{"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":"![%281%29.jpg](attachment:%281%29.jpg)","metadata":{},"attachments":{"%281%29.jpg":{"image/jpeg":"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"}}},{"cell_type":"markdown","source":"# DESCRIPTION","metadata":{}},{"cell_type":"markdown","source":"## Goal of the Competition  \n\nThe goal of this competition is to detect **`freezing of gait (FOG)`**, a debilitating symptom that afflicts many people with Parkinson’s disease.  \n\nIt is necessary to develop a machine learning model trained on data collected with a wearable 3D lower back sensor.  \n\nThis work will help researchers better understand when and why **`FOG episodes`** occur.  \n\nThis will improve the ability of healthcare professionals to optimally assess, manage and ultimately prevent **`FOG events`**. \n\n\nThis event dataset contains 3D lower back accelerometer data from subjects who experienced freeze-walking episodes, a disability symptom common among people with Parkinson's disease. Freeze gait (FOG) is perceived negatively by the likelihood of walking and experiencing movement and independence.  \n\nOur goal is to start and stop an episode of each fade, as well as the appearance in a series of three types of fade events of the march: \"StartHesitation\", \"Turn\" and \"Walking\". ","metadata":{}},{"cell_type":"markdown","source":"> Description of values:\n\n\n|   Name:   |   Type:   |     Meaning:    |\n|-----------|-----------|-----------------|\n|  **Id**   |   object  | The data series the event occured in |\n|**Time**   |   int64   | An integer timestep. Series from the tdcsfog dataset are recorded at 128Hz (128 timesteps per second), while series from the defog and daily series are recorded at 100Hz (100 timesteps per second). |\n| **AccV**  |  float64  | Acceleration from a lower-back sensor on three axes: V - vertical. | \n| **AccML** |  float64  | Acceleration from a lower-back sensor on three axes: ML - mediolateral|\n| **AccAP** |  float64  |  Acceleration from a lower-back sensor on three axes: AP - anteroposterior|                           \n| **Event** |  float64  | Indicator variable for the occurrence of any FOG-type event. Present only in the notype series, which lack type-level annotations|\n| **Valid** |  object   | There were cases during the video annotation that were hard for the annotator to decide if there was an Akinetic (i.e., essentially no movement) FoG or the subject stopped voluntarily. Only event annotations where the series is marked true should be considered as unambiguous |\n| **Task** |   object   | Series were only annotated where this value is true. Portions marked false should be considered unannotated |\n|**StartHesitation**| int64 | Indicator variable for the occurrence of each of the event types|\n| **Turn** |int64 | Indicator variable for the occurrence of each of the event types|\n| **Walking**|int64 | Indicator variable for the occurrence of each of the event types|\n\n \n NB! **Data is in units of m/s^2 for tdcsfog/ and g for defog/ and notype** ","metadata":{}},{"cell_type":"markdown","source":"![EDA-3.jpg](attachment:EDA-3.jpg)","metadata":{},"attachments":{"EDA-3.jpg":{"image/jpeg":"/9j/4AAQSkZJRgABAQEBLAEsAAD/4QD6RXhpZgAATU0AKgAAAAgABgESAAMAAAABAAEAAAEaAAUAAAABAAAAVgEbAAUAAAABAAAAXgEoAAMAAAABAAIAAAITAAMAAAABAAEAAIdpAAQAAAABAAAAZgAAAAAAAAEsAAAAAQAAASwAAAABAAeQAAAHAAAABDAyMzGRAQAHAAAABAECAwCShgAHAAAAMgAAAMCgAAAHAAAABDAxMDCgAQADAAAAAf//AACgAgADAAAAAQbUAACgAwADAAAAAQTYAAAAAAAAQVNDSUkAAAB4cjpkOkRBRmYtdXdNLVZJOjgsajoyNjQ4MTY5NzM1LHQ6MjMwNDEzMTD/4QTbaHR0cDovL25zLmFkb2JlLmNvbS94YXAvMS4wLwA8eDp4bXBtZXRhIHhtbG5zOng9ImFkb2JlOm5zOm1ldGEvIj4NCgk8cmRmOlJERiB4bWxuczpyZGY9Imh0dHA6Ly93d3cudzMub3JnLzE5OTkvMDIvMjItcmRmLXN5bnRheC1ucyMiPg0KCQk8cmRmOkRlc2NyaXB0aW9uIHJkZjphYm91dD0iIiB4bWxuczpkYz0iaHR0cDovL3B1cmwub3JnL2RjL2VsZW1lbnRzLzEuMS8iPg0KCQkJPGRjOnRpdGxlPg0KCQkJCTxyZGY6QWx0Pg0KCQkJCQk8cmRmOmxpIHhtbDpsYW5nPSJ4LWRlZmF1bHQiPkVEQSAtIDE8L3JkZjpsaT4NCgkJCQk8L3JkZjpBbHQ+DQoJCQk8L2RjOnRpdGxlPg0KCQk8L3JkZjpEZXNjcmlwdGlvbj4NCgkJPHJkZjpEZXNjcmlwdGlvbiByZGY6YWJvdXQ9IiIgeG1sbnM6QXR0cmliPSJodHRwOi8vbnMuYXR0cmlidXRpb24uY29tL2Fkcy8xLjAvIj4NCgkJCTxBdHRyaWI6QWRzPg0KCQkJCTxyZGY6U2VxPg0KCQkJCQk8cmRmOmxpIHJkZjpwYXJzZVR5cGU9IlJlc291cmNlIj4NCgkJCQkJCTxBdHRyaWI6Q3JlYXRlZD4yMDIzLTA0LTEzPC9BdHRyaWI6Q3JlYXRlZD4NCgkJCQkJCTxBdHRyaWI6RXh0SWQ+ODJiMDg0OTEtNzYyNi00MzY0LTg4Y2UtMmYyMDI1YjU4OWI4PC9BdHRyaWI6RXh0SWQ+DQoJCQkJCQk8QXR0cmliOkZiSWQ+NTI1MjY1OTE0MTc5NTgwPC9BdHRyaWI6RmJJZD4NCgkJCQkJCTxBdHRyaWI6VG91Y2hUeXBlPjI8L0F0dHJpYjpUb3VjaFR5cGU+DQoJCQkJCTwvcmRmOmxpPg0KCQkJCTwvcmRmOlNlcT4NCgkJCTwvQXR0cmliOkFkcz4NCgkJPC9yZGY6RGVzY3JpcHRpb24+DQoJCTxyZGY6RGVzY3JpcHRpb24gcmRmOmFib3V0PSIiIHhtbG5zOnBkZj0iaHR0cDovL25zLmFkb2JlLmNvbS9wZGYvMS4zLyI+DQoJCQk8cGRmOkF1dGhvcj5Db2JhIFJvYmE8L3BkZjpBdXRob3I+DQoJCTwvcmRmOkRlc2NyaXB0aW9uPg0KCQk8cmRmOkRlc2NyaXB0aW9uIHJkZjphYm91dD0iIiB4bWxuczp4bXA9Imh0dHA6Ly9ucy5hZG9iZS5jb20veGFwLzEuMC8iPg0KCQkJPHhtcDpDcmVhdG9yVG9vbD5DYW52YTwveG1wOkNyZWF0b3JUb29sPg0KCQk8L3JkZjpEZXNjcmlwdGlvbj4NCgk8L3JkZjpSREY+DQo8L3g6eG1wbWV0YT4NCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgIDw/eHBhY2tldCBlbmQ9J3cnPz7/2wBDAAIBAQIBAQICAgICAgICAwUDAwMDAwYEBAMFBwYHBwcGBwcICQsJCAgKCAcHCg0KCgsMDAwMBwkODw0MDgsMDAz/2wBDAQICAgMDAwYDAwYMCAcIDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAwMDAz/wAARCADGARcDASIAAhEBAxEB/8QAHwAAAQUBAQEBAQEAAAAAAAAAAAECAwQFBgcICQoL/8QAtRAAAgEDAwIEAwUFBAQAAAF9AQIDAAQRBRIhMUEGE1FhByJxFDKBkaEII0KxwRVS0fAkM2JyggkKFhcYGRolJicoKSo0NTY3ODk6Q0RFRkdISUpTVFVWV1hZWmNkZWZnaGlqc3R1dnd4eXqDhIWGh4iJipKTlJWWl5iZmqKjpKWmp6ipqrKztLW2t7i5usLDxMXGx8jJytLT1NXW19jZ2uHi4+Tl5ufo6erx8vP09fb3+Pn6/8QAHwEAAwEBAQEBAQEBAQAAAAAAAAECAwQFBgcICQoL/8QAtREAAgECBAQDBAcFBAQAAQJ3AAECAxEEBSExBhJBUQdhcRMiMoEIFEKRobHBCSMzUvAVYnLRChYkNOEl8RcYGRomJygpKjU2Nzg5OkNERUZHSElKU1RVVldYWVpjZGVmZ2hpanN0dXZ3eHl6goOEhYaHiImKkpOUlZaXmJmaoqOkpaanqKmqsrO0tba3uLm6wsPExcbHyMnK0tPU1dbX2Nna4uPk5ebn6Onq8vP09fb3+Pn6/9oADAMBAAIRAxEAPwD9/KKKKACiiigCvrE1xBpN1JZxJNdpEzQRu21ZHAO1SewJwM15b+zJ4l8feNrDUpviBpS6dNaXKixXyPs7PkHflNxyqHAVu+Ty2M161UN3bGZMxsI5lHyvjIH1HGR7ZH4da0jNKLjbfr1REoXkpXenToTVmWv7nxDIh4zG0g+hKf8AswaoL7x5pnh+7sbPVtQsNO1DUH8m2gmnVGu3xyIgcF/wGckDqRXN+JfA2p+K/ix4Y16HxBq2i2emxTGbSo1Plagpx/rcMFB+YcMGOF42nJBGP82gSl/LqS/Er4wWPwm8X+HrG6sdTuz4suvskTWsIdLdwUXfISRgHevTJwjHoDXWSSXF7CwjihVWXjzW3A/UDqPxqB5PtM73KuGWOZYVwcjhir/qSD/u1PJpJhkV7WU2vzZdAu6Nxz/D29cjB470StZdwindvoeZ6X+zJpei/H6fx5DeahJcxg/6CpVYUdofKYjuVEZGEzgE5BwAo9PvGW/0t2hO5sbk/wB4cj9QOK4v4UfELXdQ0DWr/wAaaLa+ETa3svll7oNFJboq/vWc8ADBG4kAhcgAV09lrOn6xp0Op6bfWU1vdqJI5Y5laG6U9CGBwfTcP1xiqqubdp620IoqCXu6X1PNPHvxm8ZaV+0B4f8AD+m+Gftnhm/jilnv/s0shAYnfIJFIRBGADtYEtjjGVrr/id8IdG+OXhWPTdYuLy4torlLlTbT+SUdQRjgdMMwIOTzxg4NQfDf4yeH/GWv+ItJ0+6kkvPDlyLW7VoWVUOXACnGGAKsuR/dz0IJ6i4urG4k37mSX/noisrfnjn6HiqlJxkrLla/q5MYxknd8yf9WG2NuyP9maaeGWEAgB9yuvQMN2T9R2P1BNhYrm1i+e6jdF5ZpIwGx9QQP0rF8V68mkaDdXklw7LZxNMJ0gJlhAGS23GHGOSPlyBiuY+Bv7RWk/tD6TeXVnaahYppkypNBMgcuWG5CCmQVwDx1yB26wqcnFzS0W5p7SKkot6s1fg9J4xvrjW5fGFnpNqsd/INJFkxO+2PIZgSeTx1w3XKjv107ebqluv9wO/0PAH8zR/aq/ww3Lf9siv88Vyfgz4jah4m+K3iTR7nwxq2l2ejxwrb6ncD/R9Q3ZJ2fKBxn+Fm4HO08FO825Jf1sCtFKL/rqdtRRRWZoFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRUN7cGCL5RmRjtQepP+c/QGgDB8T/DfQ/H3ifS9Q1TTYLy68OzedZTOWBhlO08YIDAYU4YEZAPUVrarObW5gdeW2uqj1JxgfnirMUaWVvgt8qjJZj17kn+dcV4+0PxRr/xC8JXWja1b6XotjcSS6naSwbnv0wuAvBx3AyVxu3ckAVpG8tG9Fczl7uqXY66fTtuhyW/3v3ZXI4JOOv1J5qf7Sv2Lzv4dm/8MZqXtWbZy+fooTbIuJWt8MME7ZCmfoQMj1GKzNDN8eaPDrHgK60m7tob6PULVoJopWKJIm358kZI4zgjoSOR1pvgTwXa+BPCdjp2j2cVtY2MflLaltxQgndhzkklsn5ick9RWhfD7ZFNN1VpEt4/93zAGP4nI+iirmmnY1xH/clJ/wC+vm/rV8z5bE8qvcoWU+nRaxcNarbRXbMpvo1VVmyRhHk7ngYBPY8HitO6vo7JF3HLMcKijLOfQDvUerWdvdWjfaIxIqqeO5B6j8emO9cb8CfBniDwTpGo2/ijxB/wkusTXjTreGLyyIGVQiAem5XOBwCxA4ANCinG7YNu9kjrG06TVSjXRaOFG3CBHwG4P3yPvDn7vT/eqDwd4X03wrY3EOl6fY6bbyXDyGK0t0hRm6FiFAGTjrWvVXSCTYqzbcszNwc9WJFK7tYdtS1RRRUjCiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiq+qaxa6HafaLy6t7SDcF8yaQRrknAGScZJwAO5oAsUVmyeMNLhvVtpNQtY7hvuxvIFZ/oD1/Cq3in4kaD4K0S61LVNUs7Ozsk8yaR3zsXcFBwMk5ZlHA6kCqUW3ZITkkrs265Xxb8TdN8H+KNDs75b1pteuJLOx8i1eZS6rucuVB2jsCfQnoCRi6v+0v4VtfiPZ+EUv2fVdTijeGRIz9nHmLvRTJ0DMmGHBHK88gV0kd9AHjkE6PtlWNFj+bamCvAGepP4jFX7Nx1mnqR7RS0gyyniCxuJFa4uo4cHKxyZTH13Y5/QfrUV/4r0qHxJpcEmp6fHNeLKttG1yge4YbMhBnLHHYZq9ca1DHH/q7luccwOM+3IFeP/GD4XzeM/jn4D1pfDslzHpd1m8uPtiwR2aK3mRFwCSxEmGGPvcqeCDTpRjJ2lpv+XyFUlKMbx12Pbqyb13juZIYztkkl3R8dMoAW/A5P4VccXkija9vHyM/IX479x/n1ryHxX+z7f+I/2l7DxtH4lngt9JSMTW3lNiPYhzGrbsBH3ZZcd29RSpRi2+Z20HUlJL3VfX+mepy3lo1/babDcQNNbOjSwLIDJEu1ipYZyASowT1q3H+71aRf+ekasPcgkH+a15v4e+A+n+C/jBfeObJtQuNY1qBzc200oMaqTHu2cA7vlXAYkDBAxkY9Ci1CO/ls7iFtyOWj9CpIyQR2IK4IPQ0pxircrvp+IQcvtIsXv72aGH+829vovP8APbTNTjaIx3Ua7pIM7lA5dD94f1HuPenW37++mk7JiNf5n9ePwqzWZoRC6WSz86NleNk3qwOQwxnIpump5dhCvoi/yqg8a6TbTWaqscHlM1sqjCqMYKAdscYHofatVF2IF9BigBaKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACuA/aC+Cmk/Hjw7Y6Rq019bw292lwr2siq4PKEEMGBBVz2yOorv6z7v8AeOJP71wiD6K3+OaqnKUZKUdyakYyjyy2DTov7AhgsyXa2RVhgd2LMAAAFYnkn0J69+erdc0y28RBdPureK6tHAknilXckij7qsOhBIzg8fLV66WN7aQTbfK2ndu6Y71yHwusPERsdQPiTUIZprq8kuLJLeDyWgs2P7qKQ85dRwSOeeSSaF3B9jhvEWreAtA/aD0XR9Q0/PjlrdY7LU3hLKkOHCb23AGQorqrbSQR1ANd98UfijoHwz8GNq2p30UNjZz26u0eZCuZVA4XJ/8A1VpSeFdNl8ZWt9Jp9nJfWttIkV1JCrzxqzLkByNwBye/c1cvIBrM8luwzbxDDj1c9P8Avkc/Uj0rSUotxvfTfX8uxnGEle1vu/PuU7vxTpuj6fNqmqahZ2MFpEZpDPOiraIByWOcZx1PTsPUt8P+MNF1zTXmg1bS7yOYea5iuo5FCsBtyQem3H1p2uWdr4s8Nf2fqdvb3VrfIY7yKZQ0bIv3wQeMZGPxpsOkabBDbpHZ2tnZ28It7dVhCyGMDARABlUxxgcn2HWNLGmtzP0/4jaXrOgf2jY6tYy+G1SVpNXW4XyYtjBSm7px8wLngbccnmtG1ureXSLWG12XC3gE5ELhwY2+fO7OMHIGc8hs81Q1/wCH+h634DuPDsujWsOgXI2taovkqxL7xtVMENvGeoOaz/hp8IIvBet3F1BqOo/YI7OHTLXShMRY2kcQHKrk/PnILZ5y2ck5q/csyPfukw1rwJrE/wAZtH8THxLdWem2tnJaSaKufstw5DkO7ZHPzA525zEgBwSKj+L3xFsvhN4RvvE06sBZ7We3UjZfPnCqrdA56ZPbOQeK2Pia2qWnhG+j8M2lrceIWhLWazYEaOOVZzkcZGBk4LYB4yR4/wD8LP8AEWu+E4tN8T2ei3V0Y0TUILmxEitKpBIZSxXhh0AxkVMqiilOpttbrbf+md2Ay2tjJypYbe17vbt/SPaPCPi1Nc8I6bfx2l2n9oWsd3sZVJTzFEmC2duRu7HFWLjxJKwj+z26ybmwxZ+FABJJKhlHTuQOevQHx9fjVqyBcpo77fu77dm2/TL8fhTp/jnrdzPHJI2lt5f3VMT7M+pG/BPpnOO2K5/rFPoe2uE8fbW33/8AAL3xxtvib4v8X+F7Pwq8Ol6LLOJb++jMfmQ7XGSyuctHsPCqPnOQ2BjPtA6V4afj5rrTLJu0ncoKg+Q3Q4z/AB+wpzftC6+qk79J4/6YN/8AF1U8ZGUVG23l+ZMeD8cpOWmvn/wD3CiqPhe+n1Lw5Y3Fz5f2ie3SSTYNq7ioJwMn1q9VHz8ouMnF9AorhvjF8QdS8DyacLD7L/pXm7/OjL/d2Yxhh/eNcS37QPiBTy2kj627f/F1lKtGLsz2sHw/isVRVelazvu+zse30V4f/wANDeIE76O3sYG/+OVd0/8AaP1WNh9q0/T517+UzRH9S1L28DafC2PirpJ/P/Ox7HRXC+Hfj7o2rSLHd+dpsrdPNG6Mn/fHA+rYrtre5juolkjZXjYZVlOQw9Qa0jJPY8bE4Ovh5ctaLj6/o+pJRRRVHMFFFFABRRXm/wAV/i7f+E/EcVjpv2P5YRJO00ZfDMTgDDDHAz+IqZSUVdnZgcDVxdX2NHfc9IorzH4Y/GTUvE/itdP1H7H5dxG3ktDGVO8Ddg5Y/wAIb8q9OojJSV0GOwNXCVPZVt7X+QUUUVRxhRRRQBleMPGul+A9IF7q+oWem2zyrAktzKI1aRzhVBPcn9MnoDVi6h8i1tk/uyoD7mq2t6BY+Mk+y6hZ2t9ZQyK4jniWRDKpyrAEdVPII6H6Ut3dfZoFW4PzW8iuz/3lGfm/Tkev1FVpbTcnW5PqJ+3XUdoPu/6yb/dB4X8T+gapr6IoizRjc8JJwOrKeo/r9QKj0xfs9s9xOQkk58yTcfuDoF/AAD65Pej+24HOIy059IlL/qOPzNIZXuLwR6k0yfvM26iMA/fLNx+eBU6Tx6RbpGzGSV8thRl5GPJOPr+ArlNGm8RS+P8AWV1CwtLHQ7dIf7KnhczXEoYN5vmJyF2kkDsAeh6jrdNjtkjZ4WWQsfnkLbmY+56/h2q5RsKMrnnHg3SfH178cfETa61hH4NMYfTI4TGZFclCvAG7OA5ffkbsFeMEeki3t9MjaTaAf4nPzM349TTPNWHUbhmYKqxoST0A+emNPGZVmuWEa5/dRN97646lvbt9aJycne33CjHlVr3HEMiSXlwp/dqWSP8A55j/AOKP6dB3JiiumtoorOPa92y737rHnks31OcDqfwJHD/tCal8QP8AhHNPbwDY2s90bxftYuxHkxYJGFcgbS3DH7wGNo6kdxpWkyRwM0t1JJ5zB2CYUZwM4YfMfqT+Q4ocbRUr7/1qEZXk1bYsRW0djbtlss/Lu55c+/8Anivnn4hNnxZrZH/P3L/6Ea+iBpsFvukWNfMIwXPzMR9Tz3NfO/xF/wCRs1z/AK+5f/QjXHifhR9nwb/vE/8AD+qPStPstR0rx3cWt3ofh9vDM1mkseoPtF1JdFguyVQpG3aMBsY4UZ5wOoXwsvzbdP0+Pn5ct5gIwOoKcc54B7D6C+sSy3cUbqrLLbEMCMggFf8AGoLZW8L7YZJJJLEn5JZHLtBk/dYnnbzwe3APGDXVKz2R8j7Sa+0/vOZ8N2euJ421qLWtA8OrocSQ/wBm3Fg5kuJThvMMsTINvOANpbp37dVb6LpV7HujtbJ16HEK8H0PHB9qs7tuq/8AXSL/ANBP/wBlRe2cLAyv+7ZR/rFbaw/H09jxQ7MI1Jr7T+8njjESBV+6vAHpTq5Vda8RJ8RLGzjsLOfwvNaO8uoSS+XcrcA/LGIu6lRnOBnnpgBuqolGxnGVzyv9pD/XaL/28f8AtOrn7P8ApFrqHg66ee3gmZb1wGeMMQNieoqn+0h/rtF/7eP/AGnWr+zn/wAiVd/9fz/+gR1y/wDL7+ux9lOTXD8Gu/8A7czsT4Y08jmxsz9YF/wrPv8A4Y+H9SRhNo+n5b+JIVjf/vpQD+tb1Fb2T3PlY4irF3jJr0bPLPGP7PapbPNokzRyLz9mmbcreyueR/wLI9xXHeCvH2p/DrUWjVZWt43K3FlMSu099oP3G/Q988EfQh5rzD9oDwSn2Vdet0VZIdsV0B/GhOFY+4JAz6Ef3RWFSnb3oaH1GUZw8Q/qWP8AfjLRN736a/k90+p6D4d1638TaPBfWsnmQ3C7lPceoI7EHgj1FXq8d/Z/8UNYa7caTI/7m+UzQj+7Iv3gPqvP/AK9irWnLmjc8PNsA8HiZUem69H/AFb1QUUUVZ5o2WVYYyzHaq8knsK+e4Fb4l/Eld24rqV5uIPUQqM/+i1xXrfxm8Q/2D4CvArbZrwfZYyOuXznH0UMfwriv2dvD63eu3uosoKWsQt4z/tNycfRQP8AvqsKnvSUT6vJf9lwVfHPe3LH+vVr7jmfEtvJ4A+JNw0KhfsV6LqJRwDGSHCj2wSv4GvoCwu0v7OOeNg8cyh0Yd1IyD+VeV/tG6F5Wpafqar8siG1kPoRlk/m/wCVdN8CtfOr+Bobd23Taaxtm5/hGCn/AI6QPwNFP3ZuI84/2rLqGNW691/16r8TtKKKK3PkwooooAy49TmQww2dhcSRtkedKREiY9Qx3nPqFPNcH8T/AIR+I/HnxP8ACuq2niqTSbPRZPOu7CCN9lwodW7NhiwGw7ug5AznPpkdwGQk/KV+8PSmWKllaZvvTcjPZew/z6mtKdRwfNEiVNSVmZ1rbWseoqt3aKt0i4jnkzIr7jyEZsnPyglTz069a2AKZPbpcxMkiq6sMEMMg1U2z6V93fdW/wDdJzJH9D/EPY8/XpUblkk/yapbt/fR0Pv0I/kadd2EE37yRdrKP9Yp2so+o5qG5u47hLeaNg6pMoyPfK4P/fXSkaddZnaONle3hYrKVOQzDqn4dx+HrTEcsLfXm+Jr3M2owt4W/s1SloIP9ME4kOJS+OU25468g4yM112mRWrW6zW3lyJMoZZVbf5inkHd3B61VlnifXZtsn761ERkX+6rbwP5k/hVlLX+zEzbxhYclmhRcYJOSQPUkkkd/r1JSuKMbFuuc1TVtY0vxzp9ra6bDcaHdQytdXP2gLLbyjlFVD94Nzn69RjB6C3uFuYleNtyt0NZ+tW02oGeG3kWG4+z7onI+VJN2VJ9sgUR03CWpcjvI7uFtjcgcqQVYfUHkV87fEX/AJGzXP8Ar7l/9CNfQioupWMc0kbQylM4B+eI9xn26fhXzz49hZPFGsR7mlYXUq7mA3Odx5OABk+wFc2J+FH2XBv+8VP8P6o9u8Z+OtJ+HNla6preoW2m2AjMJmnbapdim1R3JOCcDsCegNbS6jDeW6tH/pEcqBlKDergjjkcYNea+PfGfg34iWWn6fqqw6lpUcrPdQXNhK4YeUyrt+Xg7iOR2yOhNa0/xs8PrD5NvdTQW8SBVCWkqswHRV+XCjtn8gODW/tKdlrqfO/2Vjr/AMGVv8L/AMjnv2i/GXijwZ4fsbHwdb38uq39yIkkgs1vHsIQNz5UnaMgAKX4HIOOCO8+GzalrXgnSbzXo1XWHt0NzEGUrFKODwvy7vXBIByAcVz+m/Fnw1YXUcn2yUnY28rZzY3EjAHy5wBu69cknkms3X/2grDw/wCKNGt9F03V9aj169Szm+z2rpFYu5AWRyy8Z5PHZGPato1FUSpwtfv1ZzVsvxNBOvWhKMdN00l07d2dlrHxD0S18Yw+G21K1XxBJB9tgst372SNc5I7cgNxnOATjAzXRI4dQy8g8g1zUngbStS8Tpr32GzbxNZwmCO8khAmjQ5+TP8AcOWHGeGODyar+OtR8QR/DXUJPCkdk+vWpVYIr07Y8CRS6k5A3eVuCkkLu2k8VNouyj+P9bHNzSV3L8P63OZ/aQ/12i/9vH/tOtX9nP8A5Eq7/wCv5/8A0BK5f4vzarL4f8K/23Daw6t9lc3gtnLQ+dti8zZnnbuzjrx3NdJ+zrcxx+ELqNpEEjXzkLn5j8idq4/+XzX9bH2VT/knqfr/AO3M9DooBzRXQfIhVPxDpK69od1Zv9y6ieM+2QRn8OtXKD0oKjJxakt0fNfg7UZNI8U6VdN8pt7uPzMdlLBX/wDHSa+k1bcK+aPEkf8AZ+u6qi/8u97Oq+22RgP5V9KWr+Zbo394Zrnw/VH2HFyUnRrL7Sf6P9SSiimzSiGJmbhVGST2FdB8aeP/ALQ/iH7b4gs9PVh5dlGZZBn+N+mfooz/AMCrtvgt4f8A7C8AWe5ds15m6f8A4H93/wAdCj8K8hnEnxI8fn72NWvQBjqsOcD8ox+lfREMSwxKqKFVQAAOgFc9H3pOZ9Znn+y4KhgVvbmfr/w7f3HN/F7QP+Eg8A6hGq7poU+0R+u5Pm4+oBH4151+z/4h/s3xfNaFl8rU4fl93Tkf+Ol/0r2l1Drhhke9fOmqQP8AD3x5KI87tLvPNjX+9GCGUfihA/Git7slMeQf7ThK+AfVXXr/AMOkfRtFR2V2l/aRzRsGjmUOjDowPINSV0HyWwUUUUAec/Hr4Oa18VrrQJNH8U3vhsaTd+fcLAHxdL8uM7WXJXBADZU72yPXtZGvocszRbRz+7iLf+zA/lWjVW81DyX8uKNppiRlVI+QH+Jieg/U9ga0dRtKL6ExglJyXUprqclv5MU2o2Mc8gOFeExtIQMsVVnzx174HenG8uXH7udJ/wDrnbEj8y2P1rL8RfD7Sdc8V+H9a1KzS81bR55BYTFmUWhkQh9qg4OQo+8D6jFb11ebpWt7do/tGAW7+UDnBI98HHrg+hqXboCv1POvj14K8ZeOvBs1j4P1q10HXGkWVrli8KsgBwrOgcqd21uASQpGQDz2fw80ubw54Q0zS72aO51KxtIo7udE2LdTBR5kwH+2+5vXLHPNa1jZrp9okQaSTYAC7nLyH+8x7k1Uvb61vWEcczNcDf5bwjeUK8HnkcE4IPWqdRuKiKMEpcw+K1E97fFl2mQKhPfABx+WTU2mahHqVmskcscwyVYowYBgcMOO4OQR2IrmfAPiDXfE8msjVNKXQ47LUHtoJVuFnN/EiqBMAMiME5+U5I6e9N+GPw20HwPobWOl6XbWdrdyNqGzmTLynLcsSey98AYFDile+41JvVbHSywNbytNbjluXj6CT3Ho3v379iOO8c6rafEKz8TeGdH8SWuneI20raVhnH2uw3bsOyKd6/eXJ4I3jByRXZNotnj/AI9bcf8AbMVyPh74C+GdE+JGp+MLWzli1nVomhmfzm8sBiu4qmcKW2Jkj07ZOSm4rV79PXzJqKTVl8/TyKf7PPw+1r4X/Da10XWtXXUruMtMksZZ40jJGI1Z/mIXg5IHLdO580+ILeX4u1o/3byU/wDjxr1bVfBS+NdS8M3x1DWbGHRZzOiW120aXzGIgrIOrRjHTIzk9uvk/wAQkWLxTrSqoVVupQAO3zGsMbLm957t6n2HBUeWvOK2UV+aOs0z4GQ6lqN4i6tL5NqyRZWEM7ysu8jqBgKyfXk8DFM8KfAC61UXzX2qQosVwUt1to23GLAKly3G45/h4xznnA9OisJNP0bbG6pcTDy4iFysZb+LHHOOTz24wMVD4mvv+EThuNQjheRbOyLiCHmS5EeT5Sr0LHIC88tx0Jpxow2seS8/zDf2r+5f5HDQfs/2cjy+Zql9Csb7AxhUqcAZyRwOSR+Fdb8Ovhhb+BkvfLvZL6PUFRWDqFwF3dMeu6rHwz8WP428C6fq8Wl6lpq6mhufs+pRrDcRbmJw6qzYP49MVsNpjSvuLrC3cwjaT9T3/EU/ZxizOtnOMxFJ06k7xfkvXsMjg+0MYpGZbi3+5KPvFT0P49COmR9K8u/aM/aDg/Z+trO4ns/7Sm1TzI2gikESShNnzlyDsIDEYwd3rgEjqPi/qmteFfCFzc+HbGDX/ECbfsdjcNjz8uofPIBAXLYyvKjnsbdro174y0axbX9E02OSNIZ/sh8uYW0+wF8OQw+ViygqAeOpzXRTUYtTnqu17M8eo204w0fexwfxZ8Vp448MeEtXit7i1h1Sza8jjmAEirIkLAEAnkZxW78CLKbUPBtwi/Z1i+3OWMkfmNnZH0B4/Gs74/W32V9HHkxw58/JVyzN/q+pIz/OqXwA1rxLZ+Jbi3On2X/CGeU8kl+X/fpeZUbMbs7NmCSVAGc7u1ccY3ru3Y+yrO3D1O/83/tzOz+Cnwqi+DNjqWkx6pq2rLe3b6is2oTeZIu8KpQfTaCT3L57129U9V/dCG6X/l3b5v8AcPDflwf+A1crecnJ80tz5GEVFcsdgoJwKKjuLhbWFpHbaqqWJPYAZqSj5r8YnzfEOubeQ19dY9/3r19KWQ22kY9FFfNGnbte1m3T+PUbpVA95JB/8VX0zGNqVz4fqz7Liz3Y0KT3Sf8A7b/kOrlPjN4g/sLwFeKrbZr0C1j/AOB8N+S7j+FdXXjv7Qmvfb/EVnp6t8tjEZZAO7vwAfooz/wOtKsrRbPDyPC/WMbCD2Tu/Ra/i9PmR/s96CL3xVcXzrujsIPLT2d+/wCCqw/4FXs1eU/CHx/4d8F+ElhvNQEd5cStNMvkSNtPRRkKR90D8Sa6r/hePhf/AKCn/ktN/wDEVNJxjG1zuzzD4vE42c40pNLRe69l8uru/mdZXjv7Q+gfZfEVnqEa7UvITBIQOjpgj81Y/wDfNdt/wvHwv/0FP/Jab/4iuZ+LPxB8OeM/B01va6gJLyF1mhBglXLDgjJXHKlhz60VJRlFq4ZJh8XhsZCpKlJLZ+69n8um/wAjf+BWv/2x4Fht2bdLprtbN9Byn/jpA/A12deK/ALXTpfjGazZv3epRHaMfxpyP/HS/wCQr2qqpSvE5M/wvsMdOK2l7y+f/BuFFFFaHjHmPgb4o+LfEfxI8VaXrfhmXRdD0qXbYX6q2bpdxA+Zjtbco37l4UfKeea9AsruKGJRHDMEYFtyrvDdOcgnJNcv8Pfjf4Z+KE17c6LqHnQ2F1Jp1wZYmhxKmGBG4DcpDEhhwfrxRr3w1k1/4h2Gv2Wu3VvY20DRXOmwqjQXh+ba2f4T853cEPtQHG0VtUXvWa5fv/rUxg2o3T5vu/rQm8S/FKzsfGmh6DDb311qGrieS3kS2Y2qGJDuEknAB+boM9MHGRnMv/ho3iv4h6bfXFzcpHpKPPIYpdokuSybDtA7KjAnJOAFyAMV1NzdyLafZWSCZSvHlja0Q/vNH2xjseoHAqpHrT6F4fvLqCL7cyRvLHD5myUsF+SNs9C2ANxHBJJB5NJO3w7lW/mLkWj3lg1w0k0mrRzSFxHKVjMS44RQAFbHPJAJzyTUo1aF41toQ1rcMdixumxl9SB0OOTxkZrH+D/jLXfHPgCz1PXdB/4R/Up2cSWTzFjGFYgH7oIyBnBGRXIftM/F7WPhZ4csb7TPDv8Ab32y6NswMLzxwKBnlV5y7DCnoNvckCnGjKVT2a3/AK+QpVYxh7R7f18zv4/9H8OXgj+UyzSRJjsTIYx/StK4jW1EDpwsJCkf7J4/Tg/hXi+seN/HkPinQ49G0F20m6ggnmgkUzNBcyXQ85JH6xiGHceRy5Xgn5T6pJL525biG4mDfLly+059iqrROm42bHCopOxp3epQ/wCrWTe7HBVAXYDvwM1keIv7S8U6Je22kyJYSNDJHFczDcBLtIU7RncobBI4zjHrXG6h+0Vodx8PLzWtJi1TWrPT7g2M0Gm2pa6llVwjoicHbk5L9Ao45PHeaN4x0u70S1uFuILWOaJHWKZ1ikiDAEIyk/Kwzgr2IxQ6coatBzxlomYnw30LxJ4V8CaPZ+JNUsdW1e33LPcxRFFf7+0DAGcLgZ2jOOleN+M5LiXXtYa6jijuDdz71jbcq/O2MHvxivV/EPxy8P2/xBsfDC6gj+ILxTJbWyozLgo55cAqGIViF68e9eV+Owy+I9YDL5bC5l+XOdvzHiuXGc1k5K19e33H2HBXL9YqKLvaNvxW/me2ReN7DU/GH9iWd1a3GrafbefLbmTDQhsLvI6kDOOO7Y45xrf2f9mvba4mkaaZnKl24CAqeFHQDOPf1JrN0/wFo2leIH1K202ytdS1SFY7i+ggWO4mKAbQzgZOFGMHjCgdAKh+JnjGT4beAdU1m6tZtQh0iA3eIOJJNhB2kds45YZAGTgYro5U2lA+Qu0m5HVdKz7jU5LyVoLPazKcSTMMxxe3+03sOncjoeD+B/xqk/aH8CJrEOmzaVD9oktpYRcCTzGXaTtlwuUwwGQuchhxjJ1fHegyfErwdq3hnSvEEehXhVYHl0x99xp54YbtrKVDAYx8pIbg0/ZOM+WenfyF7RSjzQ17eZu6Tp1vY3kl15kzBmYK0khdpmOAzY/DChQABuwAGrQVri72lR9njI5DDMh/oPxz+FY/g/w3Nofh+002XVL64vLC1S3kupgjTT7V2+bkqc5xnnPOQc81qTIbcjdeT7m6KFQk/QbaiW+hcdjzf9oq3WCTRdvU/aMk8k/6vqa1v2dOfBV2P+n5/wD0BKxf2g1kV9H8xpGz5+N+3P8Ayz9AP61pfs+qZPCNwouJIWN7JgBVw3yR+oNcv/L7+ux9jU/5J+Hr/wC3M7iWA6WrKqmSzYENGBkxD2Hdfbt29Km0mbzLNV3bvL+UN/fA6N+IwfxoFhICD9qmPttT/wCJryX4G/AjxF8G/iN4n1zWPFUmuabrTsY4WXDJmQuJZDgDcoO3jPDE5AGK7IxTi23r+Z8XKTUkkv8AgHslcf8AG3xMvh/wNcxqw+0X3+jRD/e+8fwXP4kV0Wv+I7Pwxpj3l9cJbwLxubufQDqT7DmvBvHfjS5+IXiNZ/LlWPPk2lsPmYA+w6ux649hzjNc1Wooq3U+iyDK5YnEKpJe5F3b6O3T/Py+Re+DOg/214/tW2/udOBuXOOAeiD/AL6Of+Amvea5P4R+A28E+H/9IVft94RJcEfwH+FM/wCzz+JNdZTpQ5Y2Znn2PWKxblD4Y6Lz8/m/wsNmk8qJmJChRkk9AK+dbmeb4g+OJJI2/eatdgRkjPloTgHH+ygz+FewfGjxAdB8BXYVts17i1j+r/e/JQx/CuB/Z/0Aah4ykvCuY9Nh+U/7b5Uf+Oh/zFZ1vekoHq5B/s2Dr4972sv68219xqf8M1uf+Ywv/gH/APZ0f8M1yf8AQYX/AMBP/s69WorT2MOx5f8ArFmP/Pz8I/5HlP8AwzXJ/wBBhf8AwE/+zo/4Zrcf8xhf/AP/AOzr1aij2MOwf6xZj/z8/CP+R83zC48CeLySd1xo91ngbfMCnPT/AGl/Rq+jLO6S+tI5o2DxzKHRh0ZTyDXjf7QWgfYPGEN8q7Y9Rg2uR/z0TA/9BK/lXa/AvxAda8DRQu26TTXNqcnnaMFP/HSB+BrOj7snA9bPv9qwNHHrfZ/15NP7zs6KKK6D445XwZqXh/WtBur3w7Lo9xZnUJHmfT/LMTShtshbZwXI5JPJyD6VqX/gTR9TO6XTbPzMk+YkYjkBPJO5cN+tcX4z+BjeH/gtrXh/4cm18K6pfZktrkO4EcjMm9i3zMCUUqDzt+XAGK1fBOn61pHg7T7HVtam1XXbeILcvbojIW9CxUHgYG5iC2CcZOK2kk/ei+vXcxi5X5ZLp8jifC/7MLfD347at4wHijUZrPVo5E+wyjzWRGKFhJNIWdkULtVRjG5f7vPpOoBdUZWkWKVpZQkaJndBFgF8lc5Y4PTAG4D1J5/wDo/iq88d+If+Ekl0dtNtZE/sgWYZpgjbiTKXzkgbR/vBscYq54w8JTeJvFmkTR61rFlDoNwJpbe2kVY9SZlP7iUY5TABPoCfYi6jcpe8+n6aE04qMfdXX9Tl/jB4o8deHdZ0WLwPpqahZ3lyx1F7pFZVUbT8uxtwBG/LbScgYySa9Oju1u8W9m2FjAEjj/lmMAgD/aIIPPQHJ7A818UfCV7468EX+i6brFxoepXhjB1K2B3WxVg+1cEHlQRgMCFJOcnnR8P+F28NaXYWN3qV/feTAkZuZJTG00ir8zMFxyxBb6k+1RKScF3X3/MuKak+w7QbGRpFlhxHJDAoIYcSFiWIbv0289fr0qp4++LOi+AtFa61rUotGtY5FinlkyzRs3RVCgkk4JyBgAEnGKyvh5o9jF4M0+4+wxw3mpW6SmM5dYVIz9xiRnOcDt7KONnUvhhpfivRVsdW0+zvrEFGEFxAkiBlOd5GPmcnqSMeg5OUuXm97Yfvcvu7mvpcFuLVJlWOO2xugRBxhud3HUtnPfr6k5xviH8INB+MGkRWPiLTY7yxhuFuY4/MeFhIoZQxaMg9HYYzghjkemf8IPEmu6pb3EHijRYNBure5lt7JFuxcC6iRiA4YDA+XbgHnk8cZPd0nzQlo9R+7OOq0MG++Gnh+78Y23iSbSrN9csYTDBfMn72JCGGAfozDPYMR0NeG+P183xfrS/d3XkvX/eNfQt5PG5bzHVbeHmVicDPpn9TXMXPgjRfEk895NpFraxSSM0txcRBZZznlhn7oPq3J9BwawrRlUSV9j3chzSngKsqk4t8y6W7nG3P7QuoS26rHptjHJHgo7TMwUj2wM8ZHUdaSX4/XVyWMulWkjMpUZuWwoIxwNvv16/yrr7rwN4ba0YWOg2dxKUIjd4sRlscZJwWGcZ25rA+Cnw3uP8AhD44vHWj+HTr3nSKGsocRPGDhQT03cHpjjHfJpeyrct+Zfqdn9oZLzW+ry+9/wDyRnW/x6ltrWOFND01Yo12qiy7VA9gEwPwrnfCfinS/A/irXta03w7b2+o+IpRNfSDUZiHIycKvRRlmPHdvQAD1J/h/wCH47eFV0SxknUHKCMbpWGQFH1I5PQAHOM15n+zoutfEKTXh4w8DaboMdhMiW7pprxEk7tyYdyXC4X51GDu/KoU8RyykpadfP8AzM5ZjkfNGLwzv01en/k2h2niPwbfftCfDLSpLXX9W8J3C3huWlsJSzuqeYgjLZVtu7a/XG6NcgjIPoK281mSyrDL3JJKufqec/pWdpmmaXo0C2ttfeVBbx7hbRSqioGbO7CgHls/masDTUvj+7tVWM/8tbgF2P0Vuf8Avr8jWik+VKR8/W9m6sp0lZN6Lsui+487/aFv0vJ9HC/KyeeCNynH+r9Ca57wR8Up/BGizWMdjb3Uck5n3vMUIJCjjAP93rXsmo+BdK1pIhfWcN6Yc7DMu7bnGcdh0HAwOKzLb4X+HUuWgl0awY/ejYx/fX/EHj8jXPKnLn5os+mwec4OOBjg8TTckrvTbdvun1OJT9o2/SPB0y1bHc3J5/8AHao6x+0FrV/byRpFplrHICCShkOD/vHH5ivTR8KPDa8/2Lp//fqs74b/AAd0PwXpTfZkvLpp7hrpZr24aeZM/dVWPIULjA/E5OTQqc7ayJ/tLKYO9PDX9Xp+Lf5HkVzBqfiSe0uNSult4riRbW2utSnFtbl2ztjj3YBJwcKgJOOnFev/AA9+ENl4JdbqRhe6jggzsMCPPBCDt6Z6nnnHFWfiB8HPDnxQtbOHXNMivo9PuVu4Azsu2QAjkqQSpB5U5B7jiunrSNGEUmtWceOz7E4iLoxShT6KP6/5Ky7oKKKKo8U8d/aG1/7b4jtNODfu7GPzn5/jc4A/BR/4/XWfAXQRpHgVLhl/eajIbgn/AGfup/46M/8AAq3NT+HWh61fSXV3pdpcXExBeR0yzYAA/QAfhWtaWsdjaxwwxrHDCoREUYCqBgAfQVlGm1NyZ7mJzSnLL4YKkmrat9/6b/IkooorU8MKKKKAOL+PGg/2z4DkmVcyadILpfXaMhv/AB1ifwrifgBr/wDZvjCWxLfJqkfyj/pomW/9BL/kK9murWO9t5IZlWSORSjqw4YHgg1j2Hw20HSr2K4t9Ks4Z4W3RyImGU+1ZypvnUke9hM1pwwFTBVk3zaq3R/8Orm5RRRWh4JmyaXcaoF+1ztGisG8mBigOOzNwWHsMA9wavW9tHaRLHFGkaL0VVwBT0dZFDKQysMgjvSSPsjZj0UZNAHO634r03wLp8us6xfWum6cjSLNcXEgjjU78Jkn1xgDuSKNMvVmj3q+6e4zLI8IMvkBucDaCC2AB6AAe2eV+P8A8P8AVPid8IodH0vUrTTbiSSO6uTc2a3UcsKAsylGBH3irDjkrjgEkdrcyT6bYiC1ktmuGjPkIISctjhiN33c4JP9cVtyrlT6malLmae2hJFOUmRLe1kEdupP7xgoye5zk568kZ5rk/jhpviHxz8LtW0vw/fR6Xql8ixQXMUrIU+dd+2UAFTs3DcoyCRgg4q38I9B8U6X4GA8eahp+pa0ssjyyWabYCn8OQETcQB/dAxjgnJO1dvJPrunQ7drTF7hwf4IowAF+pd4yfofQVL9yWmtgXvx10v95g/A7wZ4h8DfDqxtfEV3b6tra7/PnEjMQpclE3kZfauFyQOntmuxF88f+st5lHquGH6c/pVlm2jJ4A5rh9W+Mc9r8WrfwpaeHdWvftWmvfJqartswy78Rl8Y5KgZzwXQY5yDWpJyt5h7sIpfI3DDa6vpckMkqxS/aJHjZ+GRg7YIBx/nivK/jf8Atf2vwb05bexjsfEGtQ3P2e7sxehW08AEkyFVYnJ2hTgfe55GG7P4Za54g1zwQtx4s0eXw/etPLmzgkWTeC5KklWYruJOAG6YyeeMz4d/s7+GfBPxO1bxhayXH9oaizxtE84a2haVkd9gxnJcYG4nHIAGeNqfs4yftdbdF1+f+RlUdSUV7PS/V9PkdN4csbnxZpWm6pdwixaWNLuKCaPdLZs6ZwQcASAMVJIJ69OladrocFl8ima+uFbcZrpzM0ZP14H+6uPwq4zPqCFYmaKL/noPvH/d/wAfy9ae6Ku2GMbc8tjsP8TXO2dCWg20g3y72beF/ix99u5+noPrUaqn2q8hkj/d4WbJAKkEYx65BQnp3H4XUUIoUdBxXJ/FD4iaH8PlsW1rVLbTE1aUWERlJHmMxB7A4AGcscAZGSM0Ri5OyFKSSuzO+B3w9vPhl4euF1HxBqfiRtUunvEur5SJbdHC4iIYsQOMnkDcT8q9K7K5/wBF1KKT+GceS315K/8Asw/EVLHLbanbsitDcRMCrKCGUjoQe34VS1a3aDS5oVfy49n7mXdt8hhyuTg8Agc44olJylzSCMVGKijlT8GtFsfjJqXjC2t7pdcu7SOK4eOdgHBCoWC/3tkMQ9CE6ZJNddHFeRxq8F1FdRkZAmTBI/3l4/8AHTTNDea4vLiaZfLaSOIeXj7hC5YZ78tjpTrm8Tw/K0kzLHYsC7zO4WO3PuSeAefofqKqUm9GEYpbFLxb8QLfwB4ZvtX1qGe1sdNhaeeWMecoUegX5vbkCjwp4tsPif4O0/XNHmMlrexi4tZWQqe4IYHkdwR9farkoh16B/PVH0/BUrIuUnHQkg8FPTsc56YqGGPSzbxNY3FvbqqBImtWXbtHAG0fKQMYxjjtil7tvMPev5F19RWTSZZh8rRo25T1Rh1B/Gp7KH7NZwx/880C/kK8a/aT/aLl+Akekn+xW1j+3JWjkKytboQmzplWzK4JAXodvXjFey2k/wBptY5NkkfmIG2OMMuRnBHqKcqcoxU2tHt8iY1Iyk4rdb/MkooorM0CiiigAooooAKKKKACiiigAooooAKKKKAKP9mtYSSSWrbfMbe8LE7GOAOP7vTtx1OMkmsD4nfF3Qfhn4Tkvdev10yOZvsyB43dzIykgBUDE9CcgEYBOaT4ZeM9a8YeE7O41rSk8O6rNv8AMs5SzsAGIDLnbwQAcdRnmofi38HNH+L+h2un68s10kdwJIWjKxtC2CCVOCR8uRg5/lWsYpTtU262M5OTheG/S4eI7W9+JHw5vl8M6x/Z11rFqpstUjXekMZwUK+u5SeRyu885AFa/wAN9DuPD/gzT7W+vG1LUreBIby8ZNrXUqDazkdskHAqz4dsLfwnpFlpMUMdva2cKW9sEG2PYoAVfYgDp37Vyfx3k8VQfD7WY/B22PXGaB7d/kLFHcLIEDcbwATlhjDHHNOPvfu1or9f1Yn7vvve3T/I7U/6fdf9MYT/AN9uP6D+f0rB8GeP9F+IVzf32g31rrC2cv2B3gfKQOoDMGPbO8HjORtIyMVR+EsOuWPw40mPxowGtxQkXThl8nhjt3bPk3bdue2c4qzoXw40P4e2ksvh3SbC1sbuZrq6t7WJQszMBmVcfxAADA4KgAdBUuMVdPfpbYak3Zrb8TpFsPObdcN5p6hMfIv4d/xz+FJdMItSgLfcaOQHJ4/hP9DSW9tDdQLJDNL5bAMGWUlSD6ZOK5Dxl4N8SeJPiF4cu9K8R/Y9D0mVzqtnJCG/tAEDCAgDOOhzwN2eSMUoxu9XYqUmldI6+2Q30v2hh8i58lT6f3j7n9B9TXD/ABa+D2j/ABz8MWGl6tHcNHb3P9qqbeXymLASBVLYPDeaR7YJBBANd5qcpg06Zl+8EO369qq6dLFZ27XDttWYhIxjJKLwoA6nPLYH96nCTi+aO4SipLlexdW6VrVZV+ZZBuX3zUUl5Fp6jzGzLIchVG5nPsOtcP4C1TxlP4q8S2/iay0vR9BhvhHoE9sd81xAQSWkJdgp5UDcqnIYAEbTXeWmnw2eTGvzN95ydzP9SeTUyjyuwRldXIR9qvj/AM+cX4NKf5gfr+FY+t+B9H8U61aQahptnqEemkXqfaYxMUnzhHy2TuADc+9dJWfov7+e8uP+e0xVT/sphf5hj+NEZNaoGk9GS3eh2l8+6S3haT++UG78+tZ+tabDZ2ohjvtQs5Lk+XEIn81nbBOAJAw6A+gA7jrWndX2yTyoVEkx5xnhB6se38zRZ2H2dzI7eZMwwzkfoPQe38zzQM8j/Zs+A2s/AG11oax4hm1SPVrhZQYtxSEruzI+/J3vkbiOPlGSeteg/EL4YaX8UvBl5oerfap7DUAol2XDK3ysrqR24ZVPTHHpXRkZri/i547m+C/gm41q10jUddihdENhZJukG5gu5cA4UZ5GCOnTmtfaVKlTm+0/kZezhTp8v2fvOg8PeENN8JeH7HTrS3jjtNMt47WDf87JHGoVQWPJwAOTUyzvffLar5cP/PYjr/uDv9Tx9ai0xZNfsLe6uopIVmjWUWki7WiyAdsg7sO46AjvjNadZNu+ppG1tCvY6ZDp8Cxx7mxjLSOXdj6ljkk+5qwOKKKkoKKKKACiiigAooooAKKKKACiiigAooooAKKKKAKyWsd1bvFIiuquRgj8R/OuO8W6L4xHxL8Ny6LqWnx+GbUSvq1tcpvuJgRhdjbSe5x8y4OclhwO0hOy7lX+9h/6f0FRmZYbi6mY7UiQBj6YBY/oaqMmmTKN0c7pvxh8N+JPHmreE4rzzNa0eFZry1aF/kVgh4ONrEb0yATjcKyfib8SLrwBq/huO30DWtdh1bUFsnktoW/0NTtO5yV55wRnHCtlumd7RfD1q2vX181rDDqF4VEtxGgWYhUUqpcckKGxg8cVqXDSQxPFdL50LgjzUHIH+0P6j8hWl4KW2hNpNbkkuoskTf6HdSYB4Cr83sOapQaVb6ezNY2U1izncwhRFRz6lc7SffGfer2n34ex3SMrGP5WcdH9CPrwfxxTtst713Qxen8bf4fz+lZl6ni/iP456/4N/aNsPBtroVxc6brSxMZwvEDNuLyxKBjauMuGc4IJG3o3sBu7i0tkW206WQAgYaVFwO5zk5/rU8kax3ltGo2qAzYHsMf1q1VVJRaVlbTXz8yKcZJu7vr9x5b+05Y/ELV/CNnD4FWOK7e5K3RS4SOQRmNgrKz4Aw5DcHOQOCMg9p4Rs9S0zQNP/taOO81WO1jS7ngYbWlCAPtB2hVLZ4AH0re60USqXgoWWn3gqdpOV3r9xTGqwyQ/vo5olYYYSxEAexPT9aggb7Mc2UsdxF3g8wHA/wBg9v8AdPH+7WnUU9jDdf6yGOT/AHlBqDQrza1GNOuJkDF7dCzRkYcHGQCPf9ah05JIbCGzhb/UqElnxxu749TnPsPfpXOfFP4D6T8WP7H+23WqWf8AYt6l9F9jufL8xl/hbIPy+64YdiMmu0jQRoFACgdgOBVPlSViVe+qG2tolpHtQd8knlmPqT3NSUUVBQUUUUAHSiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKAIX+W+j/2kIP6f/Xqkf8ASZ4Yf4ZZXlf3CNgD89v4A0UUxMXTzhRJ/E1zKD7jcw/oPyrSoopAtjzvxR8KLXUfjPovjb7dqVvNoaPpwsoZQttP5xH7xlxncCwzzyAM9OfRKKK0lJtK4oxSbsVm51Zf9mI/qR/hVmiisygooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKAP/Z"}}},{"cell_type":"markdown","source":"# Exploratory Data Analysis","metadata":{}},{"cell_type":"markdown","source":"## Imports  \n\n*******************************","metadata":{}},{"cell_type":"code","source":"import pandas as pd\nimport numpy as np\n\nimport os\nimport warnings\n\nfrom sklearn.preprocessing import StandardScaler\nfrom sklearn.preprocessing import OneHotEncoder\nfrom sklearn.ensemble import GradientBoostingClassifier\nfrom sklearn.metrics import classification_report, confusion_matrix, roc_auc_score, accuracy_score\nfrom sklearn.model_selection import train_test_split, cross_val_score\nfrom sklearn.linear_model import LogisticRegression\nfrom sklearn.ensemble import RandomForestClassifier\nfrom sklearn.model_selection import GridSearchCV\nfrom sklearn.metrics import mean_absolute_error, mean_squared_error\n\n\nimport seaborn as sns\nimport matplotlib.pyplot as plt\nimport matplotlib.patches as mpatches\nimport pickle","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:26.718903Z","iopub.execute_input":"2023-04-18T08:24:26.719904Z","iopub.status.idle":"2023-04-18T08:24:28.545712Z","shell.execute_reply.started":"2023-04-18T08:24:26.719837Z","shell.execute_reply":"2023-04-18T08:24:28.544579Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Settings  \n\n**************************","metadata":{}},{"cell_type":"code","source":"# Pandas defaults\npd.options.display.max_colwidth = 100\npd.options.display.max_rows = 500\npd.options.display.max_columns = 100\npd.options.display.float_format = '{:.2f}'.format\npd.options.display.colheader_justify = 'left'","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:28.547670Z","iopub.execute_input":"2023-04-18T08:24:28.547992Z","iopub.status.idle":"2023-04-18T08:24:28.554167Z","shell.execute_reply.started":"2023-04-18T08:24:28.547961Z","shell.execute_reply":"2023-04-18T08:24:28.552473Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# others\nwarnings.filterwarnings('ignore')","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:28.555532Z","iopub.execute_input":"2023-04-18T08:24:28.555854Z","iopub.status.idle":"2023-04-18T08:24:28.567621Z","shell.execute_reply.started":"2023-04-18T08:24:28.555822Z","shell.execute_reply":"2023-04-18T08:24:28.566442Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Constants  \n\n*****************************************","metadata":{}},{"cell_type":"code","source":"#PATH_REMOTE = ''                    # remote path to data\n\nCR = '\\n'                                     # new line\nRANDOM_STATE = RANDOM_SEED = RS = 66          # random_state\nTEST_FRAC = 0.1                               # delayed sampling fraction\n\nN_TRIALS = 10                                 # number of tries for fitting of hyperparameters\nN_CV = 4                                      # number of folds during cross-validation\nMAX_ITER = 1000                               # max number of iterations for LinearRegression\nDEGREE_POLYNOMIAL = 5                         # degree for polynomial expansion","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:28.571185Z","iopub.execute_input":"2023-04-18T08:24:28.571761Z","iopub.status.idle":"2023-04-18T08:24:28.579743Z","shell.execute_reply.started":"2023-04-18T08:24:28.571714Z","shell.execute_reply":"2023-04-18T08:24:28.578775Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Data preparation  \n\n***************************************************************************","metadata":{}},{"cell_type":"markdown","source":"## Read and Check data\n","metadata":{}},{"cell_type":"code","source":"# Daily, Defog, and Tdcsfog data:\n\ndaily_df = pd.read_csv('/kaggle/input/tlvmc-parkinsons-freezing-gait-prediction/daily_metadata.csv')\ndefog_df = pd.read_csv('/kaggle/input/tlvmc-parkinsons-freezing-gait-prediction/defog_metadata.csv')\ntdcsfog_df = pd.read_csv('/kaggle/input/tlvmc-parkinsons-freezing-gait-prediction/tdcsfog_metadata.csv')\n\n# Other three dataframes:\n\nevents_df = pd.read_csv('/kaggle/input/tlvmc-parkinsons-freezing-gait-prediction/events.csv')\nsubjects_df = pd.read_csv('/kaggle/input/tlvmc-parkinsons-freezing-gait-prediction/subjects.csv')\ntasks_df = pd.read_csv('/kaggle/input/tlvmc-parkinsons-freezing-gait-prediction/tasks.csv')","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:28.580889Z","iopub.execute_input":"2023-04-18T08:24:28.581259Z","iopub.status.idle":"2023-04-18T08:24:28.640588Z","shell.execute_reply.started":"2023-04-18T08:24:28.581225Z","shell.execute_reply":"2023-04-18T08:24:28.639345Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"daily_df.head()","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:28.642383Z","iopub.execute_input":"2023-04-18T08:24:28.642734Z","iopub.status.idle":"2023-04-18T08:24:28.668942Z","shell.execute_reply.started":"2023-04-18T08:24:28.642702Z","shell.execute_reply":"2023-04-18T08:24:28.667790Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"defog_df.head()","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:28.670266Z","iopub.execute_input":"2023-04-18T08:24:28.670630Z","iopub.status.idle":"2023-04-18T08:24:28.681597Z","shell.execute_reply.started":"2023-04-18T08:24:28.670596Z","shell.execute_reply":"2023-04-18T08:24:28.680656Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"tdcsfog_df.head()","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:28.682902Z","iopub.execute_input":"2023-04-18T08:24:28.683923Z","iopub.status.idle":"2023-04-18T08:24:28.698185Z","shell.execute_reply.started":"2023-04-18T08:24:28.683887Z","shell.execute_reply":"2023-04-18T08:24:28.697324Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"events_df.head()","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:28.699218Z","iopub.execute_input":"2023-04-18T08:24:28.700079Z","iopub.status.idle":"2023-04-18T08:24:28.717703Z","shell.execute_reply.started":"2023-04-18T08:24:28.700042Z","shell.execute_reply":"2023-04-18T08:24:28.716446Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"subjects_df.head()","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:28.722282Z","iopub.execute_input":"2023-04-18T08:24:28.722667Z","iopub.status.idle":"2023-04-18T08:24:28.737351Z","shell.execute_reply.started":"2023-04-18T08:24:28.722634Z","shell.execute_reply":"2023-04-18T08:24:28.736305Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"tasks_df.head()","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:28.738807Z","iopub.execute_input":"2023-04-18T08:24:28.739774Z","iopub.status.idle":"2023-04-18T08:24:28.751254Z","shell.execute_reply.started":"2023-04-18T08:24:28.739729Z","shell.execute_reply":"2023-04-18T08:24:28.750375Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Checking the content of dataframes","metadata":{}},{"cell_type":"code","source":"dfs = [daily_df, defog_df, tdcsfog_df, events_df, subjects_df, tasks_df]\n\noutput = \"\\n\".join([f\"{df.__class__.__name__}: {len(df)}\" for df in dfs])\nprint(output)","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:28.752460Z","iopub.execute_input":"2023-04-18T08:24:28.753100Z","iopub.status.idle":"2023-04-18T08:24:28.764494Z","shell.execute_reply.started":"2023-04-18T08:24:28.753064Z","shell.execute_reply":"2023-04-18T08:24:28.763175Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def show_info(daily_df, defog_df, tdcsfog_df, events_df, subjects_df, tasks_df):\n    data_frames = [(\"daily_df\", daily_df), \n                   (\"defog_df\", defog_df),\n                   (\"tdcsfog_df\", tdcsfog_df),\n                   (\"events_df\", events_df),\n                   (\"subjects_df\", subjects_df),\n                   (\"tasks_df\", tasks_df)]\n    for name, df in data_frames:\n        print(f\"\\033[1m{'*'*10} {name} {'*'*10}\\033[0m\")\n        df.info()\n        print(f\"{'*'*26}\\n\")","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:28.766662Z","iopub.execute_input":"2023-04-18T08:24:28.767166Z","iopub.status.idle":"2023-04-18T08:24:28.774974Z","shell.execute_reply.started":"2023-04-18T08:24:28.767131Z","shell.execute_reply":"2023-04-18T08:24:28.773865Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"show_info(daily_df, defog_df, tdcsfog_df, events_df, subjects_df, tasks_df)","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:28.776331Z","iopub.execute_input":"2023-04-18T08:24:28.778128Z","iopub.status.idle":"2023-04-18T08:24:28.837461Z","shell.execute_reply.started":"2023-04-18T08:24:28.778093Z","shell.execute_reply":"2023-04-18T08:24:28.836181Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def remove_spaces(df):\n    \"\"\"\n    The function removes extra spaces in the names of the dataframe columns\n    \n    :param df: pandas.DataFrame, dataframe to process\n    :return: pandas.DataFrame,dataframe with column names without extra spaces\n    \"\"\"\n    df.columns = df.columns.str.strip()\n    return df","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:28.838814Z","iopub.execute_input":"2023-04-18T08:24:28.839740Z","iopub.status.idle":"2023-04-18T08:24:28.845245Z","shell.execute_reply.started":"2023-04-18T08:24:28.839705Z","shell.execute_reply":"2023-04-18T08:24:28.843745Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"daily_df = remove_spaces(daily_df)\ndefog_df = remove_spaces(defog_df)\ntdcsfog_df = remove_spaces(tdcsfog_df)\nevents_df = remove_spaces(events_df)\nsubjects_df = remove_spaces(subjects_df)\ntasks_df = remove_spaces(tasks_df)","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:28.847042Z","iopub.execute_input":"2023-04-18T08:24:28.847587Z","iopub.status.idle":"2023-04-18T08:24:28.859169Z","shell.execute_reply.started":"2023-04-18T08:24:28.847532Z","shell.execute_reply":"2023-04-18T08:24:28.858075Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":" ","metadata":{}},{"cell_type":"markdown","source":"### Now I will sequentially merge the dataframes into a new dataset. I'll remove the extra columns. And on the new dataset I will conduct a feature study and train the model","metadata":{}},{"cell_type":"code","source":"defog_df = defog_df.drop('Medication', axis=1) #Remove the column with lower values\n\ndf = pd.concat([daily_df, defog_df, tdcsfog_df, events_df, subjects_df, tasks_df], axis=1, join='inner')","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:28.860487Z","iopub.execute_input":"2023-04-18T08:24:28.861080Z","iopub.status.idle":"2023-04-18T08:24:28.883209Z","shell.execute_reply.started":"2023-04-18T08:24:28.861046Z","shell.execute_reply":"2023-04-18T08:24:28.881930Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df = df.loc[:,~df.columns.duplicated() | df.columns.isin(['Medication'])]","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:28.884624Z","iopub.execute_input":"2023-04-18T08:24:28.885676Z","iopub.status.idle":"2023-04-18T08:24:28.894819Z","shell.execute_reply.started":"2023-04-18T08:24:28.885638Z","shell.execute_reply":"2023-04-18T08:24:28.893678Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#  deleting columns'Completion' and 'NFOGQ'\n\ndf = df.drop(['Completion', 'NFOGQ', 'Subject'], axis=1)\n\n# column binarization 'Init', 'Sex' , 'Medication'\n\ndf['Init'] = df['Init'].apply(lambda x: 1 if x > 0 else 0)\ndf['Sex'] = df['Sex'].apply(lambda x: 1 if x == 'M' else 0)\ndf['Medication'] = df['Medication'].apply(lambda x: 1 if x == 'on' else 0)","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:28.896866Z","iopub.execute_input":"2023-04-18T08:24:28.897748Z","iopub.status.idle":"2023-04-18T08:24:28.910946Z","shell.execute_reply.started":"2023-04-18T08:24:28.897647Z","shell.execute_reply":"2023-04-18T08:24:28.909776Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**'Init' implies FOG commit I will make it a binary value and then I will compare with other features as a feature of fixed FOG**","metadata":{}},{"cell_type":"code","source":"display(df)\ndf.info()","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:28.912377Z","iopub.execute_input":"2023-04-18T08:24:28.913308Z","iopub.status.idle":"2023-04-18T08:24:28.962114Z","shell.execute_reply.started":"2023-04-18T08:24:28.913269Z","shell.execute_reply":"2023-04-18T08:24:28.960836Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### Let's look at the values of our data again:\n\n>**Id:** The data series in which the FOG event occurred  \n>**Visit:** Laboratory visits consist of baseline evaluation, two post-treatment evaluations for different stages of treatment, and one follow-up evaluation    \n>**Beginning of recording [00:00-23:59]** test recording start   \n>**Test:**  Indicates which three types of tests were performed, from simple (1) to complex (3)   \n>**Medication** - Subjects may have been taking medication for parkinsonism during recording  \n>**Init** - FOG precedent \n>**Type** - FOG type: \"StartHesitation\", \"Turn\" and \"Walk\"   \n>**Kinetic** - Whether the event was kinetic ( 1) and involved movement, or akinetic ( 0) and static \n>**Age** - Patient's age  \n>**Sex** - Patient gender  \n>**YearsSinceDx** - How many years since the diagnosis of Parkinson's  \n>**UPDRSIII_On** - Assessment according to the Unified Parkinson's Disease Assessment Scale when taking/withdrawing treatment, respectively  \n>**UPDRSIII_Off** - Assessment according to the Unified Parkinson's Disease Assessment Scale when taking/withdrawing treatment, respectively \n>**Begin** - Time (since) the start of the task execution  \n>**End** - \nTime (since) task completion  \n>**Task** - One of the seven task types in the DeFOG protocol described on this page https://www.kaggle.com/competitions/tlvmc-parkinsons-freezing-gait-prediction/overview/additional-data-documentation  ","metadata":{}},{"cell_type":"markdown","source":"**Exploring the possibilities for combining features and objects, I decided to stop at the option with fewer features but with a full set of objects**\n\nI also decided to abandon the personalization of the study and focus on the relationship of physical signs and manifestations and remove the initialization data","metadata":{}},{"cell_type":"markdown","source":"### Data visualization","metadata":{}},{"cell_type":"code","source":"# Creating a correlation matrix\n\ncorr_matrix = df[['Age', 'Medication', 'Sex', 'YearsSinceDx', 'UPDRSIII_On', 'UPDRSIII_Off', 'Sex']].corr()\n\n\nsns.heatmap(corr_matrix, cmap=\"PuBuGn_r\", annot=True)\n\n\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:28.963725Z","iopub.execute_input":"2023-04-18T08:24:28.964129Z","iopub.status.idle":"2023-04-18T08:24:29.501931Z","shell.execute_reply.started":"2023-04-18T08:24:28.964093Z","shell.execute_reply":"2023-04-18T08:24:29.501002Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"The correlation between the features is moderate, this should be noted when training the model in order to avoid overfitting. It is also interesting to see the ratio of parameters **UPDRSIII_On** and **Sex**","metadata":{}},{"cell_type":"markdown","source":"I take the Init indicator as a basis for comparisons - since it is an indicator of the FOG event start time, and in accordance with its value it was defined as True (1) and False (2) in fact, the sample we received is a positive FOG and 65 cases have a full data set","metadata":{}},{"cell_type":"code","source":"# Creating a data subset where 'Init' is 1\n\nsubset = df[df['Init'] == 1]\n\n# Counting the number of visits with value_counts()\n\nvisit_counts = subset['Visit'].value_counts()\n\n# Building a bar chart\n\nfig, ax = plt.subplots(figsize=(5,3))\nsns.barplot(x=visit_counts.index, y=visit_counts.values, color='PaleTurquoise', ax=ax)\nax.set_xlabel('Visit', fontsize=14)\nax.set_ylabel('Count', fontsize=14)\nax.set_title('Number of visits when FOG is true', fontsize=16, fontweight='bold')\nplt.show()\n","metadata":{"scrolled":true,"execution":{"iopub.status.busy":"2023-04-18T08:24:29.503269Z","iopub.execute_input":"2023-04-18T08:24:29.504187Z","iopub.status.idle":"2023-04-18T08:24:29.696290Z","shell.execute_reply.started":"2023-04-18T08:24:29.504148Z","shell.execute_reply":"2023-04-18T08:24:29.695390Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"> Laboratory visits consist of baseline evaluation, two post-treatment evaluations for different stages of treatment, and one follow-up evaluation. That is, most of the patients received a score of 1","metadata":{}},{"cell_type":"code","source":"# Counting the number of visits with value_counts()\n\nvisit_counts = subset['Test'].value_counts()\n\n# Building a bar chart\n\n\nfig, ax = plt.subplots(figsize=(6,4))\nsns.barplot(x=visit_counts.index, y=visit_counts.values, color=\"#FFADAE\", ax=ax)\nax.set_xlabel('Test', fontsize=14)\nax.set_ylabel('Count', fontsize=14)\nax.set_title('Which of the three types of tests was performed', fontsize=16, fontweight='bold')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:29.697820Z","iopub.execute_input":"2023-04-18T08:24:29.698425Z","iopub.status.idle":"2023-04-18T08:24:29.897585Z","shell.execute_reply.started":"2023-04-18T08:24:29.698389Z","shell.execute_reply":"2023-04-18T08:24:29.896401Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"> Approximately 25 patients were able to complete both trials 1 and 2.  \n> Approximately 16 patients completed the most challenging 3 trials","metadata":{}},{"cell_type":"code","source":"# Counting the number of visits with value_counts()\n\nvisit_counts = subset['Medication'].value_counts()\n\n# Building a bar chart\n\nfig, ax = plt.subplots(figsize=(6,4))\nsns.barplot(x=visit_counts.index, y=visit_counts.values, color=\"#808080\", ax=ax)\nax.set_xlabel('Medication', fontsize=14)\nax.set_ylabel('Count', fontsize=14)\nax.set_title('Medication for parkinsonism during recording', fontsize=16, fontweight='bold')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:29.899138Z","iopub.execute_input":"2023-04-18T08:24:29.899471Z","iopub.status.idle":"2023-04-18T08:24:30.085992Z","shell.execute_reply.started":"2023-04-18T08:24:29.899439Z","shell.execute_reply":"2023-04-18T08:24:30.084871Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"> Most of the patients were taking drugs for parkinsonism at the time of the study","metadata":{}},{"cell_type":"code","source":"# Counting the number of visits with value_counts()\n\nvisit_counts = subset['Type'].value_counts()\n\n# Building a bar chart\n\nfig, ax = plt.subplots(figsize=(6,4))\nsns.barplot(x=visit_counts.index, y=visit_counts.values, color=\"palegreen\", ax=ax)\nax.set_xlabel('Type', fontsize=14)\nax.set_ylabel('Count', fontsize=14)\nax.set_title('FOG definition type', fontsize=16, fontweight='bold')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:30.087489Z","iopub.execute_input":"2023-04-18T08:24:30.087867Z","iopub.status.idle":"2023-04-18T08:24:30.292651Z","shell.execute_reply.started":"2023-04-18T08:24:30.087826Z","shell.execute_reply":"2023-04-18T08:24:30.291429Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"> Which type of FOG occurred more frequently: \"Turn\"","metadata":{}},{"cell_type":"code","source":"# Counting the number of visits with value_counts()\n\nvisit_counts = subset['Kinetic'].value_counts()\n\n#Building a bar chart\n\nfig, ax = plt.subplots(figsize=(6,4))\nsns.barplot(x=visit_counts.index, y=visit_counts.values, color=\"#D2E8D2\", ax=ax)\nax.set_xlabel('Kinetic', fontsize=14)\nax.set_ylabel('Count', fontsize=14)\nax.set_title('Was the event kinetic or akinetic', fontsize=16, fontweight='bold')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:30.294166Z","iopub.execute_input":"2023-04-18T08:24:30.294477Z","iopub.status.idle":"2023-04-18T08:24:30.483373Z","shell.execute_reply.started":"2023-04-18T08:24:30.294436Z","shell.execute_reply":"2023-04-18T08:24:30.482063Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"> In my sample, all FOG cases included movement","metadata":{}},{"cell_type":"code","source":"# Counting the number of visits with value_counts()\n\nvisit_counts = subset['Age'].value_counts()\n\n# Building a bar chart\n\nfig, ax = plt.subplots(figsize=(10,4))\nsns.barplot(x=visit_counts.index, y=visit_counts.values, color=\"#C7D8EA\", ax=ax)\nax.set_xlabel('Age', fontsize=14)\nax.set_ylabel('Count', fontsize=14)\nax.set_title('Age of the subjects', fontsize=16, fontweight='bold')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:30.485421Z","iopub.execute_input":"2023-04-18T08:24:30.486358Z","iopub.status.idle":"2023-04-18T08:24:30.880755Z","shell.execute_reply.started":"2023-04-18T08:24:30.486306Z","shell.execute_reply":"2023-04-18T08:24:30.879374Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"> As we can see, the age of patients does not have a stable increase or decrease, there are groups that stand out, but for example, the dynamics of an increase in an older group is insignificant","metadata":{}},{"cell_type":"code","source":"# Counting the number of visits with value_counts()\n\nvisit_counts = subset['Sex'].value_counts()\n\n# Building a bar chart\n\nfig, ax = plt.subplots(figsize=(6,4))\nsns.barplot(x=visit_counts.index, y=visit_counts.values, color=\"#FFDAB9\", ax=ax)\nax.set_xlabel('Sex', fontsize=14)\nax.set_ylabel('Count', fontsize=14)\nax.set_title('Sex of the subjects', fontsize=16, fontweight='bold')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:30.888668Z","iopub.execute_input":"2023-04-18T08:24:30.889032Z","iopub.status.idle":"2023-04-18T08:24:31.078714Z","shell.execute_reply.started":"2023-04-18T08:24:30.888997Z","shell.execute_reply":"2023-04-18T08:24:31.077558Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"> Men are more frequent patients with the disease","metadata":{}},{"cell_type":"code","source":"# Counting the number of visits with value_counts()\n\nvisit_counts = subset['YearsSinceDx'].value_counts()\n\n# Colors for sectors\n\ncolors = [\"#FFDAB9\", \"#C7D8EA\", \"#D2E8D2\", \"palegreen\", '#FFF0B2','#FFADAE','PaleTurquoise']\n\n# Building a pie chart\n\nfig, ax = plt.subplots(figsize=(10,7))\nwedges, _ = ax.pie(visit_counts.values, colors=colors)\nax.legend(wedges, visit_counts.index, title='YearsSinceDx', loc='center left', bbox_to_anchor=(1, 0.5), fontsize=8)\nax.set_title('Years since diagnosis of Parkinsons', fontsize=10, fontweight='bold')\n\n# Adding text labels\n\nfor i, wedge in enumerate(wedges):\n    angle = (wedge.theta2 - wedge.theta1)/2. + wedge.theta1\n    x = wedge.r * 0.80 * np.cos(np.radians(angle))\n    y = wedge.r * 0.80 * np.sin(np.radians(angle))\n    ax.text(x, y, visit_counts.index[i], ha='center', va='center', fontsize=8, fontweight='bold')\n    \nplt.show()","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:31.080344Z","iopub.execute_input":"2023-04-18T08:24:31.080698Z","iopub.status.idle":"2023-04-18T08:24:31.624815Z","shell.execute_reply.started":"2023-04-18T08:24:31.080665Z","shell.execute_reply":"2023-04-18T08:24:31.623584Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"> 7 years, 10 years and 13 years are the most common times from diagnosis to examination","metadata":{}},{"cell_type":"code","source":"fig, ax = plt.subplots(figsize=(7,5))\n\ncolors = sns.color_palette('Set2', n_colors=len(df['Medication'].unique()))\n\nfor i, (medication, data) in enumerate(df.groupby('Medication')):\n    sns.histplot(data=data, x='Test', bins=25, color=colors[i],multiple='stack',alpha=0.6, ax=ax)\n\nlabels = ['Medication=0', 'Medication=1']\n\npatches = [mpatches.Patch(color=colors[i], label=labels[i]) for i in range(len(labels))]\n\nax.legend(handles=patches,fontsize=15)\n\nax.set_title('One of three tests (Test) with or without medication', fontsize=16)\nax.set_xlabel('Test', fontsize=10)\nax.set_ylabel('Count', fontsize=10)\n\nfor axis in ['top','bottom','left','right']:\n    ax.spines[axis].set_linewidth(1.5)\n\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:31.626322Z","iopub.execute_input":"2023-04-18T08:24:31.630012Z","iopub.status.idle":"2023-04-18T08:24:32.071323Z","shell.execute_reply.started":"2023-04-18T08:24:31.629957Z","shell.execute_reply":"2023-04-18T08:24:32.069981Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"> It was easy to assume that patients who took the drug were more successful on the Test","metadata":{}},{"cell_type":"code","source":"colors = [\"#FFADAE\", \"#808080\", \"palegreen\"]\nsns.set_palette(colors)\n\nsns.pairplot(data=df, vars=['UPDRSIII_On', 'UPDRSIII_Off', 'YearsSinceDx', 'Age','Sex', 'Medication', 'Kinetic'], hue='Type')","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:32.072851Z","iopub.execute_input":"2023-04-18T08:24:32.073813Z","iopub.status.idle":"2023-04-18T08:24:46.218130Z","shell.execute_reply.started":"2023-04-18T08:24:32.073776Z","shell.execute_reply":"2023-04-18T08:24:46.216793Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"![MT.jpg](attachment:MT.jpg)","metadata":{},"attachments":{"MT.jpg":{"image/jpeg":"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"}}},{"cell_type":"code","source":"# Separation of features into numerical and categorical\n\nnumerical_features = ['Test','Medication','Init', 'Kinetic', 'Age', 'Sex', 'YearsSinceDx', 'UPDRSIII_On', 'UPDRSIII_Off']\ncategorical_features = ['Type']","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:46.219760Z","iopub.execute_input":"2023-04-18T08:24:46.220098Z","iopub.status.idle":"2023-04-18T08:24:46.225744Z","shell.execute_reply.started":"2023-04-18T08:24:46.220067Z","shell.execute_reply":"2023-04-18T08:24:46.224608Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Coding of categorical features\n\nencoder = OneHotEncoder()\ncategorical_data = encoder.fit_transform(df[categorical_features])","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:46.227236Z","iopub.execute_input":"2023-04-18T08:24:46.227557Z","iopub.status.idle":"2023-04-18T08:24:46.244655Z","shell.execute_reply.started":"2023-04-18T08:24:46.227528Z","shell.execute_reply":"2023-04-18T08:24:46.243359Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Combining coded categorical features with numeric features\n\nX = pd.concat([df[numerical_features], pd.DataFrame(categorical_data.toarray(), columns=encoder.get_feature_names())], axis=1)\n\n# Target variable\n\ny = df['Age']","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:46.246174Z","iopub.execute_input":"2023-04-18T08:24:46.246511Z","iopub.status.idle":"2023-04-18T08:24:46.264684Z","shell.execute_reply.started":"2023-04-18T08:24:46.246479Z","shell.execute_reply":"2023-04-18T08:24:46.263598Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Separation into training and test sets\n\nX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n\nX_train = X_train.fillna(X_train.median())\nX_test = X_test.fillna(X_train.median())","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:46.265743Z","iopub.execute_input":"2023-04-18T08:24:46.266349Z","iopub.status.idle":"2023-04-18T08:24:46.290494Z","shell.execute_reply.started":"2023-04-18T08:24:46.266313Z","shell.execute_reply":"2023-04-18T08:24:46.289535Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Model 1. Gradient Boosting","metadata":{}},{"cell_type":"code","source":"# Create Model and Hyperparameters\n\ngb = GradientBoostingClassifier()\nparams = {\n    'n_estimators': [50, 200, 500],\n    'max_depth': [3, 5, 10],\n    'learning_rate': [0.01, 0.1, 0.5]\n}\n\n# Finding the Best Hyperparameters\n\ngrid_search = GridSearchCV(gb, params, cv=5, scoring='accuracy')\ngrid_search.fit(X_train, y_train)\n\n# Evaluation of the quality of the model on the test sample\n\ny_pred = grid_search.predict(X_test)\ntest_score = accuracy_score(y_test, y_pred)\n\n# Deriving of results\n\nprint('Best params:', grid_search.best_params_)\nprint('Accuracy (CV):', grid_search.best_score_)\nprint('Accuracy (test):', test_score)","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:24:46.292251Z","iopub.execute_input":"2023-04-18T08:24:46.292821Z","iopub.status.idle":"2023-04-18T08:32:39.475622Z","shell.execute_reply.started":"2023-04-18T08:24:46.292778Z","shell.execute_reply":"2023-04-18T08:32:39.474284Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Model 2. Random Forest","metadata":{}},{"cell_type":"code","source":"# Defining Hyperparameters\n\nparam_grid = {\n    'n_estimators': [10, 50, 100, 200],\n    'max_depth': [2, 3, 5, None],\n    'min_samples_split': [2, 3, 5],\n    'min_samples_leaf': [1, 2, 4],\n    'max_features': ['sqrt', 'log2', None]\n}\n\n# Create a Model\n\nrf = RandomForestClassifier()\n\n# Selecting hyperparameters using GridSearchCV\n\ngrid_search = GridSearchCV(rf, param_grid, cv=5, scoring='accuracy')\ngrid_search.fit(X_train, y_train)\n\nprint('Best parameters:', grid_search.best_params_)\nprint('Best score:', grid_search.best_score_)","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:32:39.477059Z","iopub.execute_input":"2023-04-18T08:32:39.477411Z","iopub.status.idle":"2023-04-18T08:37:25.012676Z","shell.execute_reply.started":"2023-04-18T08:32:39.477370Z","shell.execute_reply":"2023-04-18T08:37:25.011339Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Saving the model using pickle\n\nwith open('rf_model.pkl', 'wb') as f:\n    pickle.dump(grid_search.best_estimator_, f)","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:37:25.014117Z","iopub.execute_input":"2023-04-18T08:37:25.015042Z","iopub.status.idle":"2023-04-18T08:37:25.023141Z","shell.execute_reply.started":"2023-04-18T08:37:25.014994Z","shell.execute_reply":"2023-04-18T08:37:25.021609Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Loading the model from the pickle file\n\nwith open('rf_model.pkl', 'rb') as f:\n    rf_model = pickle.load(f)","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:37:25.024697Z","iopub.execute_input":"2023-04-18T08:37:25.025039Z","iopub.status.idle":"2023-04-18T08:37:25.037836Z","shell.execute_reply.started":"2023-04-18T08:37:25.025006Z","shell.execute_reply":"2023-04-18T08:37:25.036548Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"y_pred = rf_model.predict(X_test)\ntest_ids = X_test.index","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:37:25.039500Z","iopub.execute_input":"2023-04-18T08:37:25.039945Z","iopub.status.idle":"2023-04-18T08:37:25.054365Z","shell.execute_reply.started":"2023-04-18T08:37:25.039910Z","shell.execute_reply":"2023-04-18T08:37:25.053005Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# y_pred_flat_padded = np.pad(y_pred, (0, 3 - len(y_pred) % 3), mode='constant')\n# submission = pd.DataFrame({'Id': test_ids})\n# submission['Id'] = submission['Id'].iloc[:5]\n# submission[['StartHesitation', 'Turn', 'Walking']] = pd.DataFrame(y_pred_flat_padded.reshape((-1, 3)), columns=['StartHesitation', 'Turn', 'Walking'])","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:37:25.055783Z","iopub.execute_input":"2023-04-18T08:37:25.056113Z","iopub.status.idle":"2023-04-18T08:37:25.073231Z","shell.execute_reply.started":"2023-04-18T08:37:25.056083Z","shell.execute_reply":"2023-04-18T08:37:25.071933Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"submission = pd.read_csv('/kaggle/input/tlvmc-parkinsons-freezing-gait-prediction/sample_submission.csv')\n","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:52:30.847132Z","iopub.execute_input":"2023-04-18T08:52:30.847559Z","iopub.status.idle":"2023-04-18T08:52:31.126749Z","shell.execute_reply.started":"2023-04-18T08:52:30.847509Z","shell.execute_reply":"2023-04-18T08:52:31.125588Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(submission)","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:52:34.017974Z","iopub.execute_input":"2023-04-18T08:52:34.019430Z","iopub.status.idle":"2023-04-18T08:52:34.030105Z","shell.execute_reply.started":"2023-04-18T08:52:34.019380Z","shell.execute_reply":"2023-04-18T08:52:34.028222Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"submission.to_csv('submission.csv',index=False)","metadata":{"execution":{"iopub.status.busy":"2023-04-18T08:52:39.132301Z","iopub.execute_input":"2023-04-18T08:52:39.132815Z","iopub.status.idle":"2023-04-18T08:52:39.579739Z","shell.execute_reply.started":"2023-04-18T08:52:39.132764Z","shell.execute_reply":"2023-04-18T08:52:39.577958Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"The result of the model evaluation on the metric was 0.63, which means that the model explains approximately 63% of the variance of the target variable (Age column). But it must be taken into account that we trained the model on an extremely truncated data sample, so overfitting is possible.","metadata":{}},{"cell_type":"markdown","source":"The purpose of my experiment was to see in a small piece of data the relationship between the occurrence of FOG and physical indicators, for example:\n\n>**Age** - we saw that FOG episodes occur at different ages and do not have a pronounced growth pattern, but it is also noticeable that before 65 years of age, the types \"StartHesitation\" (rarely) and \"Walk\" (more often) occur and starting from 65 the \"Turn\" type predominates both with and without therapy\n>**Sex** - Men in the resulting dataset become patients more often, and the density of the number of patients older than 65 increases. But for both genders, more often these are patients over the age of 65\n\n>**Medication** - Non-medicated patients included those whose FOG type was \"StartHesitation\" and \"Walking\". And those who took more often belong to the FOG \"Turn\" type, it is important to note that this type is the most common in both cases. Also, it is interesting that patients with the \"StartHesitation\" type were men not taking medication, and with the \"Walking\" type, women were not taking medication\n\n>**YearsSinceDx** - more often than 10 years passed from the moment of diagnosis to the study in women, the range of male patients is much higher (but we remember that there are more male patients) \n \n>**Kinetic** - In women with the \"Turn\" type, men are more typical with the \"Walking\" type. Those who take the drug more often have the \"Turn\" type, while those who do not take the \"Walking\" type\n\n\nI also made two algorithms that can be used on this set, here I deviated a little from the task since this is an experimental project and made the patient's age the goal, so the models predict in which age groups FOG is more possible   \n\n**Gradient Boosting** - showed a very low score of 0.49 and 0.30 on the test\n**Random Forest** - showed a higher result 0.63","metadata":{}}]}