{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Hands-on session 1: Features\n", "\n", "These exercises cover the topics introduced in the day 2 [tagging tutorial](resources/tutorial_1.ipynb) introduction of working with **Features** attached to Tags.\n", "\n", "## Radar trap example\n", "\n", "![radar trap](resources/radar_trap.jpg)\n", "\n", "While introducing the **RangeDimension** we used the example scenario of a [radar trap](../day_1/tutorial_1.ipynp), which measures the speed of passing cars. We will now extend on to this scenario by adding first identifying speeders in the data, and linking them to the raw measurements. Then we will add a \"fine\" for each speeder.\n", "\n", "The code below creates a new set of speed measurements and also provides the index of the speeder, and the related fine. It stores the raw data to a nix file. Extend the program to also store and relate the detected speed limit violations to their fine:\n", "\n", "1. Store the times of the speeders in an additional **DataArray** and link it to the speed data using a **MultiTag**\n", "2. Store the fine in a separate **DataArray** and add it to the **MultiTag** as a **Feature**. What **LinkType** should be used?\n", "3. Close the file.\n", "\n", "4. Reopen the file for ``ReadOnly``.\n", "5. Write a loop to read the actual speed (``tagged_data``), the time at which the speeder was observed and the associated fine (``feature_data``) from the **MultiTag** and print the information on the screen(``print(time, speed, fine)``).\n", "6. Close the file.\n" ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": { "image/png": { "height": 261, "width": 382 }, "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "import os\n", "import glob\n", "import nixio\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", "%matplotlib inline \n", "%config InlineBackend.figure_formats = ['retina'] # only for users with a high resolution display\n", "\n", "\n", "def radar_trap_data(duration=1000, car_probability=0.15, dt=0.01, speed_limit=50, speeder_probability=0.01):\n", " time = np.arange(0, duration, dt)\n", " car_times = time[np.random.random(len(time)) < (car_probability * dt)]\n", " car_speeds = speed_limit + np.random.randn(len(car_times)) * 0.05 * speed_limit\n", " indices = np.arange(len(car_times), dtype=int)\n", " np.random.shuffle(indices)\n", " speeders = indices[:int(np.round(speeder_probability * len(car_times)))]\n", " car_speeds[speeders] += 0.5 * speed_limit\n", " fines = car_speeds[speeders] * 1.5 + 15 # Euros ;)\n", "\n", " return car_times, car_speeds, speeders, fines\n", "\n", "\n", "def plot_raw_data(times, speeds, speed_limit=50):\n", " plt.plot(times, speeds, marker=\".\", zorder=2)\n", " plt.xlabel(\"time [s]\")\n", " plt.ylabel(\"speed [km/h]\")\n", " plt.hlines([speed_limit], 0, times[-1], color=\"grey\", ls=\"--\", zorder=1)\n", "\n", "\n", "times, speeds, speeders, fines = radar_trap_data(duration=5000)\n", "plot_raw_data(times, speeds)\n", "\n", "# store data to nix\n", "nixfile = nixio.File.open(\"radar_trap.nix\", nixio.FileMode.Overwrite)\n", "block = nixfile.create_block(\"radar trap\", \"speed_measurements\")\n", "\n", "speed_array = block.create_data_array(\"car speeds\", \"nix.irregular_sampled\", data=speeds, label=\"speed\", unit=\"km/h\")\n", "speed_array.append_range_dimension(ticks=times, unit=\"s\", label=\"time\")\n", "\n", "nixfile.close()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Exercise 2: add a string feature\n", "\n", "So far the speeder will go unpunished for we did not identify them. Let's remedy this and add the number plate information.\n", "\n", "1. Replace the radar trap function with the version below.\n", "2. Add the number plate information to the nix file **Note:** please pass the data type to the **DataArray** (``block.create_data_array(..., dtype=nixio.DataType.String, data=number_plates)``) and use it as an additional feature of the \"Speeders\" **MultiTag**.\n", "3. Close the file.\n", "\n", "4. Reopen it in ``ReadOnly`` mode.\n", "5. Extend the solution from above to also show the number plate.\n", "6. Close the file." ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [], "source": [ "\n", "def generate_number_plate():\n", " letters = 'ABCDEFGHIJKLMNOPQRSTUVWXYZ'\n", " chosen_letters = np.random.choice(list(letters), 4)\n", " number_plate = \"%s-%s-%i\" % (\"\".join(chosen_letters[:2]), \"\".join(chosen_letters[2:]), np.random.randint(999))\n", " return number_plate\n", "\n", "\n", "def radar_trap_data(duration=1000, car_probability=0.15, dt=0.01, speed_limit=50, speeder_probability=0.01):\n", " time = np.arange(0, duration, dt)\n", " car_times = time[np.random.random(len(time)) < (car_probability * dt)]\n", " car_speeds = speed_limit + np.random.randn(len(car_times)) * 0.05 * speed_limit\n", " indices = np.arange(len(car_times), dtype=int)\n", " np.random.shuffle(indices)\n", " speeders = indices[:int(np.round(speeder_probability * len(car_times)))]\n", " car_speeds[speeders] += 0.5 * speed_limit\n", " fines = car_speeds[speeders] * 1.5 + 15 # Euros ;)\n", " \n", " number_plates = []\n", " for i in range(len(fines)):\n", " number_plates.append(generate_number_plate())\n", "\n", " return car_times, car_speeds, speeders, fines, number_plates\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Exercise 3: Adding a multidimensional feature\n", "\n", "The Features we added so far are rather easy, single numbers or strings for each tagged position. But what would be a radar trap if it didn't take pictures.\n", "\n", "1. Recreate the radar trap data with the code below.\n", "2. Extend the program to add the Image data to the nix file. **Note:** In order to use the image data as a feature, the first dimension of the image data must be as long as there are positions.\n", "3. Add the Image Feature to the **MultiTag**.\n", "4. Close the file.\n", "\n", "5. Re-open for ``ReadOnly`` access.\n", "6. For each of the positions get the image data and illustrate it.\n", "7. Close the file.\n" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [], "source": [ "\n", "def generate_number_plate():\n", " letters = 'ABCDEFGHIJKLMNOPQRSTUVWXYZ'\n", " chosen_letters = np.random.choice(list(letters), 4)\n", " number_plate = \"%s-%s-%i\" % (\"\".join(chosen_letters[:2]), \"\".join(chosen_letters[2:]), np.random.randint(999))\n", " return number_plate\n", "\n", "\n", "def get_image():\n", " imgs = glob.glob(os.path.join(\"resources\", \"radar_trap_*.png\"))\n", " img = np.random.choice(imgs)\n", " img_data = plt.imread(img)\n", " return img_data\n", "\n", "\n", "def radar_trap_data(duration=1000, car_probability=0.15, dt=0.01, speed_limit=50, speeder_probability=0.01):\n", " time = np.arange(0, duration, dt)\n", " car_times = time[np.random.random(len(time)) < (car_probability * dt)]\n", " car_speeds = speed_limit + np.random.randn(len(car_times)) * 0.05 * speed_limit\n", " indices = np.arange(len(car_times), dtype=int)\n", " np.random.shuffle(indices)\n", " speeders = indices[:int(np.round(speeder_probability * len(car_times)))]\n", " car_speeds[speeders] += 0.5 * speed_limit\n", " fines = car_speeds[speeders] * 1.5 + 15 # Euros ;)\n", " \n", " number_plates = []\n", " pictures = []\n", " for i in range(len(fines)):\n", " number_plates.append(generate_number_plate())\n", " pictures.append(get_image())\n", "\n", " return car_times, car_speeds, speeders, fines, number_plates, pictures" ] } ], "metadata": { "interpreter": { "hash": "aee8b7b246df8f9039afb4144a1f6fd8d2ca17a180786b69acc140d282b71a49" }, "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.9.5" } }, "nbformat": 4, "nbformat_minor": 2 }