\n", "}\n", "\n", ".sk-estimator-doc-link:hover span {\n", " display: block;\n", "}\n", "\n", "/* \"?\"-specific style due to the `<a>` HTML tag */\n", "\n", "#sk-container-id-1 a.estimator_doc_link {\n", " float: right;\n", " font-size: 1rem;\n", " line-height: 1em;\n", " font-family: monospace;\n", " background-color: var(--sklearn-color-background);\n", " border-radius: 1rem;\n", " height: 1rem;\n", " width: 1rem;\n", " text-decoration: none;\n", " /* unfitted */\n", " color: var(--sklearn-color-unfitted-level-1);\n", " border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n", "}\n", "\n", "#sk-container-id-1 a.estimator_doc_link.fitted {\n", " /* fitted */\n", " border: var(--sklearn-color-fitted-level-1) 1pt solid;\n", " color: var(--sklearn-color-fitted-level-1);\n", "}\n", "\n", "/* On hover */\n", "#sk-container-id-1 a.estimator_doc_link:hover {\n", " /* unfitted */\n", " background-color: var(--sklearn-color-unfitted-level-3);\n", " color: var(--sklearn-color-background);\n", " text-decoration: none;\n", "}\n", "\n", "#sk-container-id-1 a.estimator_doc_link.fitted:hover {\n", " /* fitted */\n", " background-color: var(--sklearn-color-fitted-level-3);\n", "}\n", "</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>MLPRegressor(hidden_layer_sizes=(16, 8), max_iter=1000)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow\"><div><div>MLPRegressor</div></div><div><a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.6/modules/generated/sklearn.neural_network.MLPRegressor.html\">?<span>Documentation for MLPRegressor</span></a><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></div></label><div class=\"sk-toggleable__content fitted\"><pre>MLPRegressor(hidden_layer_sizes=(16, 8), max_iter=1000)</pre></div> </div></div></div></div>" ], "text/plain": [ "MLPRegressor(hidden_layer_sizes=(16, 8), max_iter=1000)" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Define and train the neural network\n", "mlp = MLPRegressor(hidden_layer_sizes=(16,8), activation='relu', solver='adam', max_iter=1000)\n", "mlp.fit(X_train_scaled, y_train)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Once the model is trained, we can compute the prediction on the test set.\n", "- We can also compute predictions beyond the whole dataset that we have available.\n", "- In this case, we do not have a way to check how well the model is doing since we do not have the actual data available.\n", "- However, it reflects a realistic forecasting scenario where we, indeed, do not know the future." ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "The next time step prediction is 34.50\n" ] } ], "source": [ "# Predict on the test set\n", "y_pred = mlp.predict(X_test_scaled)\n", "\n", "# Forecast beyond the dataset using the model\n", "last_window = time_series[-window_size:]\n", "last_window_scaled = scaler.transform(last_window.reshape(1, -1))\n", "next_step_pred = mlp.predict(last_window_scaled)\n", "\n", "print(f\"The next time step prediction is {next_step_pred[0]:.2f}\")" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- To check the performance of the model we can compute the MSE on the predictions of the test set.\n", "- We can also visualize the predictions against the real data." ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Mean Squared Error: 20.55\n" ] }, { "data": { "image/png": "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", "text/plain": [ "<Figure size 1400x400 with 1 Axes>" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "mse = mean_squared_error(y_test, y_pred)\n", "print(f\"Mean Squared Error: {mse:.2f}\")\n", "\n", "plt.figure(figsize=(14, 4))\n", "plt.plot(time_steps[2500:], time_series[2500:], label=\"Actual\")\n", "plt.plot(time_steps[-len(y_test):], y_pred, label=\"Predicted\")\n", "plt.title(\"Time Series Forecast\")\n", "plt.xlabel(\"Time Step\")\n", "plt.ylabel(\"Value\")\n", "plt.grid()\n", "plt.legend()\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Is the prediction good or not?\n", "- To get an idea, we can compare the performance against a simple baseline.\n", "- In this case, we are making a prediction at a forecast horizon $\\tau=24$ equal to the main seasonality of the time series.\n", "- The most natural baseline is to use the values of the previous day as the prediction for the next day." ] }, { "cell_type": "code", "execution_count": 13, "metadata": { "slideshow": { "slide_type": "-" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Mean Squared Error: 27.14\n" ] }, { "data": { "image/png": "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", "text/plain": [ "<Figure size 700x300 with 1 Axes>" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Compute the mse between the original time series and the time series shifhted by 24 time steps\n", "mse = mean_squared_error(y_test[24:], y_test[:-24])\n", "print(f\"Mean Squared Error: {mse:.2f}\")\n", "\n", "plt.figure(figsize=(7, 3))\n", "plt.plot(y_test[24:])\n", "plt.plot(y_test[:-24])\n", "plt.grid()\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "-" } }, "source": [ "- Any reasonable forecasting method must beat this baseline." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Windowed approaches drawbacks\n", "\n", "- What is the optimal value of $P$?\n", "- If $P$ is fixed, how can we deal with different temporal dependencies?\n", "- Let's consider the example of predicting the next word in a sentence." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "<img src=\"media/word_pred.gif\" style=\"width: 50%; display: block; margin: auto;\">\n", "\n", "- In the first case, we need to go back 4 time steps to retrieve the information we need.\n", "- In the second case, 9 steps.\n", "- How to set $P$? What is the maximum memory we will ever need? \n", "- As discussed previously, setting $P$ too high will smooth our data too much.\n", "- Finding a good tradeoff is difficult...\n", "\n", "---" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## Recurrent Neural Networks" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- RNNs are a class of neural networks designed to recognize patterns in sequences of data, such as time series data, natural language, or sequences of images. \n", "- Unlike traditional neural networks, RNNs have a memory that captures information about what has been calculated so far, essentially allowing them to make predictions based on past inputs." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- As an RNN processes a sequence $\\mathbf{x}$ it maintains information in a *hidden state* $\\mathbf{h}$. \n", "- This process allows the RNN to use previous computations as a context for making decisions about new data.\n", " \n", "<img src=\"media/RNN.gif\" style=\"width: 70%; display: block; margin: auto;\">" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "-" } }, "source": [ "- $x(t)$ is the input time series, $\\mathbf{h}(t)$ the hidden state vector, $y(t)$ the output.\n", "- The matrices $W_x$, $W_h$, and $W_o$ represents the input, hidden (recurrent), and output weights respectively." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "**Back Propagation Through Time (BPTT)**\n", "\n", "- BPTT is the algorithm used for training RNNs. \n", "- It involves unfolding the RNN through time to obtain a standard feed-forward neural network (like an MLP). \n", "- After unfolding, one can apply the standard backpropagation algorithm. \n", "- BPTT computes gradients for each parameter across *all* time steps of the input sequence. \n", "\n", "<img src=\"media/bptt.png\" style=\"width: 50%; display: block; margin: auto;\">" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Challenges and shortcomings in RNNs\n", "\n", "- In BPTT the gradient has to go through all time steps and, due to the presence of nonlinearities, it can become too small and not reach distant time steps.\n", "- This creates the problem called *vanishing gradient*. \n", "- Opposed, yet similar, is the problem of *exploding gradient* occurring when the gradients grow across the sequence. \n", "- The vanishing gradient problem makes it hard for RNNs to learn long-range dependencies because updates to the weights become insignificantly small, causing the learning to stall. \n", "- Conversely, exploding gradients can cause weights to oscillate or diverge.\n", "- Techniques such as gradient clipping and gated units (e.g., LSTM, GRU) have been developed to mitigate these issues." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "- Another important limitation of the RNNs is that they fail to exploit hardware acceleration like other neural nets.\n", "- This is due to their recurrent nature that requires computations to be done sequentially rather than in parallel.\n", "\n", "---" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## Reservoir Computing" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- RC is a familily of randomized RNNs, popularized in machine learning by Echo State Networks (ESNs).\n", "- RC and ESNs are terms ofen used interchangeably.\n", "- There are two main differences that separate an ESN from an RNN:\n", "\n", "<img src=\"media/Reservoir.gif\" style=\"width: 70%; display: block; margin: auto;\">" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "-" } }, "source": [ "- The output weights $W_o$ is the only part of the ESN that is trained.\n", "- Optimizing $W_o$ can be done with a simple linear regression algorithm.\n", "- The workflow is the following." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "1. Generate a sequence of reservoir states $\\mathbf{H} = \\{\\mathbf{h}(0), \\mathbf{h}(1), \\dots, \\mathbf{h}(T)\\}$.\n", " - This is done by applying the state update equation \n", " \n", " $$\\mathbf{h}_t = \\sigma \\left(\\mathbf{W}_i x(t)+\\mathbf{W}_h \\mathbf{h}(t-1)\\right)$$ \n", " \n", " - for each time step $t=1,2, \\dots, T$ of the input sequence.\n", " \n", " - The nonlinearity $\\sigma$ is usually an hyperbolic tangent ($\\texttt{tanh}$)." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "2. Apply a linear regression algorithm to compute a linear mapping $g(\\cdot)$ between the reservoir states and the desired output sequence $\\mathbf{y} = \\{y(1), y(2), \\dots, y(T) \\}$: \n", "\n", "$$\\mathbf{y} = g(\\mathbf{H})$$\n", "\n", "- The function $g(\\cdot)$ is called *readout*.\n", "- In a forecasting setting, $y(t)$ correspond to a future value of the input, e.g., $y(t) = x(t+\\tau)$." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Readout \n", "\n", "- The readout $g(\\cdot)$ is usually implemented through a linear regression model, e.g., Ridge Regression.\n", "- In this case $g(\\cdot)$ corresponds to a weight matrix $\\mathbf{W}_o$.\n", "- However, any other regression model can be used to implement the readout, including an MLP.\n", "\n", "<img src=\"media/Readout.png\" style=\"width: 50%; display: block; margin: auto;\">" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Note that fitting an MLP to the readout states is different from the window approach we saw before.\n", "- Window approach:\n", " - The model predicts a future value from the fixed amount of temporal information contained in the window \n", " \n", " $$x(t+\\tau) = g([x(t-P), \\dots, x(t-1), x(t)])$$\n", "\n", "- Reservoir approach:\n", " - The model predicts a future value from a single Reservoir state \n", " \n", " $$x(t+\\tau) = g(\\mathbf{h}(t))$$" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Why the Reservoir approach works\n", "\n", "- The Reservoir extracts a rich pool dynamical features from the time series.\n", "- These are embedded into the high-dimensional Reservoir state $\\mathbf{h}(t)$.\n", "- Contrary to a fixed window, $\\mathbf{h}(t)$ maintains a memory of all the previous inputs, back to the origin of the series $x(0)$.\n", "- Some of the features are relevant for the task at hand, while others are not.\n", "- The task of the readout is to select those features relevant for the task." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Say, we want make a forecast $\\tau_1$ steps ahead.\n", "- The readout will select a certain combination of dynamical features from the Reservoir.\n", "\n", "<img src=\"media/Reservoir_pred1.png\" style=\"width: 50%; display: block; margin: auto;\">" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- To predict at a different horizon $\\tau_2$ the readout will select a different group of features.\n", "- Note that in both cases the Readout produces always the same pool of features!\n", "\n", "<img src=\"media/Reservoir_pred2.png\" style=\"width: 50%; display: block; margin: auto;\">" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- The Readout is untrained and its states are generated without *supervision*, i.e., without an external guidance.\n", "- Since it does not know what task it will have to solve, the Readout is configured to produce a pool of dynamic features that is most rich and varied as possible.\n", "- In other words, the Readout trades the lack of training with a redudancy of generated features." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "### Reservoir configuration\n", "\n", "- There are several hyperparameters that can be configured in the Reservoir.\n", "- We review the main ones in the following." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "#### Spectral radius\n", "\n", "- Cleary, the contribution of more recent inputs must have a stronger influence on the current state.\n", "- This means that the Reservoir should gradually forget its past states.\n", "- This property, called *echo state property*, gives the name to the Echo State Network.\n", "- It ensures to not model noise, to forget sporadic shocks, and the initial state $\\mathbf{h}(0)$ that is uninformative.\n", "- In control theory, this translate in having a dynamic that is *contractive*.\n", "- A dynamic is contractive if two initially different states eventually converge." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- On the other hand, we want the Reservoir to produce a rich pool of features.\n", "- If the dynamics of the Reservoir is too *conctractive* this does not happen.\n", "- We have to find a sweet spot by tuning a parameter $\\rho$ called *spectral radius*." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "-" } }, "source": [ "<img src=\"media/dynamics.png\" style=\"width: 50%; display: block; margin: auto;\">" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Recall the state update equation \n", "\n", "$$\\mathbf{h}(t) = \\sigma \\left(\\mathbf{W}_i x(t)+\\mathbf{W}_h \\mathbf{h}(t-1)\\right)$$\n", "\n", "- The spectral radius is the largest eigenvalue of the state transtion matrix $W_h$.\n", "- We can *set* the spectral radius by computing the largest eigenvalue $\\lambda_\\text{max}$ of $W_h$ and then letting $W_h = \\rho \\frac{W_h}{\\lambda_\\text{max}}$." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "- A rule of thumb is to set $\\rho$ just below 1.\n", "- However, to achieve good performance it is often necessary to fine-tune $\\rho$ to values that can be lower or even higher than 1." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Another way of determining a good value of $\\rho$ is to look at the *transient* phase of the Reservoir.\n", "- This is how much time it takes to forget the initialization.\n", "- Assume two different inizializations of the Reservoir state: $\\mathbf{h}_1(0)$ and $\\mathbf{h}_2(0)$.\n", "- If the Reservoir dynamics is contractive, the effect of the different initalizations will enventually fade.\n", "- If it is chaotic, it will persist." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Let's check this with a practical example.\n", "- We consider two intializations, $\\mathbf{h}_1(0) = [0,0,\\dots,0]$ and $\\mathbf{h}_2(0) = [1,1,\\dots, 1]$.\n", "- For this example, we set the input $x$ to be always zero to not let it affect the evolution of the Reservoir state." ] }, { "cell_type": "code", "execution_count": 14, "metadata": { "slideshow": { "slide_type": "-" } }, "outputs": [], "source": [ "# Initial states\n", "initial_state_0 = np.zeros((1, 100), dtype=float) \n", "initial_state_1 = np.ones((1, 100), dtype=float)\n", "\n", "x = np.zeros((1, 100, 1)) # Zero input, it does not contribute to the state\n", "rhos = [.3, 0.99, 1.3] # We will use three different spectral radii" ] }, { "cell_type": "code", "execution_count": 15, "metadata": { "slideshow": { "slide_type": "skip" }, "tags": [ "hide-input" ] }, "outputs": [], "source": [ "def plot_states_evolution(x, rhos, h0, h1):\n", " plt.figure(figsize=(5, 3))\n", " for rho in rhos:\n", " res = Reservoir(spectral_radius=rho) # initialize the reservoir with a given spectral radius\n", " states_0 = res.get_states(x, bidir=False, initial_state=h0) # states using initial state 0\n", " states_1 = res.get_states(x, bidir=False, initial_state=h1) # states using initial state 1\n", " plt.plot(np.linalg.norm(states_0 - states_1, axis=(0,2)), label=f'rho={rho}') # L2 norm of the difference\n", " plt.legend(loc='upper right', ncol=3)\n", " plt.xlabel('$t$')\n", " plt.ylabel('$\\|\\mathbf{h}_1(t) - \\mathbf{h}_2(t)\\|_2$')\n", " plt.grid()\n", " plt.tight_layout()\n", " plt.title('Transient phase of the Reservoir')\n", " plt.show()" ] }, { "cell_type": "code", "execution_count": 16, "metadata": { "slideshow": { "slide_type": "-" } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "<Figure size 500x300 with 1 Axes>" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plot_states_evolution(x, rhos, initial_state_0, initial_state_1)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "#### Input scaling\n", "\n", "- Another critical value is the input scaling $\\omega_\\text{in}$.\n", "- Is a value that multiplies the input weights $W_i$ changing their magnitude.\n", "- This is key to control the amount of nonlinearity in the model." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Reservoir units are usually equipped with a $\\texttt{tanh}$ activation.\n", "- A small value $\\omega_\\text{in}$ maps the Reservoir inputs towards the centre of the $\\texttt{tanh}$ where is more linear.\n", "- Therefore, a small $\\omega_\\text{in}$ reduces the amount of nonlinearity in the Reservoir.\n", "- On the other hand, a large $\\omega_\\text{in}$ translates into a more nonlinear beahvior as the $\\texttt{tanh}$ is closer to saturation.\n", "\n", "<img src=\"media/tanh.png\" style=\"width: 30%; display: block; margin: auto;\">" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "-" } }, "source": [ "- Like for the spectral radius, $\\omega_\\text{in}$ should be tuned carefully.\n", "- A good starting value is usually around 0.1." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Another hypeparameter is the number of Reservoir units $N_h$. \n", "- A larger value can give better performance at the cost of higher computation time.\n", "- A good starting point is usually $N_h=300$, to be increased until there is no more gain in performance." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "- There are also other hyperparameters in the ESN, such as the sparsity of the Readout and an optional noise to inject in the state update equation.\n", "- These are usually less critical than $\\rho$ and $\\omega_\\text{in}$ and can be left to their default value in most cases.\n", "- Tuning hyperparameters in randomized architectures such as the ESN is much more important than in trainable neural networks, since there is no training that can compensate for poorly initialized models.\n", "\n", "---" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## Python example: electricity load forecasting with ESN\n", "\n", "- In the following, we will use the [Python library](https://github.com/FilippoMB/Time-series-classification-and-clustering-with-Reservoir-Computing) `reservoir-computing`. \n", "- The library allows us to perform time series classification, clustering, and forecasting with Reservoir Computing.\n", "- In this lecture, we will focus on *forecasting*." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "### Load the data\n", "\n", "We reload the same time series of energy load that we used before." ] }, { "cell_type": "code", "execution_count": 17, "metadata": { "slideshow": { "slide_type": "-" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Loaded ElecRome dataset.\n", "Data shape:\n", " X: (137376, 1)\n" ] } ], "source": [ "ts_full = PredLoader().get_data('ElecRome')\n", "\n", "# Resample the time series to hourly frequency\n", "ts_hourly = np.mean(ts_full.reshape(-1, 6), axis=1)[:, None]\n", "\n", "# Use only the first 3000 time steps\n", "time_series = ts_hourly[0:3000, :]" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- To train the ESN we need input and target data `Xtr` and `Ytr`. \n", "- We also need test data `Xte` and `Yte` to test the model and validation data `Xval` and `Yval` if we need to do hyperparameters tuning.\n", "- We will use the function `forecasting_datasets` that given a time series `X` does the following computations." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "1. Splits the dataset in consecutive chunks: `train`, `val` and `test`. \n", " - The size of the chunks is given by the values `val_percent` and `test_percent`. \n", " - If we do not need validation data, set `val_percent=0` (default) and the validation data will not be created." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "2. Create input data `X` and target data `Y` by shifting the data `horizon` time steps, where `horizon` is how far we want to predict. \n", "- For example:\n", " - `Xtr = train[:-horizon,:]`\n", " - `Ytr = train[horizon:,:]`" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "1. Normalizes the data using a scaler from `sklearn.preprocessing`. \n", " - If no scalers are passed, a `StandardScaler` is created. \n", " - The scaler is fit on `Xtr` and then used to transform `Ytr`, `Xval`, and `Xte`. \n", " - Note that `Yval` and `Yte` are **not** transformed." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- The code below exemplifies the use of the function.\n", "- Note that we loose some data due to splitting and shifting." ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Xtr: [ 0. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.]\n", "Ytr: [[ 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.]]\n", "Xval: [17. 18. 19. 20. 21. 22.]\n", "Yval: [[22 23 24 25 26 27]]\n", "Xte: [28. 29. 30.]\n", "Yte: [[33 34 35]]\n" ] } ], "source": [ "X = np.arange(36)[:, None]\n", "Xtr, Ytr, Xte, Yte, Xval, Yval, scaler = make_forecasting_dataset(X, horizon=5,\n", " test_percent=0.2, \n", " val_percent=0.3)\n", "print(\"Xtr: \", scaler.inverse_transform(Xtr.T)[0])\n", "print(\"Ytr: \", scaler.inverse_transform(Ytr.T))\n", "print(\"Xval: \", scaler.inverse_transform(Xval.T)[0])\n", "print(\"Yval: \", Yval.T)\n", "print(\"Xte: \", scaler.inverse_transform(Xte.T)[0])\n", "print(\"Yte: \", Yte.T)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Back to our real data: we want to make forecast 24h ahead. \n", "- In a real use case, we must optimize the hyperparameters on a valiation set.\n", "- However, since this is just a demonstration, we will use the default hyperparameters.\n", "- Therefore, we leave the default `val_percent=0` and the validation set is not created." ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Xtr shape: (2676, 2)\n", "Ytr shape: (2676, 1)\n", "Xte shape: (276, 2)\n", "Yte shape: (276, 1)\n" ] } ], "source": [ "# Generate training and test datasets\n", "Xtr, Ytr, Xte, Yte, scaler = make_forecasting_dataset(time_series,\n", " horizon=24, # forecast horizon of 24h ahead\n", " test_percent = 0.1)\n", "print(f\"Xtr shape: {Xtr.shape}\\nYtr shape: {Ytr.shape}\\nXte shape: {Xte.shape}\\nYte shape: {Yte.shape}\")" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Define the Reservoir and compute the states\n", "\n", "- First, we specify the Reservoir hyperparameters and we initialize it." ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [], "source": [ "res= Reservoir(n_internal_units=900,\n", " spectral_radius=0.99,\n", " input_scaling=0.1,\n", " connectivity=0.25)" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Then, we compute the sequence of the Reservoir states `states_tr` and `states_te` associated with the training and test data, respecitvely.\n", "- Since the initial states of the Reservoir mostly depend on the initialization, e.g., $\\mathbf{h}(0) = [0,0,\\dots,0]$, we want to get rid of the initial transient phase.\n", "- This can be done by setting `n_drop` that specifies how many of the initial states we drop." ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "states_tr shape: (1, 2666, 900)\n", "states_te shape: (1, 266, 900)\n" ] } ], "source": [ "n_drop=10\n", "states_tr = res.get_states(Xtr[None,:,:], n_drop=n_drop, bidir=False)\n", "states_te = res.get_states(Xte[None,:,:], n_drop=n_drop, bidir=False)\n", "print(f\"states_tr shape: {states_tr.shape}\\nstates_te shape: {states_te.shape}\")" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Fit a linear readout\n", "\n", "- We are now ready to train the readout to predict the desired output given the sequence of Reservoir states.\n", "- We start by using a linear readout implemented by a Ridge regressor.\n", "- After the readout is trained, we use it to compute the predictions $\\hat{Y}_\\text{te}$ of the test data." ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Training time: 0.0344s\n", "Test time: 0.0003s\n" ] } ], "source": [ "# Fit the ridge regression model\n", "ridge = Ridge(alpha=1.0) \n", "time_start = time.time()\n", "ridge.fit(states_tr[0], Ytr[n_drop:,:])\n", "print(f\"Training time: {time.time()-time_start:.4f}s\")\n", "\n", "# Compute the predictions\n", "time_start = time.time()\n", "Yhat = ridge.predict(states_te[0])\n", "print(f\"Test time: {time.time()-time_start:.4f}s\")" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Finally, we plot the results and compute the MSE." ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "<Figure size 1400x400 with 1 Axes>" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "fig = plt.figure(figsize=(14,4))\n", "plt.plot(Yte[n_drop:,:], label=\"True\", linewidth=2)\n", "plt.plot(scaler.inverse_transform(Yhat.reshape(-1, 1)), label=\"Predicted\")\n", "plt.grid()\n", "plt.legend()\n", "plt.title(\"True vs predicted electricity load\")\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 24, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Mean Squared Error: 21.44\n" ] } ], "source": [ "mse = mean_squared_error(scaler.inverse_transform(Yhat.reshape(-1, 1)), Yte[n_drop:,:])\n", "print(f\"Mean Squared Error: {mse:.2f}\")" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Reducing the dimensionality of the Reservoir states\n", "\n", "- As discussed earlier, the Reservoir states contain a rich, yet often redundant, pool of dynamics.\n", "- The readout job is to select only the dynamics that are useful for the task at hand.\n", "- However, training a readout on high-dimensional states is computational demanding, especially when using a sophisticated readout.\n", "- In addition, working with high dimensional data can increase the risk of multicollinearity, which destabilizes certain models, and overfitting." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "- It can be desirable reducing to some extent the redundancy in the Reservoir states.\n", "- This can be done with an unsupervised dimensionality reduction procedure.\n", "- The most common and efficient dimensionality reduction procedure is PCA." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "#### Principal Component Analysis (PCA)\n", "- PCA is a statistical procedure that utilizes an orthogonal transformation to convert a set of observations of possibly correlated variables into a set of values of linearly uncorrelated variables called *principal components*. \n", "- The number of principal components is less than or equal to the number of original variables. \n", "- By using a few components, PCA reduces the dimensionality of large data sets, by projecting the data onto a lower-dimensional space with minimal loss of information." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Our data is a sequence of length $T$ of Reservoir states, each one of size $N_h$. \n", "- They can be arranged in a matrix $\\mathbf{H} \\in \\mathbb{R}^{T \\times N_h}$.\n", "- To illustrate the procedure, let's first consider a toy example with only $N_h = 3$ features.\n", "- In addition, we will create some structure in the data, by dividing the samples in 4 groups/clusters.\n", "- This will help us to see how PCA preserves the structure in the data." ] }, { "cell_type": "code", "execution_count": 25, "metadata": { "slideshow": { "slide_type": "-" } }, "outputs": [], "source": [ "# Generate 4 clusters of points in 3 dimensions\n", "T = 300 # number of samples\n", "N_h = 3 # number of features\n", "H, clust_id = make_blobs(n_samples=T, n_features=N_h, \n", " centers=4, cluster_std=1.5, random_state=1)" ] }, { "cell_type": "code", "execution_count": 26, "metadata": { "slideshow": { "slide_type": "skip" }, "tags": [ "hide-input" ] }, "outputs": [], "source": [ "def plot_data(H, clust_id, interactive=False):\n", " if interactive:\n", " fig = go.Figure(data=[\n", " go.Scatter3d(x=H[:, 0], y=H[:, 1], z=H[:, 2], mode='markers', \n", " marker=dict(size=5, opacity=0.5, color=clust_id, colorscale='Viridis')),\n", " ])\n", " fig.update_layout(scene=dict(xaxis_title='X Axis',\n", " yaxis_title='Y Axis',\n", " zaxis_title='Z Axis',\n", " zaxis=dict(range=[-20, 10])),\n", " margin=dict(l=0, r=0, b=0, t=0),\n", " width=400, height=400 \n", " )\n", " fig.show()\n", " else:\n", " fig = plt.figure(figsize=(4, 4))\n", " ax = fig.add_subplot(111, projection='3d')\n", " ax.scatter3D(xs=H[:, 0], ys=H[:, 1], zs=H[:, 2], linewidth=0.2, alpha=0.7, c=clust_id)\n", " ax.set_xticks(())\n", " ax.set_yticks(())\n", " ax.set_zticks(())\n", " plt.show()" ] }, { "cell_type": "code", "execution_count": 27, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "<Figure size 400x400 with 1 Axes>" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Set interactive = True to get an interactive plot that can be rotated\n", "plot_data(H, clust_id, interactive=False) " ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "Reducing the data dimensionality with PCA invole the following steps:\n", "\n", "1. **Standardization** \n", "- Scale the data so that each feature has a mean of 0 and a standard deviation of 1. \n", "- This is important because PCA is affected by scale." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "2. **Covariance matrix computation**: \n", "- Calculate the empirical covariance matrix $\\mathbf{H}^T\\mathbf{H}$.\n", "- The matrix shows how changes in one variable are associated with changes in another variable." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "3. **Eigenvalues and eigenvectors computation**\n", "- The eigenvectors of the covariance matrix represent the directions of maximum variance.\n", "- In the context of PCA, the eigenvectors are the principal components.\n", "- The eigenvalues indicate the variance explained by each principal component." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "4. **Sorting eigenvectors**:\n", "\n", "- The eigenvectors are sorted by decreasing eigenvalues. \n", "- The top-$k$ eigenvectors are selected, where $k$ is the number of dimensions we want to keep.\n", "- In our case, we keep $k=2$ dimensions." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "fragment" } }, "source": [ "5. **Projection onto the new feature space**: \n", "- The original data are projected onto the selected principal components.\n", "- In our case, the 3-dimensional data are projected onto the plane spanned by the first two principale components.\n", "- The projected data are the reduced-dimensional data." ] }, { "cell_type": "code", "execution_count": 28, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [], "source": [ "# Apply PCA to reduce to 2 components\n", "pca = PCA(n_components=2)\n", "H_pca = pca.fit_transform(H)\n", "v1, v2 = pca.components_" ] }, { "cell_type": "code", "execution_count": 29, "metadata": { "slideshow": { "slide_type": "skip" }, "tags": [ "hide-input" ] }, "outputs": [], "source": [ "def plot_pca_plane(H, clust_id, v1, v2, interactive=False):\n", " mean = np.mean(H, axis=0)\n", " length = np.linspace(-15, 15, 20)\n", " another_length = np.linspace(-15, 15, 20)\n", " xx, yy = np.meshgrid(length, another_length)\n", "\n", " def compute_z(x, y, v1, v2, mean):\n", " z = np.zeros(x.shape)\n", " for i in range(x.shape[0]):\n", " for j in range(x.shape[1]):\n", " point_in_plane = mean + v1 * x[i, j] + v2 * y[i, j]\n", " z[i, j] = point_in_plane[2]\n", " return z\n", " Z_grid = compute_z(xx, yy, v1, v2, mean)\n", " \n", " if interactive:\n", " fig = go.Figure(data=[\n", " go.Surface(z=Z_grid, x=xx, y=yy, colorscale='gray', opacity=0.5, showscale=False),\n", " go.Scatter3d(x=H[:, 0], y=H[:, 1], z=H[:, 2], mode='markers', \n", " marker=dict(size=5, opacity=0.5, color=clust_id, colorscale='Viridis'))])\n", " fig.update_layout(scene=dict(xaxis_title='X Axis',\n", " yaxis_title='Y Axis',\n", " zaxis_title='Z Axis',\n", " zaxis=dict(range=[-20, 20])),\n", " margin=dict(l=0, r=0, b=0, t=0),\n", " width=400, height=400)\n", " fig.show()\n", " else:\n", " fig = plt.figure(figsize=(5, 5)) \n", " ax = fig.add_subplot(111, projection='3d')\n", "\n", " surf = ax.plot_surface(xx, yy, Z_grid, cmap='gray', edgecolor='none', alpha=0.5)\n", " ax.scatter3D(xs=H[:, 0], ys=H[:, 1], zs=H[:, 2], linewidth=0.2, alpha=0.7, c=clust_id)\n", " ax.view_init(elev=20, azim=20)\n", " ax.set_xticks(())\n", " ax.set_yticks(())\n", " ax.set_zticks(())\n", " plt.show()" ] }, { "cell_type": "code", "execution_count": 30, "metadata": { "slideshow": { "slide_type": "-" } }, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZQAAAGVCAYAAADZmQcFAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAPYQAAD2EBqD+naQAAZNtJREFUeJztvWmUrFlZ5/vb7xRjTieHM8/n1EhVUVQxWyUolIKC0IiWNtoO7aJFaFar4MWmcaFw7Xu9qz+IC6+C0thwVXAAFUGwS6hiqImqOhQ1nHkec86M8R32vh8i3jiRkZERkaciT2ZkPr9aWSczcscbOyIj9n8/41bGGIMgCIIgvECs1Z6AIAiCsD4QQREEQRC6ggiKIAiC0BVEUARBEISuIIIiCIIgdAURFEEQBKEriKAIgiAIXUEERRAEQegKIiiCIAhCVxBBEQRBELqCCIogCILQFURQBEEQhK4ggiIIgiB0BREUQRAEoSuIoAiCIAhdQQRFEARB6AoiKIIgCEJXEEERBEEQuoIIiiAIgtAVRFAEQRCEriCCIgiCIHQFERRBEAShK4igCIIgCF1BBEUQBEHoCiIogiAIQlcQQREEQRC6ggiKIAiC0BVEUARBEISuIIIiCIIgdAURFEEQBKEriKAIgiAIXUEERRAEQegKIiiCIAhCVxBBEQRBELqCCIogCILQFURQBEEQhK4ggiIIgiB0BREUQRAEoSuIoAiCIAhdQQRFEARB6AoiKIIgCEJXEEERBEEQuoIIiiAIgtAVRFAEQRCEriCCIgiCIHQFERRBEAShK4igCIIgCF1BBEUQBEHoCiIogiAIQlcQQREEQRC6ggiKIAiC0BVEUARBEISuIIIiCIIgdAURFEEQBKEriKAIgiAIXUEERRAEQegKIiiCIAhCVxBBEQRBELqCCIogCILQFURQBEEQhK4ggiIIgiB0BWe1JyCsbYwxaK0BUEqhlKp9LwiCUI8IirAkWmuKxSJBEOC6LpZ11aCNxaXxq/538feCIGwMRFCERRhjCMOQQqFAFEW4rott2wt+H3/FP0NFPIwxi4Ql/jkWpGaCI8IjCL2PCIqwAGMMxWKRcrkMgOM4WJa1YMFfavGvF5P453rRiQWn8Rpi7QjC+kAERagRhiHFYpEwDLEsa4GLqxMaF/xWwtP4c6PwNH4fX0+ERxDWLiIoAsYYyuUy5XKZKIpwHGdFF+VrEZ56N1v97eJmE4S1gwjKBicOvPu+j1JqxcVkOYibTRB6CxGUDYoxhiAIKBaLRFGEbdvLdnGtBa6Hm63RyhHhEYTmiKBsQJoF3tf7gvhC3GxRFC1yszVeS9xsgiCCsuGIA+9BEPSsVbKSdOpma/xZ3GyCIIKyYYgD76VSCa31hrBKVooXmlQQ/7xcN5tYO8JaR5nG7Zaw7oiiqGaVxIvVchYjy7JwHNl7rATNrJ367xuz2GLEzSasRWSVWMesl8D7eqabSQXiZhNWGxGUdYrWmlKptKEC7+uZbrnZ4vu2crOJtSNcK+LyWofEfbjCMOyKVaKUwnXdLs1OWG264WZrZemI6GxcxEJZR0jgXeiE61G7I262jYkIyjohDrz7vl8LossHVnghiJtNWC4iKD2OBN6F1abT2p363zUWjV6rm01EZ20hgtLDSOBd6BXEzbYxEEHpUbodeBeEtcBKutmkdmflEUHpMYwxNatEAu/CRuWFuNnqf9epm02snc4QQekhJPAuCJ0jbrbrjwhKD9AYeBchEYTu8ULdbHLg21VEUNY4EngXhLVBN91sza63HqwdEZQ1TGyVSOBdEHqDlXSzKVXpWGHbdren3TVEUNYgEngXhPXNtQiP1npF59QNRFDWGGsx8C7t3gRhdWi0VFZ7LWiHCMoaQQLvgiD0OiIoawCtdc0qAQm8C4LQm4igrDISeBcEYb0ggrJKxIH3UqkEiFUiCELvI4KyCjQG3pd7xrsgCMJaRATlOmKMwfd9SqVSTwXeJctLENYGa329EEG5TvRy4H2pJnqCIAj1iKCsMMYYwjCUwLsgCOseEZQVRALvgiB0g15xO4ugrBCxVRIEAZZlren+O4IgCN1ABKXLGGMol8vSh0sQhK7SC+uICEoXqQ+8K6Wwbbsn3gSCIPQGa309EUHpAnHgvVAoEEWRBN4FQdiQiKC8QIwxFIvFdX8A1np8ToIgdBcRlBdAfTpwXPEuCIKwEvTCpk4E5RqoD7z3UsX7C0EKGwVBaIcIyjJpDLxvBDERBEHoBBGUDmk8AEsC74IgXE96YeMqgtIBGyXwLgiC8EIQQWlDfcW7WCWCIAhLI4KyBHHgvVQqScW7IAirivTy6mHiA7CCIJDAuyAIQoeIoNQhgXdBENYqvbCpFUGporWmVCpJ4H0JpA5FEIR2iKBArQ+XHIAlCIJw7WxoQZHAe+fI6yIIq4dSqic+gxtWUOLAu+/7WJYlYiIIgvAC2XCCIoF3QRB6DUkbXoNI4F0QBGHl2DCCIoF3QRCElWXdC4oxpmaVSOBdEIRepFfWrHUtKBJ47x5ShyIIq0svfP7WpaA0Bt5FSARBEFaedScoEngXBGE9YYyRLK/VILZKJPAuCIJw/VkXgiKBd0EQhNWn5wVFAu+CIKx3pPXKCiOBd0EQhLVFTwqK1rpmlYAE3gVBENYCPScoEngXBEFYm/SMoMSB91KpBIhVIgiCsNboCUFpDLxbliViIgiCsMZY8/6iz3/+8zz11FP4vo/jONi2LWIiCMKGoxfWvTUvKH/0R3/Eww8/LC4uQRCENc6aFxTP8/B9X8REEARhjbPmBSWZTNbSg4XVo1d6CQnCeqOXPntrXlASiUQts0sQBEFYu/SEoIiFIgjCRqZXXP5rXlCSyWStFb0gCMJGpRdEZc0LSiKREEERBEHoAda8oIiFIgiC0BuseUHxPE8ERRCEDU0vuLugBwRF0oYFQRB6gzUvKJI2LAiC0Bus+eaQkjYsLJdAR3zp4pMcnb8IwA192/ixrXfiyFEHQo/SKy6vnhAUiaEIy+Ezpx/iubkLtQ/h5MRzbOUfeUXfGRQltHUAk/pNsPtXeaaCsL5Y84IiWV7CchgvzfHc/IUFO7o3DHyDW9zvQ+SglIUdnUPrI+jsJ8DqW8XZCsL6Ys37AMRCEZbDpdLsgt5HfdY8tySPYlsaHd+oFFZ0EVX6u1WZoyCsV9a8oEiWl7AcDmQ3k7K92s/DzgyeFaJQ2Kru7a4MmJOrMENBWB7SHLKLSJbX2qBX3tQpx+MHR2+pzfdCMEbZuHiWzcKwpgPW3tWYoiAsGwnKd4nYQjHG9MyLKqwuP7z5VvZkRnhq5jQKcBJvwTN/D1wVRW1tw7g/DHocVBZUatXmKwjrhZ4QFLFQhOWyP7uZ/dnN1Z9eRli6Ecv/AlDAWDdgnNuxyn+C0peBNNq9C5P4WVD26k1aEHqcNS8oUociLMD/Flb4EOhZjLUDk3gj2B24rpI/ik7+aOX76Dx28f8CNCgPCLGC76CjWVAeShUw1m6M+wZQAeh5sLaAWvMeYmEdopTqGe/MmhcUSRsWavjfwS7/JXEwROnnMMVT6NRvgz10dZyeRflfBXMJ1FDFtWVvBWOAMir4FsQ5X8aAKUJ0BpvHQGUxKo1iM6b8tyg1AAQYazPaezO4L73OT1oQeoc1LyiSNizEWOG3aIiso0wBFf4bxv53lRv0PFbx/0HpKaju6kx4CO28DCv6HspMYfQsEAEeSl8AMwuUAAVmHmXmgIsYbOb1djxnCwk1jV38JDr4JspMg+pHu/eC+7Lr9wIIwhqnJwQlCAK01ti2+Lc3NGZu8W1KVVxS8Y/B1xaICYDSl7BLfwHObjAGZTRKn6NipWigTCVgb6rfWxijUQQkzGkmS/NY9ia22JNYZhKsUWAcOzpJhAH35Sv5rIUNTq9kWEIPCEoqVcm+KZVKZDKZVZ6NsJoYaycqmqq6rqroKeAwqvgJjPMS0FeuiokOK24vfRmFwYRllAmBAMhX/7Xhaslj9YF0zRByrJAxNQlMoI1FGBYIzDkUCmX1kVBfQ4ugCALQA4KSSCQAKJfLIigbHGPthfIXUGYelIcxAYoAy4xDdBzCp9DWpquCo0+hTAEIgQhlxoEE4FIREUWlFEtTn1Ico+PfKoOqWjMKTWhcDAormiEqf4tk9jo8eUHoAdZ82kosKJLptcEJz2KX/t+qQAQYM41iCiigzBRKH0VFz6L0DIZsNRZSqForLpW3uqJilQTVn+O3/+IMGkOlmD5Gm6sjk1aIqyIMmmKYg+gC2hhO5q5wuTizMs9f2LBIllcX8TwPpZTUomxwVPFjqNidZVwU+epv6lZ9M4uKzhGl/yvK/wK2GcfggjWMip7naqwkpM7+qHw1cVMrVbk50uAbRcoy2EpXpUjjopgO05ydeY5Pnvka+aDEoJdhf3YzP7vr1Qy6bsUlZ20C5a7kyyMIa4I1LyhKKalFEbD06bpAexxAh0VKYObBuRljb8EUfv9qppfZXsnoIuCqy0sDUc1DZqhYIPGjxE4xx9IYffUXxoAxioJxSNgBH3nucYqmYu1MBzkC7fPU5T/kh/oPYelLGOWg7Vsx6d8Ge6y7L4ywrjHGSFC+m8SCIhbK6rLab2pjpVBR/ZLfLPZh0PYujuVzPDd/mVvczRxwvotlGSCJUUMoMwPKAQOGQu3uoQaUwsaAgtAoIqPwrEqA3lGVh4qqj2grTZKIgjHsS41zuDjGj216nptS44y5BXYl5lDaA+Wg8LGjJ9GF30FnP1Z5fEFYh/TEO1tqUTY4RoPpBxMCPpW3bZydlYwHYcjwjdxr+cfxf2Wfd45XbPo3SlGBFBaKFMbaDyZT9WWNo/CpSAS4FmhjUKoiVzYGpQzaKBSgMdhVUbEUgMIYm6wT8eObDnND/jIvSl8GIGvnSVhBNZB/NWKvojMQPgbuK6/HqyYI152eEBTP80RQNjCq9FmUngfS1Vs0hgG0tRPLTKBMGWONcFm9iX8at3CU5t8NfZU+u4ABAh2SsGZQ+hCQwGgPRUAsJjFWXdxTqcqHQ2MqDraqy8uqCg4KkpZGabg5fZn9iXHO+H2AwjLxHBXKBFfjJwowMyv1MgnrFAnKdxlpv7JB0SVU8c+x/L9BKYNRaVCDVQsji8n+LpE1VukYbI3x3UvPodSz3Jg4ScYqAKDQOMqnYtFEaJJoM4NFiCJskt+10ImmgMiAY139ufI/g1KaCI+Eisg6PhEW5/0sOe0yahSR1oSU0cbg2Ta2yoJ9x4q9XIKw2qx5QVFKiYWyQVHlP8cKH6taExaKHJgQYx2oLOr6DNg7wd4BQJ+bQhtD2i4QYeOgcdTVaLpBc8U3OMpjyC4tsEgWPC6VNOHY/VWfXGygpjimGrY3GMrGpt/xuRIYpqMU/UGZUbeErSJsVcSPXPLcxpfPneRU4TESlsPtA7t43eYX9czuUxDasebrUEBObdyQ6Dms8BmwUg0pt0UwuUpg27plwV1eumk/Q16ac/4W5qNMNXvLoKqC4keK+cgmYQUExmuWKVyj0f1lUU0hrqtHqQhPGV87nCn1YQy4Vog2cMHPcKqUZjpIMxn2cyEY5VzucS7nH2c2KHClPMdXL3+Pf7741At8oQRh7dATgiJZXhsQkwMCUBbGGq3zQykw5UpjxvoOw4Br2fzSntcykLqdZ8svYjYaIqIPbaAUwfFiPzYFtDHMRllykdf4qIvQUM0Iq1gtBovQVPPLTKVG5bw/SMF4PJUb5e/GD/Iv0zt5Oj/MRDTAaX+IqWiUfJTEAC9Kn8JVNp7lkLITPD17lkhHtfTQxq+VJPbNd/plWVbbL9u2sayeWFaEFWDNu7xAsrzWMq0WvU4WxHjMohM5zRhG3VrpBKw2g7UfyGGMjfb+A9gvQkXRoscZcTP8h533gPkBiE5xIf8U06Vj9DkByvVxLJ+yikg6ighFXkW4ca0KFgaFTQQqQaXhSqXCXmERVX9ficcoLBShChl0RrGweaYwyPb+TSjgtr7LjLgBAI5lV+eoSUceJnHT1aeJITK6ZkU1Y7kuseWMX4lrK6WwbZsoitqOFdYXIigdEO8W6z9MzRbLpRbHVmPbPW6z29pdu9m4dteOx7aaV+x27OTxYyzL6ninvWhnq96Cig5RqWynYq04N4PajQmC9jthtYtzQcR0tI8w8Ela0+xMTLHHmanGViAyCl+ZmlQAhBgsCmjVV6unN3Wus8hUhKakbXLa4WwxyawZZSpKsjWVRaEom3lgBqr3StgOgY4oM0gx9Ek5LqDIOknsFo6C+LVbzsK/nOOytdbLsiiWc23bttFar3oN03qgV+JsG1pQjDGUSqUF8ZlWi3/8AWn3xzXGLHvschfdTsYvN92w3eIf/y4+SqCTOXTyGtSPjecBgL0ZY70G9NlK3q61HaxK6rBSqqPFsBCWmfZzRMZwMDVLWuUoaYeMHaAwWMpQ0jYpqyJaBrCUQpPBMnMolUVjUYhcHGVwrMocp0OPknZ4IjfKTJhAG41SZRSKlO0xZ+1gzJTJ2AEDbgbXspkJDBf9YZK2SykKSNouB7KbO3qPxM+5k3Hx952+7lEUtXyvNP6d42t3cv1mmzFh/dITgrJSQXmlFKlUCtu2KRaLtQVtKaIo6mg3Fy92nboH6hfSdot0fF2lVNux8Yc5vm67+azE84vnAZ1bK1EUXT37RnkYa9+ix4qfT7PFNr5tOigwXpojF5YxRtOnZgiikDkdkvQ0mqhSLWLAEFZiJcoQoVAYIkCbaQISjPsZLAwZO8BREZdLNt/NDTIXRSgKWFh4ymYuzDEFTNku5cRufnDYw1IlQpMh4+3nhkyeK6VZbMdid2qYrJ0mDMOOXr/G59lqHDSx+Nqwkq61eC4iLOubnhCUlQ7Ke56H4zgUi8W2whUv/p1YIPULaePYZgt8vKDXiwA0d0fF8+jkAxq7HZotMI3ziKJoybHNHms5u88gCJrubJuJTDx2qcdtRqNgnZ67TFI5lI1NPgrRkY024OOSt4p42ESYaoqwXX2sBTMjMGnK2qAjg7JS5MMUxhgO57dgkSar4tLHqmVjDKH2sZRLEBkenIzYntxMZCKUukjaSXBT3zZsZYExROVnsDmHMRZYQxj7ZlCpjp5vO2ILrtO/0Uq6v+rf/yIq65eeEZSVThu2LItMJoPruuTz+QVZNs0W9thN0IxmLoL4/p3sNju1QOKxy41TdDI2DMNFbq1W8ZpOXWBLiVUj9a9Np69FFEXYloLoCOgrbLXmsF2PWb8f17KZClNs83IYDDNBmgHH4KqIo8UB9iYnSVlhpZdXlbJxCLRVbRqZIKFmSFgpCpGDQhPoaEEoPawG11OWRz4sUtIBScvlcmkWgKTtknWSnC9M8dJN+3D1EVR0Eq0UlipjtI8y8xj3Xpq11I9fv+UsyMuxJDt1I9bPYzkxnnr3l2SCdUavxZ96RlCuV9pwbK3k83kKhULb8fUZLZ34uJtZK83G1cdh2l0zdhEtx53Ubmy9W6vdgr6csbDQyuv6nIOnsPUpsFxSVsgme5aDyTmezI/ybGGIhArYlZzHVYZI2+Dcwt6BO7iUf4ox+zkydrHymEYxHdqUtUVep9jmzpDE4JhJbCvFawdyPD4/xtHSpsq8VN1iX51mKSpRsnxsLBzLJh+WiSJNyfY5PHeeFyWPAX616aQC5lDMY9RpsHcveG7LjWM0e606jcF0IvrN3GvL2dTEx1II64ueEJRkMsn09PR1ezzLsujr6yORSDA/P9/Sx22MIQxDLMvqeDFdTiyh0yD8SsY/GhevpSyu+nnU76Qbv49ZygVW/9jNvq/9rC+goqNg8kAWY+3Eik5UOgKbEoVwjkiHbLemCdw8l/wMOigwR4StFAZIBycJrDSXy4N8Z34nuxJT7Ejk6LN9HMoUIo98EJHTioSr0crDokhSWbwsfZlSkGIizFKOQuKMLltZhEZjY6Mjg28CPMfGQqGjCFu55ErzBI5m4UfQxlIBhjxGBx39fZaT0LGcBXw5mxRY3vtaa43v+yIqHSK9vLqM53mrUinveR6bNm0il8tRLBZbLqT1cYqlxsULa/3YZot7s4XUcZwFef2tMnrisZ3sSB3HaWsFxWMTiURHY2H52WjLdttF46jwyVr8AuYgOkRkIrASaGPQkWbQnUWbNFu9Etu8Eq6KyEUJtFJ4lo1vAgr+s5zKbWMq9AlNhu2JPPlq0aMCdifnyVgBdvUIR4OH0RrPynFTepJHcoOERqONwVIKS1koo2vnqSSUw2xQIG17JO1K1b9jJ0D1gZlf8Ny0dsHZvCw3X6fuo+W4p5a78YDli0q5XMbzPHF/rSN6QlBSqdSqVcorpejr68PzPKanp9tm5EDni2m8C+wkVz8IgtqOsXFs44c+jn+0m4NSakHKaLfGwlXffSyarah323WyiEVRhGOO14lJjA2UgEoab9Ly0TqNoYSvXdK2IWlpSloTYRNVWwg7yicwIQnlsNObxKLI1bb4lbNRXNXwHJRFpBNk7AkCvauaF1YZ41iVa2sd4Vg2oY5IKZdS6OPhkHGSbE8OY+wUKngMiCrJAMZgnL0Y1YduEIpmr3l8W7MEjVZ/o2YB8nbp4vVzWWoeQO3z0akrzvd90uk0jtMTS5HQhp74K66WhVJPIpFg8+bNzM3NMT8/33JsfYygnaUQ/76TyuJ4bCeLdP11O1nQ45hNu/k2xneWk4rcTog6tX6UUuiwCFEOZcVnxVNtkeKidAEbjTZzaAO5yCUIS5RNmYQdoXSZsnZQ1f8KOkXBL1OOArRbwg9CUNNYJABFqBT9poilTFU4KoTGIhdkmCnlcZVT6UyMYVffMJ7jcKY4QTkMoWq9JJVDGIbcmNpMn/LwQw+4B6VPg4kw1hhEIxD5tde6Uwsyfl06tRCW8zeEqyLU7eQPgEKhQCqVwnXlmORm9Iq7C3pEUNZKc0ilFAMDA6RSqY6slWaB6mYfyjgOU78TbOU2q49TNNJ4v6BaUd5qbP33juM0tYKAq7UhdT8vNXap+3bqAmu7eIUDqGgalMGxI7T2AIOx92HUKEqfJ9RlcmFEMQIwlLRLkhnKYQJjKvOJUBwvDuKgcCyPaT/LLrcIuGhVxlYJbBzGw1E22bM1t1eEItCKw8VBlKmcuWLbFo6y6bcr1k3GSZAluaDK3laKUbfv6qZAJTH2jUu+Dq3+1o3E1kon4+szujq9dqNLtxX12WLtxhtjai5lz2vfX01Yu/SEoKy15pCe5zE2NsbMzAwzMzMdfcAaYyBL0am1Uj+2U2ullWUT74Ku1Qrq1LpajsXkOJUdfbMd2pX5rTjRGVyngGc7ZNNFIjNabW3vYKxhHJWgXx1FU6il+J4obuZKaDHsaLTxOO9nmQg8PKvyGsyEfVz2S2zxciiSWEQEODxf2MOYPcmOxBUSVkhZOzxXGOBYabDS4l4pwijCVxHnS9MYDIXQx1YWvq5sPFzLZk96FNuyFrj42tFoRSy1Qan/Gy4Vl2gU6vrH72Qn3KqqvjGWF9c0dbKZMMZQKBQIw5BUKtVTu3LhKj0jKGvBQqlHKcXQ0BCZTIaJiYm2glcf12i1iMTWSuzqaPchDIIAz/PaLujxdV3XXdbYZnMwBs6NzzCVL+JaFttH+hke6Ft03WaLQn19S6fussaxh8+Nc/jcOJbaxVjfFZJOiaH+zezacidK1S2kzk1YymFQneViaYIrgccjs0kiU3F1ubaDhSIyIWnbI2G5GAtK1i3M41MKzjDuay6V+wmtAhdUmhPFYRSGi+UEkTEYKn93RSVbOOskyAclbCzmgwJ+FBKfypWwHAI3qFm2sVjGi3y7rLa4m2+nLrDljO10ExHPZTmuuOVYWWEYorUmk8mIqFTppdehJwRlNYPy7fA8j61btzI7O8vs7OyCNNtm/0JnLTGWUwkfu8uWckHVL8ixMMcf8HautfoFL57Td4+eZ3wmV7mmglOXJrltz2a2bOrveFGK253HY40xjM/myRd9NvWnGMhcrRavX+yCMOLE+SsoYzAGLs+OAnBuSjHYV8Rz7AZrcCewk+Ek+GaatH6eCIOrDI4JmNcWGSvBTekt3NS/vfaYR+cvcr64BYCEKpPRc9ycPUdGVdLDb/Bsni2MMBulamKigJRycbVFQEQUaTJ2gqydpPJSKa6UZskHZVK2u+j5tUuKqN9sdJLFt1x3WadWU+yK61SwluOKAyiVShhjyGazPbWYCj0iKGshKN8KpRSDg4Ok02kuXbrUUUFkJ9YKUPMrx+6DTse2wxjTkbUSE+9IJ+YKTMzmsepOoDLGcPLSNJuH+tBa11xVzbgwOce5iVlKfkhfyuOGnZtJujZPHDvPTL5UWUAuwNZNWV60Z8uChdR1XcZncwTR4kByqDVT8wW2DPXVrLb6OShgW2qIm/u3oYPvMeLO4ypNSbtc1iP0uxmemT1LWVeslfPFmdpjaKO5KT1BCgutKqdHJizDzekJHs/tos/JUIzKBCbCU9WPlDZEJiIyNlbdXA0w4+dJpQYXvH5RFHXsFm3WxWApGuN47SzD5WTbXYu7rF3mWky5XCaKIvr7+yWtuIfoCUHplTPlPc9j586dzM7OcuXKlZZiES8cruu2DO4rpWqB9VYLdePYdpaCUmrBwtRucQrDEM/zmMmVaLZ25Eo+lUtU3HDNXHYXp+Z45vRltDEEYUS+7DNbKDPUl2Y6V8SqW2DOT86xqS/D9pH+BbGB/nTlmF+MuWoWVJ4RA+lUbbe/VBzhjv4C86U5SlGIwdBvG0atcb479yzzunL/IAqZCfKMJPqrdSR50tZM5VCtyAAaAySURYIJjEoRoQmNxihDSftoY/B1SNryyEflBS3q+93UIqsVKotoqwLZ+tsaLZRWlmb873I6YC9nsxHPuVMXWKfuMt/3CYKATZs2LUoIEdYmPSUonfjdV5vYWslkMly+fHlJayX+4NdXJHfqwmhWCNnoWltq4WhcyOJFo5O6gTAMcSzIF31su1rHUP1dKuFijEbrhfUI9bvu4xcmmJrLUyz7hFFlUXZti+m5JI5TObTKrtOAixPTjPYna/M4c2WGyzM58rkixVAzlEmgqp1OtgxmsZVetPFQSi0QYhVcIGmyeGgiDA4Wc+Uig3qCOX8bkQWuUVghFE2JjJMgqTyiwCUwuvacjQHXKaNCB0sZfmDgIHNBkQulaco6pM9OMJxIMxMWUCGEKsJBsS05hKsVwRKV8PXvh04skHjzUC8czYjdVPEGptV7bbmbjeXUHEGlpioWrHbv+SAImJiYYHh4eEPWqnSaer1W6Im/0GofsHUtuK7L9u3bmZmZ4eLFi21TjJVSuK5LELRvuREHZzvZQXbqWgPaJgJordk6lOFU2mG2GOBYqho/MOwc6W/6HONq6CiKmC+UKJZ9gvBqh14/iJicL7B5MIulIIgUbrW0xK6zMI6cn+D05WlQkPBsQq2ZK5bZMdzPyECG/Vs3ATAxl2e+4DPcl6I/k8QYU2vzUZlf9agAZdVshsqJiRWhsCMIbU3SditV90DJJCmbNA75mlGkFOSCPsZ1kq0YRpP9jCb72ZcdIzIGu7pQnitOMV6ew2AYSfWzNz3WdpGINw6d/I211mitO7Io6oWik/dDJ4Ws9YWRncZJ6i3ITjZSWmsmJycZGBggmUy2HLseWeub6HpEUFaQOBMsm81y4cIFZmdnlxwbL3xxHUirD7zWmiiK8DyvbSZYp641uJp11Mq1ppTi7oM7OXZhgulCGde22T7cx46RgSXHx1X+QWSItFlY4V6J65MvBfSlPRxliDRYdiV7DCDSmvMTswsa8GaSLga4aedmhvqSRNrw6OGzTM8XqwsgbBvu4459WxcsYMoaROm5BddyrJCpKE0YV8PrSrX7WHIQRzkYDFbiTkqlb5O2CxhjmNcJvl8YJNK68hzqUmnrnTO7s6Pszo7Wfo4X0PqFuFm1ehw3ql/QW7nCfN/vqN4IFrvL6scs5TqLrY9OBKATKyhmOZlopVKJ0dFR0ul0R+OF609PCMpaKWy8VlzXZffu3czOznLx4sWmO8nGoGksLK2KIusXp6Wyyupda+388/ECEMdhGh87/n3Cc7h1z5ba/ZZaFOof3/d9lA6ZzRWxa5YNWJYi6dq4lsJoje9HZJIOezYP0p9OEAQBxXJALl/CshefvH5pcorJWZsz43Pki6UFi9iZSxNkPJudo1fFTnEQh0nm5ycIosoBvyV2Mhtkq0dqVUhZDrckt5FOJGvi+r3gpTw5fQIFzOtE5bVQil32EOVyuW2Mq55OBD5+DTtNtIDlLdCtYh+N75H4lM5O5lHv0mrHctr5GGO4cuUKIyMjZLPZttcWrj89Iyi9aKE0MjAwQCaT4ezZs0xOTrYdH/uMO6kbSSaTtfThdsSutXY7yEZrpVVgNJ5rs8VpJlfk0ImLFEo+CUdRCjSubeHYCttSJD2HrZuy3LJrjOMXJjk9Ps2jz5/meyfOM9CXZd/mQVIJh5l8iXIQVuMXFq5tcfziJNoYpudLFAPNSF+SpOfU5j81X1ggKAab7xzeRTlMkvV8ZksD+JHH9s19RBmfchSQcZLsy27GtewF7rLbB3cTGc2p/AS2iipH+GY2sycz2sS11pr6RbddunCc5NBJPGM5hanx7zvJLquveer0+dUH31tZNvWblnZjAaampgjDkIGBgZ5yB10LSxWRrlV6QlASiQRRFNVSR3sZx3HYu3cvQ0NDnDlzpqXlFbsN2qVNK6Uol8vYtt3RLnmpLKxG4kUykUh05C6Dq0WoCwoRz45TDkNc1yaTdAl1mVBrQg1J10YBO0cHeOT5M5wZn6XsV641V/ApBRH5YgnXccgVfbTRGAN+ALZtkfAcSuWQUhBgNFyayTGQTjCYSYGqWED1cxmfzTGRK6LoZ65UqXBXCsYni7x2x/6mH976lN67hvfz4k37yIcl0k4CR1kLrMEFxxe3oT6lt/4a8Wtfv2tvTPlttqOv/31jDUrjteofJ7ZI62ufmrnAoOJ2agzWt7pupyIR02nqdKlUwvd9RkdHe2rBXe/0hKDEgbhSqdTzghIzODhINpvl3Llzi856afwQxtk5nfQGi10T9ddq5gIzZulDlBo//PUZW42LTqOrrt5dprUmjCLOTsxSCkIU4IeVgr8o1NiOwphKgP3C5CwnLk5S8AOMvrpAFYol0qlKrKgUGqgG0K/OLSCKDJHWRJEGpZjLRQRhxGDaY7RvoXU7MT1HVJf4EAIoizDwyeULuM7SYlBvsXkowshnKZltF4t6IWOX6y5bjhuuk3qnWGCbWUxLiVynrrh409jJnGdmZoiiiM2bN0utyhqhpwSlXC7T19e3yrPpHo7jsGfPHoaGhjh+/HjbbgBKqY7b0MRZQp18MGOxaveBD4Kgpftxar7A5ekclqXYOTrIYF+G505dIlcoEK8zflg5ptdxFZ5rE0YBU7M+jxVKFYvMVDKoaguTqiwyQVgRkkCDayksq9L1t1AKcB2rcvQvFlqbSj9gE3LLnt0M92cWLHIDmdQCwbQATITnJXHs1otSvcXWLhtvJceWy+WOrMZY4DsVik4tV7haXNlp7GM5hZvx84s3J63I5XJEUcTWrVvXZa2KpA2vAPWCsh4ZGBjgjjvu4MyZM1y6dGnJN5ExZoGV1urDGWeCdZLQEH9wO0lbjh8/XiRijp6f4OSlqdpCfXZ8loPbhjk9PofnOJSDah0IEEYa21IUy6ZSpAiUfUM51CRdq1bLUrtDTWTAtUAbQxSBU62F0dpgOxaWBZaqJA1kkwm2DqZrFl1s3Q33pxnuzzAxV6hdHhR7xwZqi2mj+wkWn/lRH3Ru5XKp3803FiI23i/uMtBqTL3V2ukpobGY1KfpLmXhNo5tfOz6+8R9t5bq+VZ/H2BBjKlVdluc3uy6bkef+XK5TKlUYvfu3evGg9GrKNMDEqi1xvM8vvvd77Jv377Vns6KMj8/z/HjxxeJQLNFq9XpkPW3LbUwNhLXETR7rDibq/53cTpu0Q954KljV33ptftUalSUgbligB8GhGHFDWZbFpGp/A4Fjm0RBBGaiihcfQyFY1kEYVSrAYn/jVB4dqVHlmNZRFVhUUDKc9g+3MfO0UFGB7P0ZzOUy+WquBhOXp5ier6IY1tsG+5nbDBbe306rQdajqvKtm0sy1rW2E5283FhY6djO7VaV9IN1yzOthSdWkxQ6VSxe/duEolE27G9Qlwe0CtFnT0jKP39/TzwwAPcfPPNqz2dFUdrzenTpzl79mzbD5LjOLUdXSviTLAgCNpe07btSt1Iw6LazKWQSCQ4fmGc589cWfS7IKwW6FUVQisbZTQTcznCSBNFhnhNUVUTxADagFEWCedqenEUXa1fUVzNfkknPLZs6uPS1BxhtceXa1ugFH4QMpBOkky47Bob5EV7t3XcE205qerL6Ya93LGdCBtUFtPVHBsLRSfZg9B52jS0Pnen8bEcx2Hnzp3rplYl3kz3ijuvJ2QvznRary6vRizLYu/evYyMjHD48GHy+fySYxuzq5YizgTrRIDijLpOBKhcLpP2HDQWFgtjMAnPYVNfmvHZfOXMEBPhR2ApG0sZjBVnLy10eyggYYMfgm2ZWmU61fZdhspBVQnPYdfoAHffsIN/feoEQeDj2DblIKRQrixsodYYrTl5aYqhTJItwwMtffn1tTiNPaeaWXzx6xXXirRbTOMEi0bLstm/cWfhxkwraHi9GsY2XquRxiMEWv19Wx3QVj8XpVTN6mjM/mocG78OjR2vm42L5wBXLeJWr3G5XObIkSPs27eP/v7+JccJK0NPCAqsn1qU5dDX18dLXvISzpw5w8WLF9t+qD3P6ziIFy9qrdI6YxdYnKGzlDtsdLCPjDfOTCGsddY1GHZuGuCOvVs5fG6c8dk8GMjP5imWy5RDA2i0qcYxqtaJBShLoY1FGEYYFLZtYStFZDSmmmp86+4xtgxl2TLUB0bjKk2gQOuI+UKxkvFFVZiqcd0zV6YYSFdOBIzfT53U4nTa7Xo5bq1O4mD1YzuxrOKxnSRYXOvYTubQqZsqFpVOXXZxvKaTk1KPHz/Orl27GB4ebnvdtcpSYruW6RlB2UgWSj2WZbFnzx5GRkZ45plnmJuba3ufTs+P6dRdBpUFuFQqLbkA37F3jMNnx5nOl3Bsh5H+FDds24Tv++wdG2Dv2AD5ks+XH51AGU3ShnIEVl0acO3S1diKwmArwEQUQ4VrVQP4gWYuX+D2PZvJFUqcGZ9Fa0255JNIeJg41K4UYWQolANSnrMg3bhUKnV0LECcCJFKpdq+/6IoQmvdUaug2DXUiUup027T9WM7qWqPrbBOx8au0HZjwzBcVi8yY0xHzy1OBGiMcS31njxz5gxhGDI2Nia1KteJnhCUOF12IwpKTDab5aUvfSlnzpzh+PHjLXeVxWIRx3GwLKvlYhV/gDsRoFKpRCKRWPKanmNz294ttZ+bLcAXp+axbQvCyiKTsBUlA6FROFV/Vry0Rzp280AYGRwMQWRhWxWROX5xiq2b+jh9ebbST6tK4Ps4yjATRNjKUA5CLGWRdywObBtZkIZan/rarKivfhGK029bJUHEt8et/pcSqsZCyEZrsRnxterrkdpRP99W14wX9E5qShrbqixVYKm1XnSaaKvrL5U23Wz+8Xuxk8D+yZMnKRaL7N69W0TlOtATggJr71z51SC2VkZHRzl8+DBzc3NtPySduMFi/3u7dh1xqnD9QtXMDx/XVcQFbXHFv8Hg2jaR8QmDSm1JWK0b8ZWFTTU+UtfF2LIUYVgnGBiUsrAVPPLcaQbSSTAQRJXzVXy/8lg6igiMhaUqWWOOcjl7ZYodw4tPATTG1ES1my6w5ViAcTeITtxPnQb24+fV6UasU7ey7/sduwwLhULHY4vFYkfJEHE8sNPEiUuXLhEEAfv375cCyBWmpwSllxtEdpNMJsOdd97JmTNneO6559q6FVzXrRW5tSOdTlMsFpv+rj7TxPMqsYhOFstUKkWxWGRzX4pnwoC+hMNsEGB0hA3V0x81ga60QrGqicGurQjDiPp1qDKDiFDbhEFl8Y2MZjZfwhgIdSVsbwy4qhKPUZZdqaAvVHqBJb2FtQpKqQUusHaiWiqVSKfTbTc4sZB2Yl0vp3VOuVxuWgvUyHKeF1zd+XeSCVg/thXx4t/J5zce24kbsPG6jWLV+PP09DRHjhzhwIEDPZOCC9LLa8XYiEH5Viil2L17N6Ojozz99NNMTEwsOTb+cMa71VaLRaFQqFk1rRa2+EMci0WrN32xWOTSTJ7nz14hijS5ko8fhKi47TuKSOvq+SQGy01gV3t26cpNNSrfWjiWIZtJolEUyiGRuVqjYgBjVPXseAtMRGgqFljCdRfV08TEiQf1BY5LEYYhiUSiFjhtlqUVE0VRbWwz6u8TB547IbaWGosxG5MtYiGJx8a3NyOO7SyVqlv/GPF7oFVmVzy+WCwu2QOsfixctZgb41vNCiAbiyVbMT4+TrFY5NZbb61tiITu0jOC4nnehnd5NSOdTvOyl72Mc+fOcfTo0UUiUL8AxDGDetdWs8Uw/iA3+sob3QXxbj3+XWNmT/x9oRTwxOEzaGOwbJdsUpMvlgmjSrEixqC5ekSJ7/t4rgvGVMYYU/fbCinPwVOaial5ylqBrp6/YeJq/Oq8tY3B4FkwlOnHGE2x2HqnbIwhnU63FV+ouCE7TRhZTmZXp5ZCPNdOPhvTl2aZODNFZCJGdw4zsn3TkmNja6kTq7bT7Lb6mFW718D3/ZZWWL1g149tZ4XlcjkOHTrErbfeum5qVdYSPSMo4vJaGqUUO3fuZGRkhEOHDnHx4sWW440xZDIZSqVS2wWr3lpZKm0YKgtrMpmkWCwSacNMvojn2DiWxfnJWSIdoQAdlomw0Q273wVSZSo7Za0VxrKwWbxIlIKQ4xcmqztk0Fhkki6DmSTFcoApG4JQ40dRxY2mFMPZZK2+ppVVFe+oYwFotVBqrSkWix0t6nEGVicxmLhmCFoLUDzXOJ6w1N/zzHPnOfn06Upqtra4dPoKu2/awb47djcdH0VRx6174rGduKriOpJOuhE0Zrc1dnxovG6cWbZU4kR9jdEzzzzDjTfeKLUqXaZnBCVOWxWWJpVK8fKXv5zz589z6NChJRcCpRSFQqGWMtpqwah3bbXahWutKRQKnJvKc/TcFabm8pVGkAoSCQ9bqeo5JYp8sYhSEBmFUuZq2jCVlvRhVOnxpRSgIwLLrqQX1y8i1XXCGHAsq9KeJTJg2YQEhKYSO0l6Dv2ZJLZt8ey5STYPD+BGEalUasGJiEsRx0AaaeZWSiaTS7pxGjO7Gl0urQoS6xMr/JKP0QYvtfD+9Qtq4y49DENOPX2W0K/GnNCA4fT3z7Hr5u14yavXapx/3LstFtWlsrri91T8PlkqCy6+LQiCBbGleHxjsgdQ6zPXyYay06SJYrHIE088wW233dbTtSprjZ4RFLFQOkMpxY4dOxgZGeH73/8+58+fXzITKwzDpu3CG+MCUHEVeJ63KGtJa83xC5Ncns1TLPnM5kuVXXsIOvTRgJ0vAorhgTSeY1MqlYjCCEdrfGyMjlDGYNsWJqo0j6zv2wUQKQXKwjKLrZUIUJ5DKuUxPzeHthyIKu6iYtEn8EsMZlMAPH/qPDdsrxzJ24mgxq/HclxgnWYkJhKJWhPPdo9vK4dDD36fqcszGG3oH+7jhrv20bdp4cmFzZIAZsfnKJaKi1yWOtJcOTPOyK7WC2r9mTjtnn8+n+843lkoFGqWXeN160Vea025XO4oY80YU8sA66Ru6NChQ9xyyy1s2bKl5VihM3omh06C8ssjmUxy9913c/fdd2NZFqVSiXK5XPvyfb/2lc/na4Fd3/cJgqDSMr76b/wVX6M+xfWZ05c5dmGSXKHETK5I0Q/IlwPCoEyoKhaJAZQF84VyJY6iFJ5jM9iXxqsWK2LblToUA7ZlVSru66wHyxiUjtBW855G5SDEsSxcxybtKgwKoypv7zDSBGHF5VavR2EY1haqVgtlfVC5XcA8ttRSqVTb7JxyudzUWmn2+E/+2/eYujALRqEsxfx0jme/c6SptRC7wGLS/Slsp8lHXUFqMNVRM8VyuVxzP7UjtmoaBSx+PWKr0Lbt2vspjsPFVlb8ffxz/dj45/qveHPgOE4t9hOPjW+Lv1zXrf1r2zZHjhzh9OnTbZ/XaiFZXiuA1KFcG9u3b2d0dJRDhw5x4sSJ2u2NLVdi0WgWC6h3PcSpqKlUimLZ59J0DlU9R8QoC6MUlWxeC2U0EQqDhWPbeJ7NnQd34to2h05cJNCaTY7DXL5ciXtYdvWAMAv0VfeHNqbWv0tHmggLRZ3LiUpjycBYJC0HLIVj2xT9EJSFMhAYhWvZ7NwyXLEMDJSDgITr1GIF9R/cVgKTTCY7ygKLF9X6LKxGmtX2NFIu+UyemwKl0AZsZYFlmJ/KM352ii17RhddPz5ZEUAlFcPbhrl4/NKCv/vIjmFS2Uo8qTH+sVRBoTFmQV3JUhlb9WnQnRyJsJzU5k6bnMbC2kl90bPPPkuhUOCmm27qqQV8rdEzgtJpMZmwGM/zeOlLX8r27dt5+OGHWzabjGscHMdZZBHWB+VLpRLFckC+HKJ0UGv+GJTKtSyxWq9gpQhUiqyrSNmQ9BS37x7m1KVpCmVNqj/BbL5SxDiVCyuiUXWrxe43z3HoS3lMzcd9uhTGdrB1WOvZon0HL5GiWApI2oay9gkjDZaDaxx2Dg3gmJBDR05z5so0JT8k4TnsHB3kwLbhZbmrUqkUvu+3XQDjZIVOXWDxGSP1lPIl/CjAcixQlW7MduSgrYhivtDy2vHO/cBdu3GTNpPnpzDApi0D7L1td22xD4Kg9pzaLdSxq6pdYkNcLd9JHY7v+zWroVMB6qQP2XJqa06dOkW5XOb222+XAshrpGcEJZlMMj8/v9rT6Gm2bdvGm970Jp588kkOHz685LggCAiCoG0mWMJzSFqaQLlgNJ4D6aQHZb9yfknV+rEdm7QDIwMZBvoq55L0pRK1Vi1PHDtPrlT5ACcdi0LZx3ZdHAwJz2bX6BC7xwZ5+tQlgkgzkytijMGKAozlYCmDY1sYBTdsH+H7py5jtGYom8IPIjYPZXnxge30ZzOcvDDOsQuVmh3LUgRhxPGLk5XzU0YGKBQKZLNZisViywUozgJzHKflRkdrTT6fr12zFfGuvnHzlEgn6NuUJV89FAwFESEJO8HmPaMtrxmGIVEUkU6n2fOinex50c6WzylOAGhcqBuFI3YVxvNstauPT/qsD+w3Um/ttKrZqafZwWXNMKbzxqmTk5M8+eST3H777at+WFcn811r9JSgSAzlheO6Li972cvYvXs3jz32GIVCofa7ZqmYceaSMYsP2AK4ac9WvnfyEmChbIt02uHmvVtJODaHz01QLIckEzbDfRlecnA7UKmdqXcZRdgYVbl2NuMQaEW+5OMrRSqZIJlMMl0I8CPIZjIUA8iX/YpVBGhlYZTFyFA/u7aMMLZpgNNXZtDGsGOkn/50quZaujxbxKj6OEzlQ3t5rsjebZXFOW6E2Ni+HhZ/yOPXqNZepiEdOr6tWCwu6qxbLvqcPXyeoBSyacsAY7tH0VrX3Dr1onLwJXv5/jefp5QvgVLYrs3uO7ZXOgHUJUrUJ1NcOTPBmWfPUcyVSKQ99t+xj5EdQ00zqWLiVN1OLLXYUmhXBAssSK1ut1DGWYWdWHXLyQDrtF4ml8uRy+V4+ctfviAWJbSnJw7YAvjIRz7CU089xac+9anVnsq6IQxDnnzySZ5++umWH3JjDH19fUsK+kyuyPnJOYwx7NwyynA2UXN7zRbKOJYim1oY+I0DokfPXuapExeYL5SxLQvPqcY+MGSSHplkAhyXfD5Pppre6gchM7kSkda4jo2lFJmkx1vvfTEmap6JNF8sc+TcBCcvTxFqSHouac+upSwP9aV46Q0Ld+9KKdLp9ALRXYpUKkUQBG0ztmIX2OWzV/j+Q88TBFeFYGznCLe86obaIt/oVovCiO9/83nmp3L0b+pjy74xxnaNLOhYAFDKlzn25EnOHr6Al3Rx3Mq+MQxChseGsD2HzECKnTdvY+ZypXv16K7hRW6eTCZDPp9vawHE4tvJor6UW6+RuFaqXReGmJUQoEwmwyte8Qqy2WzbsStBvOmKkxZ6gZ6yUCQo310cx+GlL30pe/bs4aGHHmJ6errpOKUUuVyOZDKJ67qL/g6D2VQtLRcqH4RMJkMul2Mw03yHFwQB565Mc/TSLJ5jV44E1pr5UoQFeK5DKuFiMJSLBcoRZKs7b891GMomKZQDRgYyjA5k2Ld1mNCv9IGyLGvBghFpzRNHz9cywcq+Ty4MQWVIu5UP6qbs4qppYwz5fL6WMtxJh+d28YLYBXb2mQsEflhrAKCUYvzsBNOXNrNp6yDjZycJygHbD2zFS1Rai3z/289XBEDBzJU5pi/P4hcD9ty6s5oBleTQN77PuSMXKM6X8MsBxZkSqYEUrmszP5mjOFuif7ifK6cmeO7bR8luyqAsi5PfO8Otr76RTVuGavOJ3Z5xO5alEgtimtXhxNeqJ06XbjUmdru1ysCrn09cB9QqWaLeIqufa2NNUUwQBDz88MPceeedq1qr0itiAj0mKBKUXxlGR0f5iZ/4Cb73ve9x5MiRRa3J43/jE/bS6XTTFiv1Lp+4Ir3+tL/GheHizOWK+0E5bBroI1/yyRXLWAoG+zKV+yiDbTtEhQJBVDlb2+gIx3UZ8DzuvX0/Se9q+3etda3PVhzcPX1lmnw5ABSe41CyQspByNx8jsSmfjYPZNi3dVNtjo3PO5fLLdiF1/+ufjGKd9718ZLGMfFrcPnCFXSkUBh0XSeAM4fP8ewjz5OfLaAsi+ceO8K+O3YzsnmEC6cuoaxKRlsxXyIoh4z/8zhXzo+T3ZTh8CPHmLucwygDymC7FRePP+1jOzZRGBL5EQnfZXpiFhNAZEeksyn82TJPfuNpXv5jL1m0gHXq1jLGkM1mO7LoVqJeBzq3VKB1I9R6HnzwQV75yldKrUoH9IygbPTzUFYa27a588472bVrF1//+tcZHx9fNKY+y8t1XVzXbfuBtCyLdDrdNLOskM/jl+P7G1KJFCYsUygHBH7domAMg2mX0C8R+uAlU4R+me3D/egwoBAuzgqKg8ZRFDGfK6LrxmQTNglHEUWau/aMMjyQxXGcls8lrlnJZrMts+Sg4oOPH3upliFu0qWYKwEKx3KITAgGLhy/RH6uQBREKMsikfY49uQpyjeUsS2HMAyYHp8l9CvZcMYYnn/kKFFU2ZlrNCYCZRRahyRSVTHwK1aG4zoEfkgUapQFpgxUPTrF+RIzV2YZ2jy44Pn4vt9RtppSqlbYWGmds7RFF7esyWQybQVoqWSFZhSLxY6zuuJkiXaPH4Yh3/rWt7j77rvZvbt5qxqhggiKsIDh4WHe+ta3cujQIR5//PEld4VBEOD7Pn19fRSLxSUXD601uVyOTCZTK+SLGcymmJovVNN+FUG5EoweHsiSL151MW3qS3Pn/m2cujLNxGwehWHzzjFu3DG2YIFpXLjjWokdW0Y4eXm6srsHMOC5FoODKYYHsrWx2Wy2bc1CuVwmnU7X3ED11Fs2lXoae1GDzZidB7Zz7MkToCpBeks5oDRT52ZqbhvLttBlg2UsJs/NcOnUZcp5v5olXdccMQgxeqFlpY0GHwITYjs2tutUYk3ZDFEQQqgwyqAsUJEFyoBRpNOpWk1Os6LJOKGiHXGMrN3YMAwXuLVaiUC9W6veKm68X9yxOf47xL9vNra+UHepx49ve/zxxykUCtx8882tn3wX6SV3F/SQoIjL6/phWRZ33nknu3fv5qGHHuLy5cuLMpdicrkcSqna7rWxbUt8v/gwsPog967RfsZn5picK9YO1BpJONy5dwwvkeT0hXH60gn60wnCMGDHpiw7h/tqjz03N0d/fz+5XK7l87GBHSMDnLw4QVyO7zo2u0dGF1gl8e4WaPley+fz2LZNKpVqu7uNExpiq8YvB5x59hz52QLGNZgoAEuRGkxy6dh4pb1L3D0gBMe18afKlIMShWKhIhQaLGNj7OpiHWu0afgXMGWNa7ls3j2MXwoplgoEfkCoKlaHlVKUwiJBKcRzXc6fvoh2Km1wliIOlnfShiZuxNmOOFOuE7FaTrKA67odbUQ7saoAnnjiCYrFInfeeWfPLfbXg57J8vriF7/IBz7wAR555JHVnsqGwhjDoUOH+Pa3v9024DkwMEAul2ubMdbf31/pSlzdFU7M5pkrlOjPpBjpvxocjy2BBb3Dqte26uYRp3a223DMFMpM531MFLBrdBDXad7GRSlVW7RaET+XduPiOZZLZR755yeqrq4Kjmtz+w/ewsyVOb734LPkJvOgFdjV52lbKFuR6U8zP5XDVDsIGAy2djCWbuzsv/C5WArP9fCSHgObs+Sm8+ioYpnYjk0inaAwV0RHmuxghoSXxE1b3P6aW0n3pZa8btwuppP29n19fW1FH6h1FnihB8Fd69hOY0UAe/bs4eUvf/mKHtYVn8/TSWuctULPWChxc8hOCpmE7qGU4sUvfjF79+7lG9/4BpcvX246Bqg15at3NTSOM8bUqqfjzrs7q32vmgmR53kkk0nm8kWePnmJyfkitoLRoT5evG8rTt1OOpPJLAqC15NOp9k2QkeumDi43m5cXDQILGnFxbeffe4C5Tkfu9qPzADah/PPXyaZSRAVI3RoMFpDJYcArQxuyiU3USAsa5SqO4iKEGPiEy+boxSUA5/ADynMFekfzhBVYyi2o8n0pUlnUthxanEUEM4rTn3vHHe85pYF12p0K8HVdOnGMfXupfjQtnaV7fFCXu+NaOYKixMlXNdd8pr142dnZ2tFuktdExa2y4+f01LvpaNHj5LP53nNa16zood19dpa1zOCspyzsYXuMzAwwFve8haeeeYZHnjggZbWQGytxE0nW41rtFaW4vHjV5jN5TG6Mm4+N8/83CwvObB9wbhmVk0zYkFr956KU4HbubaUUvT19TE/P7/kIjB+aYKCXyDhJgm1T7yeTY5PsnN4G0W/SGSFaKMxGFRkoSwISwFGx/36Gx8YtFYobIwVLWzRXB2vtEKFCiyYGY+wEpUWLkEQcP5UgcGxARy18PWfuDBZi5O1olAodGyBxBXw7SyQcrlcu2a7fmlxB+x214zjeO3+jnH9VCc9wC5cuMC//Mu/8LrXvY5UamlrbiPRM4IiQfnVRynFHXfcwd69e/nqV7/K8ePHlxw3NzeH67qkUqklXUJKKebn53Fdt6WL6cpMjsnpGSzHwfWShH4JRcVV5ocRXp3rqlAoYFlWrQ5mqQUhfi/19/e3bOkTd12OXVtLufOMMczNzZHJZPB9f5FATl2aYXZijuJcEZM1uE7FvRPpkGQmweiuERzPwRgITIAJDcbSGKOwAwtjt3AjKoMmwNZuRVSazA1tsJRFZEXYxq7dT6Eo5cpkhxYuBZZ7tb9XO3dRvFgvNa4+rdq27Vq8otViXSwWa0kSrfpqxenscVZdK8IwXFBXsxSxhZ1Opxdcs9l9crkc//qv/8q9997LwMBAy8ffCPSMoHR6zKqw8vT39/O2t72NZ599lm9/+9ttC/kymcySbrCYKIrIZrOLWpcAMFfCdq/67C3brezMI4PjeiQTlZ5LjUVpja6oZmIQu+niuTbDmMpRx43NC5e6njFmQaPF5x87xtnnLoAxFOZK5GeL9A1n8TyHZCLJzpu24zg22w5s4fLJcUr5MlEYEZswAQYrqHQYrjdBTEPwJDQRVF1gZsGvDChQloXt2FULCBxsvD6XIFxoiRit2XbD5koD0GKxo8SH+HWMG0K2QinVUfp1sVis1ZV0UoPSyTWh4lKLU7pbEVvQ7Syl+fl5/uEf/oE3vOENjIyMtH389UzPBOWfe+457rzzTq5cudJzfsX1Qnz2RD35fJ6vfOUrPPvssy3v67ouyWSy7cLkeR6JRGLBwhCEEQ8+fZKobsG3bIf+vgwv27+55fvBcRySyWTbhcZxHFKpVNv5deLaiunv7+fi2Ut8918OXT0oLNIU5yt9vXbfsp0dN2xjx77tFAoFxs9N8siXnmB2Yu6qi6sOy9igwKjWcR1lFJZlLxynwPUcokhX0ogdm0TKY3BogOFdgxRzZXIzebyEy5a9Y+y+dceC55dOp/F9n3LJ5+zz58nPFkikPXbeuI1kXTcEx3HwPK+tVVOfxNGOOFjfyUFonYhfPM9mHbWb0Wmxpuu6vP71r2fHjh1tx3ZCFEW117NX6BkLJXZ5SVB+bZHJZHjb297Grbfeype//OUlP8yxP35goNLRd6kdp+/7lMvl2jitNZ7rcMOOUQ6fm0BXjwK2LcWBsX42bdq0ILMsfm/U/+v7PsPDw7VU16W63UZRxNjYWG3xaHatmC1btixo9b7k2KJFys1UYiLV2/pSfWDgB974KtyEw8yVWcKCYfrUHCMjowQzIUGpfvdcN18NSlksDqbUj1YYXTnl0VSr8C3Hoi+dpVz0cVwb13MJSyGUbO6468VsOzjG+LlJinNFtt+wDS/hLnouOjT86198g7npAkopClMBpyYu8oP3v4qhsavuHqXUorPdm73eQNM6nWb1Pa1+1/h9/NitHrfZ9VrdFnd8aDc+l8sRRdGijde10mtrXc8ISuyWiDOEhLXFTTfdxIEDB5q2KF/qg9rphzlmar7Ad4+ew7FsXnHzTlKJtb9z+943nsWc+NKiOEAqk+Dtb3s7f/MH/8jUc2XKpTK55yO27dzF4JYxLhy7hImuwXlQ9YgppTCYq7pjgR3aWGEKExrSQxk27xpheNsQF78+xfQTBfyCD5Yi9/2zvPZnXs0rfuyuBZd+4P/7JoN6jKHBzQsf8mKC+95x3/LnKqw7ek5Q6ovPhLVF7EZYKTb1pXn9S25YseuvBC+65yYe+ptHmLgwVbtNR5oX3XMzX/nkA5w5fAFlK0ykiSLN+JlJskPpSu8tomWLirKo/Q2iMEJHlfgJGqLSVauwNF/k0onLeCmP+al5olCz86ZKxpxfCvjq//w6B16yl5Ftm2r3GT872XTHPHFuatFtwsakZ44l66SCWVhZeiTctqawLIt3/M7buOUVB8kOpNm0ZYB7f/IV/Ogvv5aTT5+tLdCpvhRuwsUA89NFMv1pUpkklm1hOxaWbeEmXQZG+hjePsSe23eSGaieFR93LLYVfQNZEmmvmtxQnUSTP5sxEPghk+cnKcyViIJo0e+f/vrCuNjASF/T98DAyOq0d1/vNFrxvUDPWSjSwn51iLNi4jTNpVxTvfYBuB4MjPTz9ve9ecFti3z8SjG2c5iLJ69UAu8RJNIejmujbAsTabykx+4X7eTSiSug4aZX3MDkhWmmLk1jtMaybUxkKM6WqsF3i6hJcB9VKXg02lAqlHFch0RqofvQGIOyF/4tX/kTd/P0N5+nMHc14O64Nq94090v/EUS1gU9IyhKKTlXfpWIeyxZlrWg2R4sLSb1MQMRnMUopdj/4t08/c3na69JdlOWg0MZDty1l8e//BR9w31MnJ1kZqJyCJbBcPTx4xTzJbbsHcOyFKM7h0EZLh67jLIiwKAchUJVOgorqDcqlKVqpzwqy8JLJhgY6SM7lFkwPzfhcNd9dyy4bWCkn//wuz/Fg5/7DhPnJukf6ecVP/4S9t2xZyVfKqGH6BlBASluvN7EZ1AYY2rngTQGl5udixIfjhTTmKUTWznxz/XjNhI//quvpzBf4sShU4RhxODoAK//uXu5/TW3cNsP3Mw3/+4RbKfi6ho/M0FQCtHa4CU95iYqZ7RYrs3lk+PoSJPOJgn9sNL6XlUq422nkvEUu7Qc1ybdlyLwI1J9CV75E3fzU+9/M//yyQc48t0TlVMdt2/i9T/3g/QNLXZlje0c4Sd/403X+6XakPSii7ln6lCMMYyNjfH5z3+eu+66q/0dhBdEfFAVVIK88Xncy2WpXkz11IvNUtlfjd+vJ6YvzzI3Nc+Og1uxGxpWRmHE4/9yiH/8f78GxjA/Oc/4+alqam7F9ViYLxIGEen+FBjITecqYRMFrueiowjLtukf7WPnwa0U5ots27+FH/zpV3HX62+vPdb8dI78bIGxXSPX9LcWuksURXiet6KJLt2md2aKVMtfL2IxiesJXNe95mstVT8Ai0Wm3rJpVW+w3txpQ5sHGNp8tY5j8uI03/r7R5m6OM3Q5kGUUrhe5aM6uHmQmfE5wiAiiiqt7B3XvvrcFXgpj3KhjOu5JNMJHNdmy94xhrcN8et/9p9qZ8w30jeUbWqVCKtDL76fe0pQ5EyUladeTFY6DbidGCzlTqtvz9Jo3fS6O23y4jSf+u2/JF8NfJ9+9jwYKBXLJNMJLNti+w3bmDg3hWUplIJEupIBOXVxmijUuEmX4W2b8NIuruvQt6kiEne85pYlxURYm/TK+zamZ95dSqmOz6AWro16MYlP3FtNOhWcepFZysqpr3ju9Pqrwbf//rGamNRQ0L8pWzmt0VZ4SZf9L97N/f/HW5i6NMOX/virRNowONpPbjrPja84yDv+27/jq5/6OmeeP4/rOtz8qhu59+2vWJ0nJWwYekZQAMnyWkHqxcTzvJ7woS/XnRaLTWzJNLv/agvO5MXpprdv2TPGD/37H+DIY8dIZJK89EfvIJVNsfuWHQyO9vPU/36acjFgz207edkb78SyLH7iPW+4LnMWuk/TJqk9QE8JSnxGgdBdoigiiqKeEpN2rCV3Wrno88g/VZo+bj+whTt+6EVLHrE7tGWQ5x89ytz4PNoY0v1p+oezDG0eYPctO9h9y+LGg3tv28Xe23a1nMNyicJKVlhjkoAgtKKnBEXShrtPLCaWZdW6um4EWgnCUtZNvYXT7FrNrJupSzP8xYc+x+zkPAp48PPf4aG/fYRf+9gvNo1nDI71c/HYZcLqgj4/MU/oB7zqrS99YU+4Q+Ym5/nHj3+VU8+cRVmKfbft5k2/dh+ZuqOZBWEpekpQksmkCEoXqS9Y3Ehi0o7lWjet3GkPfPYh5ibnyc/kuXJmgqAccv7IRaavzPCej/9HhrcMLbjWoQeeYedN25i6OINfDnATLn2DGUI/ZPLCNIm0R3Yws2hO3eJz//cXOX/s6jHPzz92jNIflPmF3/vpFXtMoTnSemWFkbTh7hCfw621XrJgUVia5QjOxRNXiCLNpWrxoVKVvicXj1/mC3/4ZX7+w2+vvfb5mQITFyZxEy6b94zGD0B+tsDH//OnUJbCcR32v3gPb/uNHydVdw5JNzh75AJnn7+AVefmUkpx6ukzTF6cZnjrUIt7C0IPNYcEcXl1gzj4LmKyctQXaGYH08xP5dDRwkOxbMfmzLPnKM4Xax0JnKRNuj+1wOqJwogLxy5WzzGp7P+OPXmSL/7Rv7zgoO2hf3uGv/r9L/CX/+ff88S/Ps3c+FylyVcDkdbkptufhCgIPWWhiMvrhdHNgkWhM+667w6efvD5BbdZtqoWLFooy6oJum3b3PlDt/Gtv38Uqxq0n7kyi5f0akITc+yJE5RLleLFa0kW+Nqnv8G3vvAYyqqMOfzYMe7+kTvIDKQp5hYmvgyNDbLjxm0v4FUQNgo9tTVNJBKSNnyNNBYsiphcH150z03c/9tvITOQxnYt0n1Jth3Ygpd02X3rdtJ9qQXjf/jn7uH1/+E1bNu/mZEdw+y9fRfbD25Z1OFZhxodXXVdhmFIGIYEQVD7PgxDoiha4OI0xlCYL/LYV56siQmAsiyeeuAZfuDfvRyn2v/LGIOXcPiRX3zNkllpglBPT1ko4vK6NtZaweJG4+4fuYP+4T6+/In/zfTlWZRSbNk7xo+/a/Eph0opXvWWu3nVWyot4eenc/zRr/15rbljLCq7btlOIrXwoLlOW9mcPXyeYq5USwmOfxf4AX2bMrznj3+Zp/7397Esiztff5tkeAkd01OCkkwmmZ+fX+1p9BQiJmuDG+7ex/4793Dm2XMkMwm27tvc/k5U+mvd9wuv4V//4kH8ko/WmrFdo7zxna9bNLbTZIEte8fwkl6t1iS+3bIttuwbJdWX5FVvfWktY61V5wFhZem117mnBEUq5ZfHeixY7GVs27qmAsS77rudF91zE89863myQ1kOvmTvNS00tXNXBjPc/ppbeeJrh2rvicqxxDcxtHlwyVY29dfolVY2wvWlpwQllUqJy6tDNmrB4nolkfJ4yetubz+wQ37snT/M8LZBjjx2HID9d+7l1VWrZD21shGuLz0lKFKH0hlSsCi0QynFK998N698c2fH966lVjYbgV7s4wU9JiiS5dUaKVgUVovr1cqm2fWFtUNPCYo0h1ya+hMWRUyEtUQ3W9mIdbO26SlBEQulOVKwKPQy1yo4zdxpwCKh6VXBkV5eK4wE5RdzPU9YFITVYDnuNKBWwNnqvuJOWxl6avWRoPxCpMZE2Ohca7JAYyp0vTUj7rRrp6cERSrlryJiIgjtWc4x0vG/rWpvGt1QKyk4vShgPSUoscurPki3EZGCRUHoDss9RrpZ14DGnzeyO62nBEUsFClYFITrhdTeLJ+eE5SNnOUlBYuCsHZYjjttua1sZmZm8H2f3bt3r9j8V4KeEpSNeh6KFCwKQu+xXHdaLDZaa375l3+ZoaEhPvvZz163+XaDnhKUjejyik/zM8aImAjCOmEp68YYw+///u/z/PPP88gjj6zG1F4QPSUoyWSSIAhqO/X1jhyKJQgbi3/8x3/kYx/7GA899BCjo6OrPZ1l01Nb3WQyCbAh2q+ImAjCxuK5557jne98J3/6p3/Ki1/84tWezjXRU4KSSFROqFvvbq/GGhOpfheE9c3MzAz3338/73znO7n//vtXezrXjDI91Ce5VCqRSqU4fPgwW7duXe3prAhSsCgIG4soivipn/optNZ86Utf6ukNZE/N3HVdlFLr1kKRgkVB2FgYY/joRz/KkSNHePTRR3taTKDHBEUpRTKZXJe1KFKwKAgbjy9+8Yt8/OMf56GHHmJ4eHi1p/OC6TlB8Txv3QXlpWBREDYezz77LP/pP/0nPvnJT3LHHXes9nS6Qk8JCqy/WpRYTKTGRBA2DtPT09x///28613v4u1vf/tqT6dr9KSgrAcLRQoWBWFjEoYhv/RLv8SBAwf4yEc+sm76eEGPCYpSal3085IaE0HYmBhj+L3f+z1OnDjBI4880vNB+EZ67tn0ustLjusVhI3LF77wBf7kT/6Eb33rW2zatGm1p9N1RFCuI1JjIggbl2eeeYZf/dVf5c///M+57bbbVns6K0LPOe17VVBETARh4zI1NcX999/Pu9/9bt72tret9nRWjJ6zUJLJZM8F5aVgURA2LmEY8ou/+IvceOON/N7v/d66CsI30nOC4nleTwXlpWBREDYuxhg+/OEPc/r0aR555JF175noOUHpJZeXFCwKwsbm7/7u7/jkJz/Jt7/9bYaGhlZ7OitOzwlKr5zaKAWLgrCxefrpp3nXu97Fpz/9aW699dbVns51oecEZa1bKFKwKAjC5OQk999/P+9973t561vfutrTuW70nKCsZQslzuQCpGBREDYoYRjyC7/wC9x66618+MMfXtdB+EZ6TlA8z1uTgiIFi4IgGGP4nd/5Hc6dO7chgvCN9JygJJNJpqenV3saC5AaE0EQAP7mb/6GT33qU3z7299mcHBwtadz3ek5QVlrMRQRE0EQAA4dOsS73/1u/uIv/oJbbrlltaezKvScoKylGIoULAqCADAxMcHP/MzP8Ou//uu85S1vWe3prBo9JyhrpduwFCwKggAQBAG/8Au/wO23386HPvShDRWEb6TnBGUtWChSsCgIAlSC8B/60Ie4ePEiDz/88IZ3efecoKx2lpcULAqCEPO5z32OT3/603znO99hYGBgtaez6vScoKyWhSIFi4Ig1PPUU0/xnve8h89+9rPcfPPNqz2dNYEISgdIwaIgCPXEQfj3ve99vPnNb17t6awZek5QrndQXgoWBUGoJwgCfv7nf54777yT//bf/tuGDsI30pOCcr3OQ5EaE0EQ6jHG8MEPfpArV67wxS9+UdzeDfScoCSTSXzfxxizojsDERNBEBr5q7/6Kz7zmc/w8MMPSxC+CT0pKCsdQ5GCRUEQGnnyySd573vfy1/91V9x4403rvZ01iQ9JygrnTYsBYuCIDQyPj7Oz/7sz/Jbv/Vb/NiP/dhqT2fN0nOrZezyWglETARBaCQIAn7u536Ou+66i//6X/+rBOFb0HMWSjKZXJGgvBQsCoLQiDGG3/7t32ZycpJ/+qd/knWhDT0nKIlEohYw70YKrxQsCoKwFJ/97Gf5y7/8Sx5++GH6+/tXezprnp4TlFQqBUCxWHzBgtJYsGjbtoiJIAgAPPHEE/z6r/86f/3Xf80NN9yw2tPpCXpOUBKJBMALjqNIwaIgCEtx+fJlfuZnfoYPfOADvPGNb1zt6fQMyhhjVnsSyyEIAjzP47nnnmP79u3XdA2pMREEYSl83+dNb3oTmzdv5nOf+5x4LZZBz1kocZzjWlOHRUwEQVgKYwwf+MAHmJmZ4Z//+Z9FTJZJzwkKXPsxwFKwKAhCKz7zmc/w13/91zzyyCP09fWt9nR6jp4TFKXUNQmKiIkgCK14/PHH+Y3f+A0+//nPc/DgwdWeTk/Sc4ICy28QKQWLgiC04tKlS/zsz/4sH/zgB/nRH/3R1Z5Oz9KTguJ5XsdZXlKwKAhCK3zf5+d//ud59atfzfvf/36phH8B9JygKKU6qpaXgkVBENphjOG3fuu3mJub4ytf+YqsES+QnhMUaN8gUgoWBUHohP/1v/4Xn//853n00UfJZrOrPZ2epycFpVWDSClYFAShEx599FF+8zd/k7/927/lwIEDqz2ddUFPCspSQfl6MXEcB8fpyacnCMIKc+nSJf79v//3fOhDH+K+++5b7emsG3pyxW3m8pKCRUEQOqFcLvOOd7yDe++9l9/8zd+UIHwX6UlBaTy1sV5MpMZEEISlMMbw/ve/n3w+zyc+8QlZK7pMTwpKfWGjFCwKgtAp//N//k/+/u//XoLwK0TPCorv+1KwKAhCxzz88MO8//3v54tf/CL79u1b7emsS3pyBU4mkxSLRRETQRA64tKlS7zjHe/gwx/+MD/8wz+82tNZt/Rc+3qA++67j/n5eX7lV36FG264gYMHD5LNZiW4JgjCIsrlMm984xvZu3cvn/nMZ2TzuYL0pKC8733v4+GHH6ZUKnH06FFmZ2fZunUrBw4cYN++fRw8eJADBw6wf/9+9u3bRyKRELERhA2IMYb//J//M0888QTf/OY3yWQyqz2ldU1PCko9WmuuXLnCkSNHOHr0KIcPH+bYsWMcPXqU48ePUy6X2bVrF/v37+fAgQMcOHCAgwcPsn//fnbt2oVt2yI2grBO+bM/+zN+93d/l0cffZS9e/eu9nTWPT0vKEthjCGKIs6dO1cTmiNHjnDs2DGOHTvGyZMnsSyLffv2sX//fvbv31+zbA4cOMCWLVtQSonYCEKP8p3vfIc3v/nN/MM//IPETa4T61ZQWmGMwfd9Tp06VROao0ePcvToUY4dO8bZs2fJZrPs27evJjD11s3Q0JAIjSCsYS5cuMA999zD+973Pv7Lf/kv8nm9TmxIQWmFMYZCocCxY8c4fPhwTWjir/HxcYaHhxe40GLrZv/+/WQyGXnzCsIqUiqVeMMb3sANN9zApz/9aQnCX0dEUJaBMYaZmZmaVRO70GLLZm5uju3bty9wocXCs3fvXjzPE7ERhBXEGMO73/1uvve97/HNb36TVCq12lPaUIigdAFjDMYYLl++XLNq6t1ox48fJwxDdu3atciqOXDgADt37sSyLBEbQXiBfOITn+CjH/0ojz76KHv27Fnt6Ww4RFBWmDg54OzZsxw5cmSBG+3YsWOcOnUKx3FqyQGN8ZqxsTFJDhCEDvjWt77FW97yFv7pn/6J1772tas9nQ2JCMoqEicHnDhxYkEWWvzv+fPn6e/vX5AcUG/ZDA4OitAIAnD+/HnuuecePvCBD/De9753taezYRFBWaMYY8jn8wtiNbEb7dixY0xMTDAyMrJAaOprbFKplIiNsCEolUr8yI/8CLfccguf+tSnJAi/ioig9CDGGKamphalPMfxmvn5eXbs2FETmnqrZs+ePbiuK2IjrAu01vzar/0azz77LA8++KAE4VcZEZR1RJwccPHixZplUy84J0+eJIoidu/e3dSFtn37dkkOEHqKP/mTP+H3f//3eeyxx9i9e/dqT2fDI4KyQTDGEIYhZ86cWSQ2x48f59SpU3ieV7NoGjsHjI6OSnKAsKb45je/yVvf+la+9KUv8ZrXvGa1pyMggiJQEZtyuczx48cXWTXHjh3jwoULDAwM1OI09fGaAwcO0N/fL0IjXFfOnTvHPffcwwc/+EHe8573rPZ0hCoiKEJLjDHMz88vEJn6gs6pqSnGxsYWCE0sNvv27SOZTIrYCF2lWCxy3333cfvtt/Nnf/ZnEoRfQ4igCNeMMYaJiYlFLrS4AWehUGDnzp01F1p9FtqePXtwHEfERlgWWmt+9Vd/lcOHD/Pggw+STCZXe0pCHSIoQteJkwPOnz+/qNPz0aNHOXHiBAB79uxpmvK8fft2idcITfnjP/5j/uAP/oBHH32UXbt2rfZ0hAZEUITrSpwccOrUqQWWTWzVnD59mlQqVTscrb4f2sGDBxkeHhYXxwblwQcf5G1vextf/vKXuffee1d7OkITRFCENYMxhlKpxPHjxxfU2MSWzaVLlxgcHFwUr4m/+vr6xKpZp5w5c4Z77rmH3/md3+Hd7373ak9HWAIRFKEnMMYwNze3KAst/pqZmWHLli1NU57lGOjeplAocN9993HnnXfyyU9+Uv6OaxgRFKHn0VozPj6+yIUW19iUSiV27drV9MC03bt3yzHQaxitNe985zs5fvw4X//61yUIv8YRQRHWLXFywLlz5xZ1eo47B1iWtSA5oN6y2bp1qyQHrCLGGD7+8Y/zP/7H/+Cxxx5jx44dqz0loQ0iKMKGxBhDEAQtj4FOp9NNOz0fPHiQTZs2idCsMN/4xjf4yZ/8Sb7yla9wzz33rPZ0hA4QQRGEBowxFIvFJY+BvnLlCps2bWp6DPSBAwfkGOgucPr0ae655x5+93d/l3e9612rPR2hQ0RQBGEZGGOYnZ1t2un52LFjzM7OsnXr1qaZaPv27ZNjoDugUCjw+te/nrvvvps//dM/lderhxBBEYQuobXmypUrTY+BPnHiBOVyueUx0JIcUHkN/+N//I+cOXOGf/u3fyORSKz2lIRlIIIiCCtMfAx0nBzQ2Kbm1KlTWJZVi9c09kTbvHnzhkgOMMbwsY99jD/8wz/k8ccfZ9u2bas9JWGZiKAIwioSHwN98uTJBfGaI0eOcPz4cc6ePUs2m12y0/PQ0NC6EZqvf/3rvP3tb+erX/0qr371q1d7OsI1IIIiCGuU+Bjo+g7P9ZbN+Pg4w8PDSx4DnU6ne0ZsTp06xb333stHP/pR3vnOd672dIRrRARFEHoQYwzT09NN4zXHjx9nbm6O7du315IB6utr9u7du6aOgc7n87zuda/jla98JX/8x3+8ZuYlLB8RFEFYR8TFnJcuXaqJTCw6x44d4/jx44RhWDsGutGq2blz53U9BlprzS/90i9x/vx5HnjgAQnC9zgiKIKwQYiTA86cObPgWIH6Y6Bd12Xfvn1Ne6KNjY11NTnAGMMf/uEf8kd/9Ec8/vjjbN26tSvXFVYPERRBEGrHQJ84caLpMdDnz5+nv79/wWFp9ckBAwMDyxaaBx54gJ/+6Z/ma1/7Gq961atW6JkJ1xMRFEEQWmKMIZfLLYjV1DfgnJycZHR0tGkx5/79+0mlUovE5uTJk9x777389//+3/mVX/mVVXpmQrcRQREE4ZoxxjA1NVVzn9ULzbFjx8jlcouOgd6xYwcf+chHeO1rX8vHP/5xCcKvI0RQBEHoOnFywIULFxYdA/3d736XcrnMhQsX8DxvtacqdBERFEEQritxvEbONll/iKAIgiAIXcFa7QkIgiAI6wMRFEEQBKEriKAIgiAIXUEERRAEQegKIiiCIAhCVxBBEQRBELqCCIogCILQFURQBEEQhK4ggiIIgiB0BREUQRAEoSuIoAiCIAhdQQRFEARB6AoiKIIgCEJXEEERBEEQuoIIiiAIgtAVRFAEQRCEriCCIgiCIHQFERRBEAShK4igCIIgCF1BBEUQBEHoCiIogiAIQlcQQREEQRC6ggiKIAiC0BVEUARBEISuIIIiCIIgdAURFEEQBKEriKAIgiAIXUEERRAEQegKIiiCIAhCVxBBEQRBELqCCIogCILQFURQBEEQhK4ggiIIgiB0BREUQRAEoSuIoAiCIAhdQQRFEARB6AoiKIIgCEJXEEERBEEQuoIIiiAIgtAVRFAEQRCEriCCIgiCIHQFERRBEAShK4igCIIgCF1BBEUQBEHoCiIogiAIQlcQQREEQRC6ggiKIAiC0BVEUARBEISuIIIiCIIgdAURFEEQBKEriKAIgiAIXUEERRAEQegKIiiCIAhCVxBBEQRBELqCCIogCILQFf5/k1KffIaaJ44AAAAASUVORK5CYII=", "text/plain": [ "<Figure size 500x500 with 1 Axes>" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Plot the hyperplane spanned by the first two principal components\n", "plot_pca_plane(H, clust_id, v1, v2, interactive=False)" ] }, { "cell_type": "code", "execution_count": 31, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "<Figure size 400x400 with 1 Axes>" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Plot the 2D projection\n", "plt.figure(figsize=(4, 4))\n", "plt.scatter(H_pca[:, 0], H_pca[:, 1], c=clust_id, cmap='viridis', alpha=0.7)\n", "plt.xticks([], []), plt.yticks([], [])\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- We can further reduce the number of dimensions. \n", "- In this case, we can go down to 1 dimension.\n", "- It boils down to projecting the data into the direction of maximum variation." ] }, { "cell_type": "code", "execution_count": 32, "metadata": { "slideshow": { "slide_type": "-" } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "<Figure size 400x150 with 1 Axes>" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "H_pca_1d = PCA(n_components=1).fit_transform(H)\n", "\n", "# Plot the 1D projection\n", "plt.figure(figsize=(4, 1.5))\n", "plt.scatter(H_pca_1d, np.zeros_like(H_pca_1d), c=clust_id, cmap='viridis', alpha=0.7)\n", "plt.xticks([], []), plt.yticks([], [])\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- 💡 Check [this blog](https://setosa.io/ev/principal-component-analysis/) for an excellent visual explanation and interactive examples about PCA." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "#### Reducing the dimensionality of the Reservoir states\n", "\n", "- Let's go back to our Reservoir states, which in our case are very high-dimensional vectors. \n", "- In particular, since `n_internal_units=900`, we end up with a sequence of length `T` of vectors with size `900`.\n", "- We will use PCA to reduce the dimension of the reservoir states to `75`." ] }, { "cell_type": "code", "execution_count": 33, "metadata": { "slideshow": { "slide_type": "fragment" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "states_tr shape: (2666, 75)\n", "states_te shape: (266, 75)\n" ] } ], "source": [ "pca = PCA(n_components=75)\n", "states_tr_pca = pca.fit_transform(states_tr[0])\n", "states_te_pca = pca.transform(states_te[0])\n", "print(f\"states_tr shape: {states_tr_pca.shape}\\nstates_te shape: {states_te_pca.shape}\")" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Below, we fit the readout on the new data and compute the results on the test set." ] }, { "cell_type": "code", "execution_count": 34, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Training time: 0.0028s\n", "Test time: 0.0001s\n", "Mean Squared Error: 22.55\n" ] } ], "source": [ "# Fit the ridge regression model\n", "ridge = Ridge(alpha=1.0) \n", "time_start = time.time()\n", "ridge.fit(states_tr_pca, Ytr[n_drop:,:])\n", "print(f\"Training time: {time.time()-time_start:.4f}s\")\n", "\n", "# Compute the predictions\n", "time_start = time.time()\n", "Yhat_pca = ridge.predict(states_te_pca)\n", "print(f\"Test time: {time.time()-time_start:.4f}s\")\n", "\n", "# Compute the mean squared error\n", "mse = mean_squared_error(scaler.inverse_transform(Yhat_pca.reshape(-1, 1)), Yte[n_drop:,:])\n", "print(f\"Mean Squared Error: {mse:.2f}\")" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Even with a drastic reduction of the dimensionality, the performance remained more or less unchanged.\n", "- On the other hand, we reduce the computation time both in training and in test.\n", "- This reduction in computing time is particularly significant when using more sophisticated readouts, like the one we will try next." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "### Fit a GBRT readout\n", "\n", "- Mapping the Reservoir states to the desired output is a standard regression problem, which can be solved by one of the many standard regression models in [scikit-learn](https://scikit-learn.org/stable/supervised_learning.html).\n", "- For example, we can use a Gradient Boost Regression Tree (GBRT), which gives us predictions for different quantiles.\n", "- In this way, we can compute confidence intervals in our predictions.\n", "- This is a very simple way to implement probabilistic forecasting." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- In the following, we will fit a different model for the 0.5, 0.05 and 0.95 quantiles.\n", "- The 0.5 quantile will give us the most likely prediction for the future values.\n", "- The 0.05 and 0.95 quantiles together will us a 90\\% confidence interval for our prediction." ] }, { "cell_type": "code", "execution_count": 35, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Training time: 11.51s\n" ] } ], "source": [ "time_start = time.time()\n", "\n", "# Quantile 0.5\n", "max_iter = 100\n", "gbrt_median = HistGradientBoostingRegressor(\n", " loss=\"quantile\", quantile=0.5, max_iter=max_iter)\n", "gbrt_median.fit(states_tr[0], Ytr[n_drop:,0])\n", "median_predictions = gbrt_median.predict(states_te[0])\n", "\n", "# Quantile 0.05\n", "gbrt_percentile_5 = HistGradientBoostingRegressor(\n", " loss=\"quantile\", quantile=0.05, max_iter=max_iter)\n", "gbrt_percentile_5.fit(states_tr[0], Ytr[n_drop:,0])\n", "percentile_5_predictions = gbrt_percentile_5.predict(states_te[0])\n", "\n", "# Quantile 0.95\n", "gbrt_percentile_95 = HistGradientBoostingRegressor(\n", " loss=\"quantile\", quantile=0.95, max_iter=max_iter)\n", "gbrt_percentile_95.fit(states_tr[0], Ytr[n_drop:,0])\n", "percentile_95_predictions = gbrt_percentile_95.predict(states_te[0])\n", "\n", "print(f\"Training time: {time.time()-time_start:.2f}s\")" ] }, { "cell_type": "code", "execution_count": 40, "metadata": { "slideshow": { "slide_type": "subslide" } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "<Figure size 1400x400 with 1 Axes>" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Plot the results\n", "fig = plt.figure(figsize=(14,4))\n", "plt.plot(Yte[n_drop:,:], 'k--', label=\"True\", linewidth=2)\n", "plt.plot(scaler.inverse_transform(median_predictions[:,None]), label=\"Median prediction\", color=\"tab:blue\")\n", "plt.fill_between(np.arange(len(Yte[n_drop:,:])), scaler.inverse_transform(percentile_5_predictions[:,None]).ravel(), scaler.inverse_transform(percentile_95_predictions[:,None]).ravel(), alpha=0.3, label=\"90% CI\", color=\"tab:blue\")\n", "plt.grid()\n", "plt.legend()\n", "plt.title(\"Predicted electricity load using Gradient Boosting Regression Trees\")\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "subslide" } }, "source": [ "- Finally, we repeat the training on the states reduced with PCA." ] }, { "cell_type": "code", "execution_count": 41, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Training time: 5.80s\n" ] } ], "source": [ "time_start = time.time()\n", "\n", "# Quantile 0.5\n", "max_iter = 100\n", "gbrt_median = HistGradientBoostingRegressor(\n", " loss=\"quantile\", quantile=0.5, max_iter=max_iter)\n", "gbrt_median.fit(states_tr_pca, Ytr[n_drop:,0])\n", "median_predictions = gbrt_median.predict(states_te_pca)\n", "\n", "# Quantile 0.05\n", "gbrt_percentile_5 = HistGradientBoostingRegressor(\n", " loss=\"quantile\", quantile=0.05, max_iter=max_iter)\n", "gbrt_percentile_5.fit(states_tr_pca, Ytr[n_drop:,0])\n", "percentile_5_predictions = gbrt_percentile_5.predict(states_te_pca)\n", "\n", "# Quantile 0.95\n", "gbrt_percentile_95 = HistGradientBoostingRegressor(\n", " loss=\"quantile\", quantile=0.95, max_iter=max_iter)\n", "gbrt_percentile_95.fit(states_tr_pca, Ytr[n_drop:,0])\n", "percentile_95_predictions = gbrt_percentile_95.predict(states_te_pca)\n", "\n", "print(f\"Training time: {time.time()-time_start:.2f}s\")" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "-" } }, "source": [ "- The improvement in terms of computing time now is significant." ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## Summary\n", "\n", "In this lecture we saw:\n", "\n", "- Neural Networks, the MLP, and how it can be used as a windowed model to perform time series forecasting.\n", "- The limitation of windowed approaches and how they can be solved by RNNs.\n", "- The ESN, a randomized RNN from the family of Reservoir Computing approaches that is fast and easy to train.\n", "- Examples of forecasting with MLP and ESN on real-world electricity data.\n", "\n", "---" ] }, { "cell_type": "markdown", "metadata": { "slideshow": { "slide_type": "slide" } }, "source": [ "## Exercises\n", "\n", "### Exercise 1\n", "\n", "- In the MLP example, change the `window_size` ($P$) from 12 to 24. \n", "- Comment on the results and explain why the performance improve. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Exercise 2\n", "- In the MLP, change `forecast_horizon` from 24 to 36.\n", "- Do the same change also in the ESN.\n", "- With a forecast horizon of size 24 we used as a baseline the MSE between the time series and its shifted version by 24 lags, `mse = mean_squared_error(y_test[24:], y_test[:-24])`. Is this a good baseline also when the forecast horizon changes? What could be a better baseline?\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Exercise 3\n", "\n", "Contrarily from model-based approach such as ARIMA, the MLP and the ESN do not strctly require the data to be stationary. However, by removing trend and seasonality the MLP and the ESN can focus only on predicting the residuals.\n", "\n", "- Remove trend and seasonality and train both MLP and ESN to train the residuals. \n", "- In testing, remember to put back the trend and seasonality when computing the predictions.\n", "- Compare the results with those obtained by training the MLP and the ESN on the original data." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Exercise 4\n", "\n", "For this exercise, we will consider a new dataset consisting of hourly water temperatures from Gulf of Mexico near Key West, Florida. \n", "The hourly temperatures are provided from October 3, 2016 to October 3, 2017\n", "The dataset consists of a data frame with 6572 observations on the following 3 variables.\n", "\n", "- `DateTime` Date and time of reading (format mm/dd/yyyy h:00).\n", "- `WaterTemp`: Water temperature (in degrees Fahrenheit).\n", "- `t`: Time index (1 to 673)." ] }, { "cell_type": "code", "execution_count": 38, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Raw data:\n", "------------------\n", " DateTime WaterTemp t\n", "0 10/3/2016 0:00 86.2 1\n", "1 10/3/2016 1:00 86.2 2\n", "2 10/3/2016 2:00 86.2 3\n", "3 10/3/2016 3:00 86.2 4\n", "4 10/3/2016 4:00 86.0 5\n" ] }, { "data": { "image/png": "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", "text/plain": [ "<Figure size 1400x400 with 1 Axes>" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "temps = sm.datasets.get_rdataset(\"KeyWestWater\", \"Stat2Data\").data\n", "print(\"Raw data:\\n------------------\\n\", temps.head())\n", "plt.figure(figsize=(14,4))\n", "plt.plot(temps[\"WaterTemp\"])\n", "plt.grid()\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "- Apply the MLP on the temperature data.\n", "- Perform a small grid search to try different values for the following hyperparameters:\n", " - `hidden_layer_sizes` (e.g., try more layers `[16, 16, 8]`, layers with more units `[32, 16]`, etc...).\n", " - `learning_rate_init`.\n", " - `activation` (try at least `tanh` and `relu`).\n", "- Apply the ESN on the temperature data.\n", "- Perform a small grid search to try different values for the following hyperparameters:\n", " - `spectral_radius`.\n", " - `input_scaling`.\n", " - `n_internal_units`.\n", " - `connectivity`.\n", "\n", "Report the configuration of MLP and ESN that achieves the best performance on the test data." ] } ], "metadata": { "celltoolbar": "Slideshow", "kernelspec": { "display_name": "sta2003_nb", "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": "3.10.16" }, "toc": { "base_numbering": 1, "nav_menu": {}, "number_sections": true, "sideBar": true, "skip_h1_title": false, "title_cell": "Table of Contents", "title_sidebar": "Contents", "toc_cell": false, "toc_position": {}, "toc_section_display": true, "toc_window_display": false }, "varInspector": { "cols": { "lenName": 16, "lenType": 16, "lenVar": 40 }, "kernels_config": { "python": { "delete_cmd_postfix": "", "delete_cmd_prefix": "del ", "library": "var_list.py", "varRefreshCmd": "print(var_dic_list())" }, "r": { "delete_cmd_postfix": ") ", "delete_cmd_prefix": "rm(", "library": "var_list.r", "varRefreshCmd": "cat(var_dic_list()) " } }, "types_to_exclude": [ "module", "function", "builtin_function_or_method", "instance", "_Feature" ], "window_display": false } }, "nbformat": 4, "nbformat_minor": 2 }