diff --git a/content/GA_1_1/Task_2_solution.ipynb b/content/GA_1_1/Task_2_solution.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..4943203610ed1c2ee2726f6c5980ffbacd294250
--- /dev/null
+++ b/content/GA_1_1/Task_2_solution.ipynb
@@ -0,0 +1,1234 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# GA 1.1, Part 2: Data-driven approach: Which model is better?\n",
+    "\n",
+    "<h1 style=\"position: absolute; display: flex; flex-grow: 0; flex-shrink: 0; flex-direction: row-reverse; top: 60px;right: 30px; margin: 0; border: 0\">\n",
+    "    <style>\n",
+    "        .markdown {width:100%; position: relative}\n",
+    "        article { position: relative }\n",
+    "    </style>\n",
+    "    <img src=\"https://gitlab.tudelft.nl/mude/public/-/raw/main/tu-logo/TU_P1_full-color.png\" style=\"width:100px\" />\n",
+    "    <img src=\"https://gitlab.tudelft.nl/mude/public/-/raw/main/mude-logo/MUDE_Logo-small.png\" style=\"width:100px\" />\n",
+    "</h1>\n",
+    "<h2 style=\"height: 10px\">\n",
+    "</h2>\n",
+    "\n",
+    "*[CEGM1000 MUDE](http://mude.citg.tudelft.nl/): Week 1.1, Friday, Sep 6, 2024. This assignment does not need to be turned in.*"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "21f9833788f64e78a35bc8cac535e76d",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "## Overview\n",
+    "\n",
+    "In this assignment we will fit two models to observations of ice break-up date and reflect on their performance.\n",
+    "\n",
+    "The data we are working with represents the number of days since the new year that it took for the ice in a river to completely melt and break apart. The record goes from 1917 to 2019, which is in total 103 years of measurements.\n",
+    "\n",
+    "**Goal of the assignment:** reflect on model performance and model accuracy in the context of its use.\n",
+    "\n",
+    "We will follow these steps:\n",
+    "\n",
+    "1. Import data using `numpy`;\n",
+    "2. Explore the data;\n",
+    "3. Fit a linear regression model;\n",
+    "4. Reflect on its goodness of fit;\n",
+    "5. Interpret confidence intervals;\n",
+    "6. Compare the linear model with a non-linear one;\n",
+    "7. And finally...choose a model to make a bet in the Ice Classic!"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {
+    "cell_id": "10af2251a77c4f84a32d01bed7350da9",
+    "deepnote_cell_type": "code",
+    "deepnote_to_be_reexecuted": false,
+    "execution_millis": 1559,
+    "execution_start": 1694006986292,
+    "source_hash": null
+   },
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "import matplotlib.pyplot as plt\n",
+    "import scipy.stats as sci\n",
+    "import scipy.optimize as opt\n",
+    "\n",
+    "%matplotlib inline"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "b1d3e3d2f92c4de29aba4aa61c525867",
+    "deepnote_cell_type": "markdown",
+    "jp-MarkdownHeadingCollapsed": true,
+    "tags": []
+   },
+   "source": [
+    "## 1. Import data"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We will import the dataset by using the `numpy` function `loadtxt`. If you open the *data-days.csv* file, you will notice that the comma is used as decimal separator for the number of recorded days. For this reason, we will import the data in the following steps:\n",
+    "\n",
+    "1. Read the file *data-days.csv* as strings. To this end, we will use the option `dtype=str` within the function `loadtxt`. Note that we also need to set the delimiter between columns as `delimeter=;` and specify to skip the first row (which contains the header of the columns) as `skiprows=1`.\n",
+    "    \n",
+    "2. Replace the commas by dots. We will use the `numpy` function `char.replace` to this end. The first argument is the numpy array of strings where we want to make changes, the second argument is the character we want to look for and change, and the third argument is what we want to use as replacement.\n",
+    "   \n",
+    "3. Transform it back to `float` data. We will use `data.astype(float)` for this.\n",
+    "\n",
+    "And our data is ready to start playing!\n",
+    "\n",
+    "We also printed for you the first 10 elements of the array, so you can see how it actually looks."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {
+    "cell_id": "47af53f26ebc41b2a90a63dd3b02ab0f",
+    "deepnote_cell_type": "code",
+    "deepnote_to_be_reexecuted": false,
+    "execution_millis": 33,
+    "execution_start": 1694006987852,
+    "source_hash": null
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array([[1917.       ,  119.4791667],\n",
+       "       [1918.       ,  130.3979167],\n",
+       "       [1919.       ,  122.60625  ],\n",
+       "       [1920.       ,  131.4486111],\n",
+       "       [1921.       ,  130.2791667],\n",
+       "       [1922.       ,  131.5555556],\n",
+       "       [1923.       ,  128.0833333],\n",
+       "       [1924.       ,  131.6319444],\n",
+       "       [1925.       ,  126.7722222],\n",
+       "       [1926.       ,  115.66875  ]])"
+      ]
+     },
+     "execution_count": 2,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data = np.loadtxt('data/days.csv', dtype=str, delimiter=';', skiprows=1)\n",
+    "data = np.char.replace(data, ',', '.')\n",
+    "data = data.astype(float)\n",
+    "\n",
+    "data[0:10]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "jp-MarkdownHeadingCollapsed": true,
+    "tags": []
+   },
+   "source": [
+    "## 2. Preliminary analysis"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "1b15837c64e748b89fafad1f8007a399",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "One of the first steps when getting familiar with new data is to see the dimensions of the data. To this end, we can use the `numpy` function `shape`."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "cell_id": "2aaf111f8cdd40959fb3496188242efa",
+    "deepnote_cell_type": "code",
+    "deepnote_to_be_reexecuted": false,
+    "execution_millis": 32,
+    "execution_start": 1694006987878,
+    "source_hash": null
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(103, 2)"
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "np.shape(data)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "e901697b36064391b4a62d78c955dd6b",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "The result is a (103, 2) array, i.e., a matrix with 103 rows and 2 columns. The first column contains the year of record, while the second one contains the measured data."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "43139b45e34f4252a6a270336ca401ba",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "We can also compute the mean and the standard deviation of the variable of interest (second column) to get a sense of how the variable behaves."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {
+    "cell_id": "efc536f1b0804d50ae01f3013a9d1367",
+    "deepnote_cell_type": "code",
+    "deepnote_to_be_reexecuted": false,
+    "execution_millis": 20,
+    "execution_start": 1694006987934,
+    "source_hash": null
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Mean: 123.647\n",
+      "Standard deviation: 6.516\n"
+     ]
+    }
+   ],
+   "source": [
+    "mean = np.mean(data[:,1])\n",
+    "std = np.std(data[:,1])\n",
+    "\n",
+    "print(f'Mean: {mean:.3f}\\n\\\n",
+    "Standard deviation: {std:.3f}')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "2a1ddf21c02141dc8dfebee83c602a73",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "We can also quickly plot them to see the scatter of the data and the evolution in time."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {
+    "cell_id": "a67a4c42e202489e8801bfb67ca95fbc",
+    "deepnote_cell_type": "code",
+    "deepnote_to_be_reexecuted": false,
+    "execution_millis": 229,
+    "execution_start": 1694007029078,
+    "source_hash": null
+   },
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.scatter(data[:, 0], data[:, 1], label='Measured data')\n",
+    "plt.xlabel('Year [-]')\n",
+    "plt.ylabel('Number of days/year [-]')\n",
+    "plt.title(f'Number of days per year between {data[0,0]:.0f}-{data[-1,0]:.0f}')\n",
+    "plt.grid()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "37edb558a10b47d88b3e4f683da56221",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "In the figure above, we have plotted the year of the measurement in the x-axis and the number of days until the ice broke during that year in the y-axis. We can see that there is a significant scatter. Also, there seems to be a trend over time: as we go ahead in time (higher values in the x-axis), the number of days until the ice broke seems to decrease.\n",
+    "\n",
+    "We have identifid a trend but **can we model it**?"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "a05297ff6298401192d49f8e257ff9ec",
+    "deepnote_cell_type": "markdown",
+    "jp-MarkdownHeadingCollapsed": true,
+    "tags": []
+   },
+   "source": [
+    "## 3. Fit a linear regression model: is it a good model?"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We are going to create a model which allows us to predict the number of days until the ice broke as function of the year. For that, we are going to assume a linear relationship between the variables (a linear model) and we will fit it using linear regression. This is, we will fit a regression model $days=m\\cdot year+q$, where $m$ represents the slope of the line, and $q$ is the intercept.\n",
+    "\n",
+    "We will do it using functions which were already coded for us. We will use the `scipy.stats` library which contains the `linregress` function. For more info see [here](https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.linregress.html).\n",
+    "\n",
+    "In the next code block we define the function `regression` which requires x and y values as input and computes the values of $R^2$, $m$, and $q$. \n",
+    "\n",
+    "Remember that $R^2$ is a goodness of fit metric and you should have read about it [here](https://mude.citg.tudelft.nl/book/modelling/gof.html)."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {
+    "cell_id": "4d72dacfdbb543d9925a40a30eac2944",
+    "deepnote_cell_type": "code",
+    "deepnote_to_be_reexecuted": false,
+    "execution_millis": 22,
+    "execution_start": 1694007048665,
+    "source_hash": null
+   },
+   "outputs": [],
+   "source": [
+    "def regression(x, y):\n",
+    "    '''\n",
+    "    Determine linear regression\n",
+    "\n",
+    "    Input: x = array, x values\n",
+    "           y = array, y values\n",
+    "\n",
+    "    Output: r_sq = coefficient of determination\n",
+    "            q = intercept of the line\n",
+    "            m = slope of the line\n",
+    "    '''\n",
+    "\n",
+    "    regression = sci.linregress(x, y)\n",
+    "    r_sq = regression.rvalue**2\n",
+    "    q = regression.intercept \n",
+    "    m = regression.slope \n",
+    "\n",
+    "    print(f'Coefficient of determination R^2 = {r_sq:.3f}')\n",
+    "    print(f'Intercept q = {q:.3f} \\nSlope m = {m:.3f}')\n",
+    "\n",
+    "    return r_sq, q, m\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {
+    "cell_id": "31d247d5cafc4c9690c39bc736f04758",
+    "deepnote_cell_type": "code",
+    "deepnote_to_be_reexecuted": false,
+    "execution_millis": 20,
+    "execution_start": 1694007051941,
+    "source_hash": null
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Coefficient of determination R^2 = 0.151\n",
+      "Intercept q = 291.239 \n",
+      "Slope m = -0.085\n"
+     ]
+    }
+   ],
+   "source": [
+    "r_sq, q, m = regression(data[:,0], data[:,1])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<div style=\"background-color:#AABAB2; color: black; vertical-align: middle; padding:15px; margin: 10px; border-radius: 10px\">\n",
+    "<p>\n",
+    "<b>Task 1:</b> Based on the obtained coefficient of determination, assess the performance of the model. Answer in the cell below using markdown:\n",
+    "<ol>\n",
+    "    <li> What does the coefficient of determination mean in this context?</li>\n",
+    "    <li> Is the developed model accurate? </li>\n",
+    "\n",
+    "</ol>\n",
+    "</p>\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "9a082db4cf644c9c9854af0a458c0b8a",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "<div style=\"background-color:#FAE99E; color: black; vertical-align: middle; padding:15px; margin: 10px; border-radius: 10px\">\n",
+    "<p>\n",
+    "<b>Solution:</b>   \n",
+    "<li> 1. Coefficient of determination ($R^2$) measures the percentage of the variance in our observations explained by the model. Thus, the higher, the better. As we can see, the value of $R^2$ is quite low. Only $\\approx15\\%$ of the variance is explained by the model, which is very low. Therefore, the linear model is not able to explain the scatter in our observations. </li> \n",
+    "<li> 2. Based on the answer to the previous question, the linear model is not an accurate model. Whether this low level of accuracy is good enough or not, depends on the use we want to give to the model. Would you bet 3\\$? And 1,000\\$? </li> \n",
+    "</p>\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "143e5d1bf8324d9f80fca4af9a0d162c",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "We can also plot the data and the fitted model to see how the fit looks. To do so, we can make computations using the previous equation $days=m\\cdot year+q$ with the fitted intercept $q$ and slope $m$. We have already defined a function which does it for you."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {
+    "cell_id": "650d74731c11439684ba4de949b2127b",
+    "deepnote_cell_type": "code",
+    "deepnote_to_be_reexecuted": false,
+    "execution_millis": 16,
+    "execution_start": 1694007056091,
+    "source_hash": null
+   },
+   "outputs": [],
+   "source": [
+    "def calculate_line(x, m, q):\n",
+    "    '''\n",
+    "    Determine y values from linear regression\n",
+    "\n",
+    "    Input: x = array\n",
+    "           m = slope of the line\n",
+    "           q = intercept of the line\n",
+    "\n",
+    "    Output: y = array\n",
+    "    '''\n",
+    "\n",
+    "    y = m * x + q\n",
+    "\n",
+    "    return y"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {
+    "cell_id": "cd3c13f84f124211af2d5e251754d636",
+    "deepnote_cell_type": "code",
+    "deepnote_to_be_reexecuted": false,
+    "execution_millis": 10,
+    "execution_start": 1694007059430,
+    "source_hash": null
+   },
+   "outputs": [],
+   "source": [
+    "line = calculate_line(data[:,0], m, q)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {
+    "cell_id": "4424bf1d4bf244baaa24cfcf0b5782bc",
+    "deepnote_cell_type": "code",
+    "deepnote_to_be_reexecuted": false,
+    "execution_millis": 246,
+    "execution_start": 1694007094334,
+    "source_hash": null
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5, 1.0, '(b) Observed and predicted number of days')"
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1200x400 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "fig, axes = plt.subplots(1, 2,figsize = (12, 4))\n",
+    "\n",
+    "axes[0].scatter(data[:,0], data[:,1], label = 'Observations')\n",
+    "axes[0].plot(data[:,0], line, color='r', label='Fitted line')\n",
+    "axes[0].set_ylabel('Number of days/year [-]')\n",
+    "axes[0].set_xlabel('Year [-]')\n",
+    "axes[0].grid()\n",
+    "axes[0].legend()\n",
+    "axes[0].set_title('(a) Number of days as function of the year')\n",
+    "\n",
+    "axes[1].scatter(data[:,1], line)\n",
+    "axes[1].plot([105, 145],[105, 145], line, color = 'k')\n",
+    "axes[1].set_xlim([105, 145])\n",
+    "axes[1].set_ylim([105, 145])\n",
+    "axes[1].set_ylabel('Predicted number of days/year [-]')\n",
+    "axes[1].set_xlabel('Observed number of days/year [-]')\n",
+    "axes[1].grid()\n",
+    "axes[1].set_title('(b) Observed and predicted number of days')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "b564cab0d51c40f2aa3c7ebb9affaade",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "<div style=\"background-color:#AABAB2; color: black; vertical-align: middle; padding:15px; margin: 10px; border-radius: 10px\">\n",
+    "<p>\n",
+    "<b>Task 2:</b>   \n",
+    "    <p>Interpret the figures above. Do the previous plots fit with the result given by the coefficient of determination? Are they aligned?</p>\n",
+    "<ol>\n",
+    "\n",
+    "</ol>\n",
+    "</p>\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "9a082db4cf644c9c9854af0a458c0b8a",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "<div style=\"background-color:#FAE99E; color: black; vertical-align: middle; padding:15px; margin: 10px; border-radius: 10px\">\n",
+    "<p>\n",
+    "<b>Solution:</b>   \n",
+    "    \n",
+    "    \n",
+    "In the plot (a), we observe that the observations have a high scatter around the fitted line, but how much is that? That is why we use metrics!\n",
+    "    \n",
+    "In the plot (b), we compare the observations with the predictions of the model. The perfect fit would correspond to the 45-degrees line in black. Thus, the model has a poor performance as we already quantified using the coefficient of determination. Both results are aligned.\n",
+    "</p>\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "67e44fda81f24273b8d28edd35a50d87",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "We can also assess the scatter using the Root Mean Square Error ($RMSE$). Don't you remember it? Go back to the [book](https://mude.citg.tudelft.nl/book/modelling/gof.html)!\n",
+    "\n",
+    "Let's see how our model performs for this metric."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {
+    "cell_id": "b9e8d87cb37e4950ad597f3cc9a71b49",
+    "deepnote_cell_type": "code",
+    "deepnote_to_be_reexecuted": true,
+    "execution_millis": 30,
+    "execution_start": 1693551237449,
+    "source_hash": null
+   },
+   "outputs": [],
+   "source": [
+    "def RMSE(data, fit_data):\n",
+    "    '''\n",
+    "    Compute the RMSE\n",
+    "\n",
+    "    RMSE = [sum{(data - fit_data)^2} / N]^(1/2)\n",
+    "\n",
+    "    Input: data = array with real measured data\n",
+    "           fit_data = array with predicted data\n",
+    "    \n",
+    "    Output: RMSE\n",
+    "    '''\n",
+    "\n",
+    "    diff_n = (data - fit_data)**2\n",
+    "    mean = np.mean(diff_n)\n",
+    "\n",
+    "    error = mean**(1/2)\n",
+    "    print(f'RMSE = {error:.3f}')\n",
+    "    return error"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {
+    "cell_id": "a8a3a34164ce4347b238194128ae1ce2",
+    "deepnote_cell_type": "code",
+    "deepnote_to_be_reexecuted": true,
+    "execution_millis": 28,
+    "execution_start": 1693551237454,
+    "source_hash": null
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "RMSE = 6.003\n"
+     ]
+    }
+   ],
+   "source": [
+    "RMSE_line = RMSE(data[:,1], line)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "b564cab0d51c40f2aa3c7ebb9affaade",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "<div style=\"background-color:#AABAB2; color: black; vertical-align: middle; padding:15px; margin: 10px; border-radius: 10px\">\n",
+    "<p>\n",
+    "<b>Task 3:</b>   \n",
+    "    Based on the obtained RMSE, assess the performance of the model. Answer in the cell below using markdown:\n",
+    "<ol>\n",
+    "    <li> What does RMSE mean in this context?</li>\n",
+    "    <li> Is the developed model accurate according to RMSE? </li>\n",
+    "\n",
+    "</ol>\n",
+    "</p>\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "604f11f41ac54d59b11422bac89e6411",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "<div style=\"background-color:#FAE99E; color: black; vertical-align: middle; padding:15px; margin: 10px; border-radius: 10px\">\n",
+    "<p>\n",
+    "<b>Solution:</b>   \n",
+    "<li> 1. $RMSE$ represents the mean error between the observations and the predictions of the model. This means that the mean error is $\\approx 6$ days. Therefore, the linear model is not able to explain the scatter in our observations. </li> \n",
+    "<li> 2. Based on the previous interpretation, the linear model is not accurate. Whether this low level of accuracy is good enough or not, depends on the use we want to give to the model. Would you bet $3? And $1,000? </li> \n",
+    "</p>\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Finally, we can compute the bias of our model using $rbias$."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def rbias(data, fit_data):\n",
+    "    '''\n",
+    "    Compute the relative bias\n",
+    "\n",
+    "    rbias = [sum{(fit_data-data) / |data|}]/N\n",
+    "\n",
+    "    Input: data = array with real measured data\n",
+    "           fit_data = array with predicted data\n",
+    "    \n",
+    "    Output: relative bias\n",
+    "    '''\n",
+    "    bias = np.mean((fit_data-data)/data)\n",
+    "\n",
+    "    print(f'rbias = {bias:.3f}')\n",
+    "    return bias"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "rbias = 0.002\n"
+     ]
+    }
+   ],
+   "source": [
+    "rbias_line = rbias(data[:,1], line)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "b564cab0d51c40f2aa3c7ebb9affaade",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "<div style=\"background-color:#AABAB2; color: black; vertical-align: middle; padding:15px; margin: 10px; border-radius: 10px\">\n",
+    "<p>\n",
+    "<b>Task 4:</b>   \n",
+    "    Based on the obtained relative bias, assess the performance of the model. Answer in the cell below using markdown:\n",
+    "<ol>\n",
+    "    <li> What does rbias mean in this context?</li>\n",
+    "</ol>\n",
+    "</p>\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "604f11f41ac54d59b11422bac89e6411",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "<div style=\"background-color:#FAE99E; color: black; vertical-align: middle; padding:15px; margin: 10px; border-radius: 10px\">\n",
+    "<p>\n",
+    "<b>Solution:</b>   \n",
+    "<li> 1. $rbias$ provides an standardized measure of the systematic tendency of our model to under- or over-prediction. It is very low for our model, so it does not have a clear tendency to under- or overestimate and, thus, does not seem to be biased. </li> \n",
+    "</p>\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "b564cab0d51c40f2aa3c7ebb9affaade",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "<div style=\"background-color:#AABAB2; color: black; vertical-align: middle; padding:15px; margin: 10px; border-radius: 10px\">\n",
+    "<p>\n",
+    "<b>Task 5:</b>   \n",
+    "    Based on all the computed GOF metrics, Is the fitted model a good model? Would you...\n",
+    "    <li> ...bet 3€?</li>\n",
+    "    <li> ...bet 1,000€?</li>\n",
+    "   \n",
+    "<ol>\n",
+    "</ol>\n",
+    "</p>\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<div style=\"background-color:#FAE99E; color: black; vertical-align: middle; padding:15px; margin: 10px; border-radius: 10px\">\n",
+    "<p>\n",
+    "<b>Solution:</b>   \n",
+    "No right answer, that depends on the risk you want to take ;)\n",
+    "</p>\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "19496dbefd3247f09c2226579c7b665f",
+    "deepnote_cell_type": "markdown",
+    "jp-MarkdownHeadingCollapsed": true,
+    "tags": []
+   },
+   "source": [
+    "## 4. Confidence Intervals"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "One way of assessing the uncertainty around the predictions of a model are confidence intervals. They give us insight into the precision of their predictions by transforming them into probabilities. In short, the 95% confidence interval (significance $\\alpha=0.05$) shows the range of values within which my observation would be with a probability of 95%. Here, we want you to focus on their interpretation. In the following weeks (1.3), you will learn more about how to compute them. "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<div style=\"background-color:#C8FFFF; color: black; vertical-align: middle; padding:3px 20px;\">\n",
+    "<p>\n",
+    "<b>Note:</b>   \n",
+    "The confidence intervals as computed here and later in part 5 are based on a simplification, in week 1.3 you will see how to compute the confidence intervals correctly. This will also result in a different 'shape' (curved intervals).\n",
+    "</p>\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {
+    "cell_id": "adf17125447a424f915d421e541b8a6e",
+    "deepnote_cell_type": "code",
+    "deepnote_to_be_reexecuted": true,
+    "execution_millis": 9,
+    "execution_start": 1693551237510,
+    "source_hash": null
+   },
+   "outputs": [],
+   "source": [
+    "def conf_int(x, y, alpha):\n",
+    "    '''\n",
+    "    Compute the confidence intervals\n",
+    "\n",
+    "    Input: x = array, observations\n",
+    "           y = array, predictions\n",
+    "           alpha = float, confidence interval\n",
+    "\n",
+    "    Output: k = float, width of the confidence interval\n",
+    "    '''\n",
+    "    sd_error = (y - x).std()\n",
+    "    k = sci.norm.ppf(1-alpha/2)*sd_error\n",
+    "\n",
+    "    return k"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5, 1.0, 'Number of days as function of the year')"
+      ]
+     },
+     "execution_count": 16,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "k = conf_int(data[:,1], line, 0.05)\n",
+    "ci_low = line - k\n",
+    "ci_up = line + k\n",
+    "\n",
+    "#plot\n",
+    "plt.scatter(data[:,0], data[:,1], label = 'Observations')\n",
+    "plt.plot(data[:,0], line, color='r', label='Fitted line')\n",
+    "plt.plot(data[:,0], ci_low, '--k')\n",
+    "plt.plot(data[:,0], ci_up, '--k')\n",
+    "plt.ylabel('Number of days/year [-]')\n",
+    "plt.xlabel('Year [-]')\n",
+    "plt.grid()\n",
+    "plt.legend()\n",
+    "plt.title('Number of days as function of the year')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "b564cab0d51c40f2aa3c7ebb9affaade",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "<div style=\"background-color:#AABAB2; color: black; vertical-align: middle; padding:15px; margin: 10px; border-radius: 10px\">\n",
+    "<p>\n",
+    "<b>Task 7:</b>   \n",
+    "    What can you conclude from the previous plot? Think about the accuracy of the prediction compared to the precision you need in your bet (up to minutes!).\n",
+    "<ol>\n",
+    "</ol>\n",
+    "</p>\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "604f11f41ac54d59b11422bac89e6411",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "<div style=\"background-color:#FAE99E; color: black; vertical-align: middle; padding:15px; margin: 10px; border-radius: 10px\">\n",
+    "<p>\n",
+    "<b>Solution:</b>   \n",
+    "If you consider that you need to place a bet with not only the day but also the hour and minute at which the ice would break, the model is not accurate enough. You can see that the confidence interval spans almost 20 days!\n",
+    "</p>\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "a19fb5b7e1cd4c32b435b8b967933700",
+    "deepnote_cell_type": "markdown",
+    "jp-MarkdownHeadingCollapsed": true,
+    "tags": []
+   },
+   "source": [
+    "## 5. Non-linear models"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "As we have seen, the data-driven linear model is not really a good choice for representing the data we have. Let's try with one which is slightly more complicated: a non-linear model.\n",
+    "\n",
+    "In this section, we will analyze the fitting of a quadratic model as $days = A year^2 + B year + C$. The steps are the same as in the previous section, so we will go fast through the code to focus on the interpretation and comparison between the two models."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "ca8a6b9d68234ae69a799a3f4f3866a2",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "You do not need to worry about this right now, but in case you are curious: we will make use of the `scipy.optimize` library, which contains the `curve_fit` function. For further info on the function see [here](https://docs.scipy.org/doc/scipy/reference/generated/scipy.optimize.curve_fit.html)."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {
+    "cell_id": "ebd3c8f852c24c9d8128ae35482824de",
+    "deepnote_cell_type": "code",
+    "deepnote_to_be_reexecuted": true,
+    "execution_millis": 22,
+    "execution_start": 1693551238912,
+    "source_hash": null
+   },
+   "outputs": [],
+   "source": [
+    "def parabola(x, a, b, c):\n",
+    "    '''\n",
+    "    Compute the quadratic model\n",
+    "\n",
+    "    y = a * x^2 + b * x + c\n",
+    "\n",
+    "    Input: x = array, independent variable\n",
+    "           a, b, c = parameters to be optimized\n",
+    "\n",
+    "    Output: y = array, dependent variable\n",
+    "    '''\n",
+    "\n",
+    "    y = a * x**2 + b * x + c\n",
+    "    return y"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "metadata": {
+    "cell_id": "6897b364543142b4adc2d85ffbcb7d42",
+    "deepnote_cell_type": "code",
+    "deepnote_to_be_reexecuted": true,
+    "execution_millis": 29,
+    "execution_start": 1693551238923,
+    "source_hash": null
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Optimal estimation for parameters:\n",
+      "a = -1.277e-03, b = 4.942, c = -4654.244\n",
+      "\n",
+      "Covariance matrix for parameters:\n",
+      "Sigma = [[ 5.60366689e-07 -2.20560328e-03  2.16981823e+00]\n",
+      " [-2.20560328e-03  8.68165065e+00 -8.54118418e+03]\n",
+      " [ 2.16981823e+00 -8.54118418e+03  8.40337452e+06]]\n"
+     ]
+    }
+   ],
+   "source": [
+    "popt_parabola, pcov_parabola = opt.curve_fit(parabola, data[:,0], data[:,1])\n",
+    "\n",
+    "print(f'Optimal estimation for parameters:\\n\\\n",
+    "a = {popt_parabola[0]:.3e}, b = {popt_parabola[1]:.3f}, c = {popt_parabola[2]:.3f}\\n')\n",
+    "\n",
+    "print(f'Covariance matrix for parameters:\\n\\\n",
+    "Sigma = {pcov_parabola}')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "10efd81771064ce0ac4d50095be06e23",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "Therefore, our parabola now looks like $days = -1.277 \\cdot 10^{-3} \\cdot year^2 + 4.942 \\cdot year - 4654.244$.\n",
+    "\n",
+    "Now that we have fitted it, we can use it to compute predictions."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "metadata": {
+    "cell_id": "182a633995af49a9ba16ac67f2d0181d",
+    "deepnote_cell_type": "code",
+    "deepnote_to_be_reexecuted": true,
+    "execution_millis": 55,
+    "execution_start": 1693551238924,
+    "source_hash": null
+   },
+   "outputs": [],
+   "source": [
+    "fitted_parabola = parabola(data[:,0], *popt_parabola)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "6cc6a36c668f4ee2b26d41b27d335691",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "We can also determine the confidence intervals for this fit and see how it looks!"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "metadata": {
+    "cell_id": "4221e9ef1da1463eb5f524d9a458ab0b",
+    "deepnote_cell_type": "code",
+    "deepnote_to_be_reexecuted": true,
+    "execution_millis": 13,
+    "execution_start": 1693551238973,
+    "source_hash": null
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5, 1.0, 'Number of days as function of the year')"
+      ]
+     },
+     "execution_count": 20,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "k = conf_int(data[:,1], fitted_parabola, 0.05)\n",
+    "ci_low_2 = fitted_parabola - k\n",
+    "ci_up_2 = fitted_parabola + k\n",
+    "\n",
+    "#plot\n",
+    "plt.scatter(data[:,0], data[:,1], label = 'Observations')\n",
+    "plt.plot(data[:,0], fitted_parabola, color='r', label='Fitted line')\n",
+    "plt.plot(data[:,0], ci_low_2, '--k')\n",
+    "plt.plot(data[:,0], ci_up_2, '--k')\n",
+    "plt.ylabel('Number of days/year [-]')\n",
+    "plt.xlabel('Year [-]')\n",
+    "plt.grid()\n",
+    "plt.legend()\n",
+    "plt.title('Number of days as function of the year')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "And finally compute the GOF metrics."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "RMSE = 5.918\n",
+      "Coefficient of determination = 0.175\n",
+      "rbias = 0.002\n"
+     ]
+    }
+   ],
+   "source": [
+    "RMSE_parabola = RMSE(data[:,1], fitted_parabola)\n",
+    "R2_parabola = 1-((data[:,1]-fitted_parabola)**2).mean()/(data[:,1].var())\n",
+    "print(f'Coefficient of determination = {R2_parabola:.3f}')\n",
+    "rbias_parabola = rbias(data[:,1], fitted_parabola)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "cell_id": "b564cab0d51c40f2aa3c7ebb9affaade",
+    "deepnote_cell_type": "markdown"
+   },
+   "source": [
+    "<div style=\"background-color:#AABAB2; color: black; vertical-align: middle; padding:15px; margin: 10px; border-radius: 10px\">\n",
+    "<p>\n",
+    "<b>Task 8:</b>   \n",
+    "    Based on the previous plot and the computed GOF metrics, is the quadratic model better than the linear one?\n",
+    "<ol>\n",
+    "</ol>\n",
+    "</p>\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<div style=\"background-color:#FAE99E; color: black; vertical-align: middle; padding:15px; margin: 10px; border-radius: 10px\">\n",
+    "<p>\n",
+    "<b>Solution:</b>   \n",
+    "No! The GOF metrics are actually similar while the non-linear model is more complicated (e.g., it includes more parameters to fit).\n",
+    "    \n",
+    "At the same time we can see that the observations do fall quite well within the 95% confidence bounds. Moreover, the 'misfit' is clearly due to large year-to-year variability in the observed number of days. This variability might be explainable by the year-to-year weather conditions (e.g., number of heatwaves). The question is then: can we find a model by including a weather-related explanatory variable to get a better fit? But also, is this really needed? That depends on whether you are interested in predicting the number of days on the short term (e.g., next year) or the number of days in the long term (e.g., in 10 years or longer).\n",
+    "    \n",
+    "We will revisit this problem in the part on Sensing and Observation Theory in week 1.4.\n",
+    "</p>\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<div style=\"background-color:#AABAB2; color: black; vertical-align: middle; padding:15px; margin: 10px; border-radius: 10px\">\n",
+    "<p>\n",
+    "<b>Task 9:</b>   \n",
+    "    Consider that you are using the predictions of the model you found better in the previous question:\n",
+    "<ol>\n",
+    "    <li>  Would you bet &#36;3?</li> \n",
+    "    <li>  Would you bet &#36;1,000?</li> \n",
+    "</ol>\n",
+    "</p>\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<div style=\"background-color:#FAE99E; color: black; vertical-align: middle; padding:15px; margin: 10px; border-radius: 10px\">\n",
+    "<p>\n",
+    "<b>Solution:</b>   \n",
+    "No right answer, that depends on the risk you want to take ;)\n",
+    "</p>\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "**End of notebook.**\n",
+    "<h2 style=\"height: 60px\">\n",
+    "</h2>\n",
+    "<h3 style=\"position: absolute; display: flex; flex-grow: 0; flex-shrink: 0; flex-direction: row-reverse; bottom: 60px; right: 50px; margin: 0; border: 0\">\n",
+    "    <style>\n",
+    "        .markdown {width:100%; position: relative}\n",
+    "        article { position: relative }\n",
+    "    </style>\n",
+    "    <a rel=\"license\" href=\"http://creativecommons.org/licenses/by/4.0/\">\n",
+    "      <img alt=\"Creative Commons License\" style=\"border-width:; width:88px; height:auto; padding-top:10px\" src=\"https://i.creativecommons.org/l/by/4.0/88x31.png\" />\n",
+    "    </a>\n",
+    "    <a rel=\"TU Delft\" href=\"https://www.tudelft.nl/en/ceg\">\n",
+    "      <img alt=\"TU Delft\" style=\"border-width:0; width:100px; height:auto; padding-bottom:0px\" src=\"https://gitlab.tudelft.nl/mude/public/-/raw/main/tu-logo/TU_P1_full-color.png\" />\n",
+    "    </a>\n",
+    "    <a rel=\"MUDE\" href=\"http://mude.citg.tudelft.nl/\">\n",
+    "      <img alt=\"MUDE\" style=\"border-width:0; width:100px; height:auto; padding-bottom:0px\" src=\"https://gitlab.tudelft.nl/mude/public/-/raw/main/mude-logo/MUDE_Logo-small.png\" />\n",
+    "    </a>\n",
+    "    \n",
+    "</h3>\n",
+    "<span style=\"font-size: 75%\">\n",
+    "&copy; Copyright 2024 <a rel=\"MUDE\" href=\"http://mude.citg.tudelft.nl/\">MUDE</a> TU Delft. This work is licensed under a <a rel=\"license\" href=\"http://creativecommons.org/licenses/by/4.0/\">CC BY 4.0 License</a>."
+   ]
+  }
+ ],
+ "metadata": {
+  "deepnote": {},
+  "deepnote_execution_queue": [],
+  "deepnote_notebook_id": "283160ea7b7649c7a8ca381512535534",
+  "deepnote_persisted_session": {
+   "createdAt": "2023-09-06T14:01:47.518Z"
+  },
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.11.4"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}