{"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":"# **Back ground info**","metadata":{}},{"cell_type":"markdown","source":"Ref: https://www.amc.seoul.kr/asan/healthinfo/disease/diseaseDetail.do?contentId=31884  \n'Freezing of Gait' refers to a condition in which the **center of gravity moves excessively** when walking or **the movement of both sides is unbalanced**, making it impossible to walk normally.  \n  \nIt is caused by a disorder in the extrapyramidal system of the brain. Movement disorders such as hand tremor, muscle stiffness, posture disorders, and gait disorders appear. **Typical Parkinson's disease patients have difficulty** bending their bodies and **starting walking**. Once you start walking, **it is difficult to change direction, avoid obstacles, or stop**. During walking, **the movement of the upper extremity or the movement of the torso and pelvis decreases**, and the posture response is also impaired. Therefore, even if the **center of the body shakes a little, it easily falls down**.","metadata":{}},{"cell_type":"markdown","source":"# **Library Load**","metadata":{}},{"cell_type":"code","source":"import numpy as np\nimport pandas as pd\nimport matplotlib.pyplot as plt\nimport seaborn as sns\nimport sklearn\nimport random\nimport os\nfrom sklearn.preprocessing import LabelEncoder\nfrom sklearn.metrics import mean_absolute_error\nimport warnings\nwarnings.filterwarnings('ignore')","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2023-03-13T04:52:21.003189Z","iopub.execute_input":"2023-03-13T04:52:21.003562Z","iopub.status.idle":"2023-03-13T04:52:21.009777Z","shell.execute_reply.started":"2023-03-13T04:52:21.003534Z","shell.execute_reply":"2023-03-13T04:52:21.008626Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"Base = '/kaggle/input/tlvmc-parkinsons-freezing-gait-prediction/'","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:21.016214Z","iopub.execute_input":"2023-03-13T04:52:21.017567Z","iopub.status.idle":"2023-03-13T04:52:21.022383Z","shell.execute_reply.started":"2023-03-13T04:52:21.017516Z","shell.execute_reply":"2023-03-13T04:52:21.021411Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# **Data Load & EDA**\n### **→ Availiable csv files**","metadata":{}},{"cell_type":"markdown","source":"### **1. Subjcets**","metadata":{}},{"cell_type":"code","source":"subject = pd.read_csv(Base + 'subjects.csv')\nsubject.head()","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:21.024899Z","iopub.execute_input":"2023-03-13T04:52:21.025262Z","iopub.status.idle":"2023-03-13T04:52:21.054022Z","shell.execute_reply.started":"2023-03-13T04:52:21.025225Z","shell.execute_reply":"2023-03-13T04:52:21.052938Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"subject.isna().sum()","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:21.055735Z","iopub.execute_input":"2023-03-13T04:52:21.056007Z","iopub.status.idle":"2023-03-13T04:52:21.065364Z","shell.execute_reply.started":"2023-03-13T04:52:21.055982Z","shell.execute_reply":"2023-03-13T04:52:21.063895Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"subject.describe()","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:21.066953Z","iopub.execute_input":"2023-03-13T04:52:21.067253Z","iopub.status.idle":"2023-03-13T04:52:21.094174Z","shell.execute_reply.started":"2023-03-13T04:52:21.067225Z","shell.execute_reply":"2023-03-13T04:52:21.093490Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# replace Nan with 1 in Visit\nsubject[\"Visit_\"] = subject[\"Visit\"].fillna(1)\n\n# replace Nan with 43.0 in UPDRSIII_Off\nsubject[\"UPDRSIII_Off_\"] = subject[\"UPDRSIII_Off\"].fillna(43.0)\n\n# replace Nan with 35.0 in UPDRSIII_On\nsubject[\"UPDRSIII_On_\"] = subject[\"UPDRSIII_On\"].fillna(35.0)\n\nsubject.head()","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:21.095786Z","iopub.execute_input":"2023-03-13T04:52:21.097017Z","iopub.status.idle":"2023-03-13T04:52:21.118005Z","shell.execute_reply.started":"2023-03-13T04:52:21.096957Z","shell.execute_reply":"2023-03-13T04:52:21.117234Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"fig,ax = plt.subplots(nrows=2, ncols=1,figsize=(5,5))\n\nsns.histplot(data=subject, x=\"Age\",ax=ax[0])\nax[0].set_title(\"Distribution of Age\")\n\nsns.histplot(data=subject, x=\"YearsSinceDx\",ax=ax[1])\nax[1].set_title(\"Distribution of YearsSinceDx\")\n\nfig.tight_layout()","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:21.120702Z","iopub.execute_input":"2023-03-13T04:52:21.121553Z","iopub.status.idle":"2023-03-13T04:52:21.467450Z","shell.execute_reply.started":"2023-03-13T04:52:21.121501Z","shell.execute_reply":"2023-03-13T04:52:21.465542Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"fig,ax = plt.subplots(nrows=2, ncols=1,figsize=(5,5))\n\nsns.boxplot(data=subject, x=\"UPDRSIII_On\",ax=ax[0])\nax[0].set_title(\"Distribution of UPDRSIII_On\")\n\nsns.boxplot(data=subject, x=\"UPDRSIII_Off\",ax=ax[1])\nax[1].set_title(\"Distribution of UPDRSIII_Off\")\n\nfig.tight_layout()","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:21.469124Z","iopub.execute_input":"2023-03-13T04:52:21.469523Z","iopub.status.idle":"2023-03-13T04:52:21.726581Z","shell.execute_reply.started":"2023-03-13T04:52:21.469485Z","shell.execute_reply":"2023-03-13T04:52:21.725077Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sns.histplot(data = subject, x= 'NFOGQ')","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:21.728111Z","iopub.execute_input":"2023-03-13T04:52:21.729243Z","iopub.status.idle":"2023-03-13T04:52:21.934808Z","shell.execute_reply.started":"2023-03-13T04:52:21.729206Z","shell.execute_reply":"2023-03-13T04:52:21.933610Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sns.countplot(data = subject, x='Visit')","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:21.936186Z","iopub.execute_input":"2023-03-13T04:52:21.936562Z","iopub.status.idle":"2023-03-13T04:52:22.073477Z","shell.execute_reply.started":"2023-03-13T04:52:21.936528Z","shell.execute_reply":"2023-03-13T04:52:22.072645Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Check the Subjects data by Sex,Age\nsubject[\"Age\"].hist(by=subject['Sex'])","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:22.076516Z","iopub.execute_input":"2023-03-13T04:52:22.078684Z","iopub.status.idle":"2023-03-13T04:52:22.363888Z","shell.execute_reply.started":"2023-03-13T04:52:22.078645Z","shell.execute_reply":"2023-03-13T04:52:22.362497Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Man has more larg boundary for age","metadata":{}},{"cell_type":"code","source":"# Check the Subjects data by Sex,YearsSinceDx(how long from desease)\nsubject[\"YearsSinceDx\"].hist(by=subject['Sex'])","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:22.365205Z","iopub.execute_input":"2023-03-13T04:52:22.366122Z","iopub.status.idle":"2023-03-13T04:52:22.637165Z","shell.execute_reply.started":"2023-03-13T04:52:22.366062Z","shell.execute_reply":"2023-03-13T04:52:22.636294Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Check NFOGQ ","metadata":{}},{"cell_type":"code","source":"# Check the Subjects data by Sex,YearsSinceDx(how long from desease)\nsubject[\"NFOGQ\"].hist(by=subject['Sex'])","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:22.638180Z","iopub.execute_input":"2023-03-13T04:52:22.638611Z","iopub.status.idle":"2023-03-13T04:52:22.914696Z","shell.execute_reply.started":"2023-03-13T04:52:22.638583Z","shell.execute_reply":"2023-03-13T04:52:22.913627Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Grouping Age and Checking wit GroupAge,NFOGQ\nbins = 3\nsubject[\"Group_Age\"] = pd.cut(subject.Age, bins,labels=['0','1','2'])\nsubject.head()","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:22.915863Z","iopub.execute_input":"2023-03-13T04:52:22.916196Z","iopub.status.idle":"2023-03-13T04:52:22.939800Z","shell.execute_reply.started":"2023-03-13T04:52:22.916159Z","shell.execute_reply":"2023-03-13T04:52:22.938700Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Check the Subjects data by Group_Age,NFOGQ\nfig,ax = plt.subplots(nrows=1,ncols=1)\nsubject[\"NFOGQ\"].hist(by=subject['Group_Age'],ax=ax)\nfig.tight_layout()","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:22.942993Z","iopub.execute_input":"2023-03-13T04:52:22.943351Z","iopub.status.idle":"2023-03-13T04:52:23.440994Z","shell.execute_reply.started":"2023-03-13T04:52:22.943319Z","shell.execute_reply":"2023-03-13T04:52:23.439654Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"NFOGQ has no relationship with Group_Age","metadata":{}},{"cell_type":"markdown","source":"Check UPDRSIII_On","metadata":{}},{"cell_type":"code","source":"# Check the Subjects data by Sex, UPDRSIII_On\nsubject[\"UPDRSIII_On\"].hist(by=subject['Sex'])","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:23.442358Z","iopub.execute_input":"2023-03-13T04:52:23.442872Z","iopub.status.idle":"2023-03-13T04:52:23.714424Z","shell.execute_reply.started":"2023-03-13T04:52:23.442839Z","shell.execute_reply":"2023-03-13T04:52:23.713134Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Man has more boundary for UPDRSIII_On  \nIs this situation from age?","metadata":{}},{"cell_type":"code","source":"# Check the Subjects data by Group_Age,UPDRSIII_On\nfig,ax = plt.subplots(nrows=1,ncols=1)\nsubject[\"UPDRSIII_On\"].hist(by=subject['Group_Age'],ax=ax)\nfig.tight_layout()","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:23.715988Z","iopub.execute_input":"2023-03-13T04:52:23.716415Z","iopub.status.idle":"2023-03-13T04:52:24.183622Z","shell.execute_reply.started":"2023-03-13T04:52:23.716372Z","shell.execute_reply":"2023-03-13T04:52:24.182204Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"It seems old person has more large boundary for UPDRSIII_On","metadata":{}},{"cell_type":"markdown","source":"Check UPDRSIII_Off","metadata":{}},{"cell_type":"code","source":"# Check the Subjects data by Sex, UPDRSIII_Off\nsubject[\"UPDRSIII_Off\"].hist(by=subject['Sex'])","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:24.187286Z","iopub.execute_input":"2023-03-13T04:52:24.187772Z","iopub.status.idle":"2023-03-13T04:52:24.463266Z","shell.execute_reply.started":"2023-03-13T04:52:24.187728Z","shell.execute_reply":"2023-03-13T04:52:24.462005Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Check the Subjects data by Group_Age,UPDRSIII_Off\nfig,ax = plt.subplots(nrows=1,ncols=1)\nsubject[\"UPDRSIII_Off\"].hist(by=subject['Group_Age'],ax=ax)\nfig.tight_layout()","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:24.464319Z","iopub.execute_input":"2023-03-13T04:52:24.464624Z","iopub.status.idle":"2023-03-13T04:52:24.947336Z","shell.execute_reply.started":"2023-03-13T04:52:24.464594Z","shell.execute_reply":"2023-03-13T04:52:24.946094Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"It seems age has relationship with UPDRSIII_Off but middle(1) has largest boundary in the group","metadata":{}},{"cell_type":"markdown","source":"#### **2. task**","metadata":{}},{"cell_type":"code","source":"task = pd.read_csv(Base + 'tasks.csv')\ntask.head()","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:24.948519Z","iopub.execute_input":"2023-03-13T04:52:24.948831Z","iopub.status.idle":"2023-03-13T04:52:24.966324Z","shell.execute_reply.started":"2023-03-13T04:52:24.948804Z","shell.execute_reply":"2023-03-13T04:52:24.965310Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"task.isna().sum()","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:24.967851Z","iopub.execute_input":"2023-03-13T04:52:24.968209Z","iopub.status.idle":"2023-03-13T04:52:24.979653Z","shell.execute_reply.started":"2023-03-13T04:52:24.968174Z","shell.execute_reply":"2023-03-13T04:52:24.978410Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"task[\"Between_Begin_End\"] = task[\"End\"] - task[\"Begin\"]\nprint(task.shape)\ntask.head()","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:24.980985Z","iopub.execute_input":"2023-03-13T04:52:24.981359Z","iopub.status.idle":"2023-03-13T04:52:25.000815Z","shell.execute_reply.started":"2023-03-13T04:52:24.981320Z","shell.execute_reply":"2023-03-13T04:52:24.999515Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"task.Task.value_counts()","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:25.002171Z","iopub.execute_input":"2023-03-13T04:52:25.002928Z","iopub.status.idle":"2023-03-13T04:52:25.015058Z","shell.execute_reply.started":"2023-03-13T04:52:25.002885Z","shell.execute_reply":"2023-03-13T04:52:25.013719Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"task.groupby(\"Task\")[\"Between_Begin_End\"].mean().sort_values(ascending=False)","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:25.020812Z","iopub.execute_input":"2023-03-13T04:52:25.021234Z","iopub.status.idle":"2023-03-13T04:52:25.032981Z","shell.execute_reply.started":"2023-03-13T04:52:25.021198Z","shell.execute_reply":"2023-03-13T04:52:25.031940Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# **Data Load & EDA**\n### **→ train data**","metadata":{}},{"cell_type":"markdown","source":"### 1. defog","metadata":{}},{"cell_type":"code","source":"defog_list = os.listdir(Base + \"/train/defog\")\ntdcsfog_list = os.listdir(Base + \"/train/tdcsfog\")","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:25.034265Z","iopub.execute_input":"2023-03-13T04:52:25.035692Z","iopub.status.idle":"2023-03-13T04:52:25.042995Z","shell.execute_reply.started":"2023-03-13T04:52:25.035653Z","shell.execute_reply":"2023-03-13T04:52:25.041818Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"defog_list[0]","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:25.044475Z","iopub.execute_input":"2023-03-13T04:52:25.044838Z","iopub.status.idle":"2023-03-13T04:52:25.057334Z","shell.execute_reply.started":"2023-03-13T04:52:25.044805Z","shell.execute_reply":"2023-03-13T04:52:25.055527Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"defog_df = pd.read_csv(Base + \"/train/defog/\"+defog_list[0])\ndefog_df.head()","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:25.058797Z","iopub.execute_input":"2023-03-13T04:52:25.059889Z","iopub.status.idle":"2023-03-13T04:52:25.189808Z","shell.execute_reply.started":"2023-03-13T04:52:25.059855Z","shell.execute_reply":"2023-03-13T04:52:25.188392Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sns.lineplot(data = defog_df,x=\"Time\" , y= \"AccV\")","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:25.190907Z","iopub.execute_input":"2023-03-13T04:52:25.191203Z","iopub.status.idle":"2023-03-13T04:52:25.525138Z","shell.execute_reply.started":"2023-03-13T04:52:25.191174Z","shell.execute_reply":"2023-03-13T04:52:25.523627Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sns.lineplot(data = defog_df,x=\"Time\" , y= \"AccML\")","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:25.526972Z","iopub.execute_input":"2023-03-13T04:52:25.527286Z","iopub.status.idle":"2023-03-13T04:52:25.874726Z","shell.execute_reply.started":"2023-03-13T04:52:25.527259Z","shell.execute_reply":"2023-03-13T04:52:25.873611Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sns.lineplot(data = defog_df,x=\"Time\" , y= \"AccAP\")","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:25.876125Z","iopub.execute_input":"2023-03-13T04:52:25.876499Z","iopub.status.idle":"2023-03-13T04:52:26.214516Z","shell.execute_reply.started":"2023-03-13T04:52:25.876452Z","shell.execute_reply":"2023-03-13T04:52:26.213446Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Some signal processing skill need...","metadata":{}},{"cell_type":"code","source":"print(\"StartHesitation: \",defog_df.StartHesitation.value_counts())\nprint(\"-----\"*10)\nprint(\"Turn: \",defog_df.Turn.value_counts())\nprint(\"-----\"*10)\nprint(\"Walking: \",defog_df.Walking.value_counts())","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:26.215974Z","iopub.execute_input":"2023-03-13T04:52:26.216286Z","iopub.status.idle":"2023-03-13T04:52:26.228133Z","shell.execute_reply.started":"2023-03-13T04:52:26.216259Z","shell.execute_reply":"2023-03-13T04:52:26.226681Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"defog_df.groupby([\"Turn\",\"Walking\"]).count()","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:26.229673Z","iopub.execute_input":"2023-03-13T04:52:26.229921Z","iopub.status.idle":"2023-03-13T04:52:26.256449Z","shell.execute_reply.started":"2023-03-13T04:52:26.229896Z","shell.execute_reply":"2023-03-13T04:52:26.255755Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(\"Valid: \",defog_df.Valid.value_counts())\nprint(\"-----\"*10)\nprint(\"Task: \",defog_df.Task.value_counts())","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:26.257523Z","iopub.execute_input":"2023-03-13T04:52:26.257783Z","iopub.status.idle":"2023-03-13T04:52:26.267039Z","shell.execute_reply.started":"2023-03-13T04:52:26.257758Z","shell.execute_reply":"2023-03-13T04:52:26.265748Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(\"Turn status 1\")\ndefog_df[defog_df[\"Turn\"]==1].describe().iloc[:,:4]","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:26.268090Z","iopub.execute_input":"2023-03-13T04:52:26.268482Z","iopub.status.idle":"2023-03-13T04:52:26.776299Z","shell.execute_reply.started":"2023-03-13T04:52:26.268421Z","shell.execute_reply":"2023-03-13T04:52:26.775436Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(\"Turn status 0\")\ndefog_df[defog_df[\"Turn\"]==0].describe().iloc[:,:4]","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:26.778264Z","iopub.execute_input":"2023-03-13T04:52:26.778678Z","iopub.status.idle":"2023-03-13T04:52:26.822071Z","shell.execute_reply.started":"2023-03-13T04:52:26.778642Z","shell.execute_reply":"2023-03-13T04:52:26.820939Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"fig,ax = plt.subplots(ncols=1,nrows=2,figsize=(10,5))\n\nsns.lineplot(data = defog_df.iloc[80000:,:],x=\"Time\" , y= \"AccAP\",ax=ax[0])\nsns.lineplot(data = defog_df.iloc[80000:,:],x=\"Time\" , y= \"Turn\",ax=ax[0])\n\nsns.lineplot(data = defog_df.iloc[80000:,:],x=\"Time\" , y= \"AccAP\",ax=ax[1])\nsns.lineplot(data = defog_df.iloc[80000:,:],x=\"Time\" , y= \"Walking\",ax=ax[1])","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:26.823794Z","iopub.execute_input":"2023-03-13T04:52:26.824045Z","iopub.status.idle":"2023-03-13T04:52:27.327530Z","shell.execute_reply.started":"2023-03-13T04:52:26.824021Z","shell.execute_reply":"2023-03-13T04:52:27.326399Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"fig,ax = plt.subplots(ncols=1,nrows=2,figsize=(10,5))\n\nsns.lineplot(data = defog_df.iloc[80000:,:],x=\"Time\" , y= \"AccML\",ax=ax[0])\nsns.lineplot(data = defog_df.iloc[80000:,:],x=\"Time\" , y= \"Turn\",ax=ax[0])\n\nsns.lineplot(data = defog_df.iloc[80000:,:],x=\"Time\" , y= \"AccML\",ax=ax[1])\nsns.lineplot(data = defog_df.iloc[80000:,:],x=\"Time\" , y= \"Walking\",ax=ax[1])","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:27.329378Z","iopub.execute_input":"2023-03-13T04:52:27.329808Z","iopub.status.idle":"2023-03-13T04:52:27.845548Z","shell.execute_reply.started":"2023-03-13T04:52:27.329774Z","shell.execute_reply":"2023-03-13T04:52:27.844371Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"fig,ax = plt.subplots(ncols=1,nrows=2,figsize=(10,5))\n\nsns.lineplot(data = defog_df.iloc[80000:,:],x=\"Time\" , y= \"AccV\",ax=ax[0])\nsns.lineplot(data = defog_df.iloc[80000:,:],x=\"Time\" , y= \"Turn\",ax=ax[0])\n\nsns.lineplot(data = defog_df.iloc[80000:,:],x=\"Time\" , y= \"AccV\",ax=ax[1])\nsns.lineplot(data = defog_df.iloc[80000:,:],x=\"Time\" , y= \"Walking\",ax=ax[1])","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:27.846900Z","iopub.execute_input":"2023-03-13T04:52:27.847292Z","iopub.status.idle":"2023-03-13T04:52:28.354585Z","shell.execute_reply.started":"2023-03-13T04:52:27.847254Z","shell.execute_reply":"2023-03-13T04:52:28.352967Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"defog_df[defog_df[\"Turn\"]==1]","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:28.356096Z","iopub.execute_input":"2023-03-13T04:52:28.356948Z","iopub.status.idle":"2023-03-13T04:52:28.376966Z","shell.execute_reply.started":"2023-03-13T04:52:28.356906Z","shell.execute_reply":"2023-03-13T04:52:28.375743Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Index at 93448, change in Turn status  \nCompare before 93448 and after 93448 to 93552","metadata":{}},{"cell_type":"markdown","source":"#### **Check Turn status change with each value(AccV, AccML, AccAP)**","metadata":{}},{"cell_type":"code","source":"slice_defog_df = defog_df.loc[93343:93352]\nslice_defog_df","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:28.378799Z","iopub.execute_input":"2023-03-13T04:52:28.379150Z","iopub.status.idle":"2023-03-13T04:52:28.395622Z","shell.execute_reply.started":"2023-03-13T04:52:28.379115Z","shell.execute_reply":"2023-03-13T04:52:28.394491Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"fig,ax = plt.subplots(nrows=3,ncols=1,figsize=(10,8))\n\nsns.lineplot(data = slice_defog_df,x=\"Time\" , y= \"AccV\",ax=ax[0])\nax[0].set_title(\"AccV\")\n\nsns.lineplot(data = slice_defog_df,x=\"Time\", y= \"AccML\",ax=ax[1])\nax[1].set_title(\"AccML\")\n\nsns.lineplot(data = slice_defog_df,x=\"Time\", y= \"AccAP\",ax=ax[2])\nax[2].set_title(\"AccAP\")\n\nfig.tight_layout()","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:28.397069Z","iopub.execute_input":"2023-03-13T04:52:28.397398Z","iopub.status.idle":"2023-03-13T04:52:28.865847Z","shell.execute_reply.started":"2023-03-13T04:52:28.397364Z","shell.execute_reply":"2023-03-13T04:52:28.864904Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Nothing dramaticaly change at index 93348","metadata":{}},{"cell_type":"code","source":"# Moving average\ndefog_df.loc[93341:93352][[\"AccV\",\"AccML\",\"AccAP\"]].rolling(window=2).mean()","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:28.867164Z","iopub.execute_input":"2023-03-13T04:52:28.868212Z","iopub.status.idle":"2023-03-13T04:52:28.882408Z","shell.execute_reply.started":"2023-03-13T04:52:28.868176Z","shell.execute_reply":"2023-03-13T04:52:28.880738Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"tmp_df = defog_df.loc[93341:93352][[\"AccV\",\"AccML\",\"AccAP\"]].rolling(window=2).mean()\ntmp_df[\"Time\"] = defog_df.loc[93341:93352][\"Time\"]\n\nfig,ax = plt.subplots(nrows=3,ncols=1,figsize=(10,8))\n\nsns.lineplot(data = tmp_df,x=\"Time\" , y= \"AccV\",ax=ax[0])\nax[0].set_title(\"AccV moving average\")\n\nsns.lineplot(data = tmp_df,x=\"Time\", y= \"AccML\",ax=ax[1])\nax[1].set_title(\"AccML moving average\")\n\nsns.lineplot(data = tmp_df,x=\"Time\", y= \"AccAP\",ax=ax[2])\nax[2].set_title(\"AccAP moving average\")\n\nfig.tight_layout()","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:28.884404Z","iopub.execute_input":"2023-03-13T04:52:28.884832Z","iopub.status.idle":"2023-03-13T04:52:29.366362Z","shell.execute_reply.started":"2023-03-13T04:52:28.884798Z","shell.execute_reply":"2023-03-13T04:52:29.365074Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"Hmm, It seems Moving average about AccML, AccAP has litle bit strange factor with this graph ","metadata":{}},{"cell_type":"code","source":"tmp_df = defog_df.copy()\ntmp_df[\"AccV_delta\"] = (tmp_df.AccV - tmp_df.AccV.shift()).fillna(0)\ntmp_df[\"AccML_delta\"] = (tmp_df.AccML - tmp_df.AccML.shift()).fillna(0)\ntmp_df[\"AccAP_delta\"] = (tmp_df.AccAP - tmp_df.AccAP.shift()).fillna(0)\n\nfig,ax = plt.subplots(nrows=3,ncols=1,figsize=(10,8))\n\nsns.lineplot(data = tmp_df.loc[93341:110000],x=\"Time\" , y= \"AccV_delta\",ax=ax[0])\nsns.lineplot(data = tmp_df.loc[93341:110000],x=\"Time\" , y= \"Turn\",ax=ax[0])\nax[0].set_title(\"AccV delta\")\n\nsns.lineplot(data = tmp_df.loc[93341:110000],x=\"Time\", y= \"AccML_delta\",ax=ax[1])\nsns.lineplot(data = tmp_df.loc[93341:110000],x=\"Time\" , y= \"Turn\",ax=ax[1])\nax[1].set_title(\"AccML delta\")\n\nsns.lineplot(data = tmp_df.loc[93341:110000],x=\"Time\", y= \"AccAP_delta\",ax=ax[2])\nsns.lineplot(data = tmp_df.loc[93341:110000],x=\"Time\" , y= \"Turn\",ax=ax[2])\nax[2].set_title(\"AccAP delta\")\n\nfig.tight_layout()","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:29.368280Z","iopub.execute_input":"2023-03-13T04:52:29.368705Z","iopub.status.idle":"2023-03-13T04:52:30.062275Z","shell.execute_reply.started":"2023-03-13T04:52:29.368661Z","shell.execute_reply":"2023-03-13T04:52:30.060982Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"#### **Check Turn status change with complex(AccV, AccML, AccAP)**\nref : https://www.mdpi.com/347228","metadata":{}},{"cell_type":"markdown","source":"![image.png](attachment:d69344df-c5dd-414e-b6d7-590015c2f06a.png)","metadata":{},"attachments":{"d69344df-c5dd-414e-b6d7-590015c2f06a.png":{"image/png":"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"}}},{"cell_type":"markdown","source":"### **2.Feature Engineering**","metadata":{}},{"cell_type":"code","source":"defog_df[\"Stride\"] = defog_df[\"AccV\"] + defog_df[\"AccML\"] + defog_df[\"AccAP\"]\ndefog_df[\"Stride\"]","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:30.063597Z","iopub.execute_input":"2023-03-13T04:52:30.063878Z","iopub.status.idle":"2023-03-13T04:52:30.076787Z","shell.execute_reply.started":"2023-03-13T04:52:30.063850Z","shell.execute_reply":"2023-03-13T04:52:30.075708Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"fig,ax = plt.subplots(ncols=1,nrows=2,figsize=(10,5))\n\nsns.lineplot(data = defog_df.iloc[80000:,:],x=\"Time\" , y= \"Stride\",ax=ax[0])\nsns.lineplot(data = defog_df.iloc[80000:,:],x=\"Time\" , y= \"Turn\",ax=ax[0])\n\nsns.lineplot(data = defog_df.iloc[80000:,:],x=\"Time\" , y= \"Stride\",ax=ax[1])\nsns.lineplot(data = defog_df.iloc[80000:,:],x=\"Time\" , y= \"Walking\",ax=ax[1])","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:30.077888Z","iopub.execute_input":"2023-03-13T04:52:30.078157Z","iopub.status.idle":"2023-03-13T04:52:30.583739Z","shell.execute_reply.started":"2023-03-13T04:52:30.078131Z","shell.execute_reply":"2023-03-13T04:52:30.582312Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"tmp_df = defog_df.loc[93341:93352][[\"Stride\"]]\ntmp_df[\"Time\"] = defog_df.loc[93341:93352][\"Time\"]\n\nfig,ax = plt.subplots(nrows=1,ncols=1,figsize=(5,5))\n\nsns.lineplot(data = tmp_df,x=\"Time\" , y= \"Stride\",ax=ax)\nax.set_title(\"Stride\")","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:52:30.584916Z","iopub.execute_input":"2023-03-13T04:52:30.585260Z","iopub.status.idle":"2023-03-13T04:52:30.755052Z","shell.execute_reply.started":"2023-03-13T04:52:30.585228Z","shell.execute_reply":"2023-03-13T04:52:30.753925Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def sqrt_df(x):\n    return np.sqrt(abs(x))\n\ndefog_df[\"Step\"] = defog_df[\"Stride\"].apply(sqrt_df)","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:53:16.654732Z","iopub.execute_input":"2023-03-13T04:53:16.655131Z","iopub.status.idle":"2023-03-13T04:53:16.792531Z","shell.execute_reply.started":"2023-03-13T04:53:16.655092Z","shell.execute_reply":"2023-03-13T04:53:16.791347Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"defog_df","metadata":{"execution":{"iopub.status.busy":"2023-03-13T04:53:17.623499Z","iopub.execute_input":"2023-03-13T04:53:17.624210Z","iopub.status.idle":"2023-03-13T04:53:17.645926Z","shell.execute_reply.started":"2023-03-13T04:53:17.624138Z","shell.execute_reply":"2023-03-13T04:53:17.644781Z"},"trusted":true},"execution_count":null,"outputs":[]}]}