{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"**I would like to share some tips about the data of this competition with you.\nAfter analyzing the text files of the training and test data, there were significant points in them that no one mentioned.\nHere are some tips:**\n","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19"}},{"cell_type":"markdown","source":"\n**1. In text files, timestamps and features values are not sorted by time! You need to sort them out before working with them.**\n\n\n","metadata":{}},{"cell_type":"markdown","source":"**2. Wi-Fi signals are updated every 2 seconds but waypoints are given every 10 seconds.**\n","metadata":{}},{"cell_type":"markdown","source":"**3. The number of Wi-Fi signals in each update is different.**\n","metadata":{}},{"cell_type":"markdown","source":"**4. It seems that it is better to use signals that are closer to the waypoints based on time to training the model.**\n","metadata":{}},{"cell_type":"markdown","source":"**5. In my opinion, 3 features of Wi-Fi, magnetic field and beacon can be used for modeling.**\n","metadata":{}},{"cell_type":"markdown","source":"**6. To improve the model, the floor feature (which has been predicted with high accuracy) can be added to the feature set.**\n","metadata":{}},{"cell_type":"markdown","source":"**7. It is better to use BSSID of WiFi which is equivalent to device MAC address.**\n\n","metadata":{}},{"cell_type":"markdown","source":"**In the image below, you can see the time steps of the various features in the text files. Wi-Fi signals are updated every 2 seconds but waypoints are given every 10 seconds. Magnetic field and beacon features are almost present.**\n\n","metadata":{}},{"cell_type":"code","source":"import matplotlib.pyplot as plt\nimport matplotlib.image as mpimg\nimg = mpimg.imread('../input/timesteps/time-steps.jpg')\nplt.figure(figsize=(20,15))\nimgplot = plt.imshow(img)\nplt.show()","metadata":{"trusted":true},"execution_count":null,"outputs":[]}]}