{"cells":[{"metadata":{},"cell_type":"markdown","source":"# Diet prediction on Indian food"},{"metadata":{},"cell_type":"markdown","source":"![istockphoto-868408746-612x612.jpg](attachment:istockphoto-868408746-612x612.jpg)","attachments":{"istockphoto-868408746-612x612.jpg":{"image/jpeg":"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"}}},{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"import numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"Dataset : https://www.kaggle.com/nehaprabhavalkar/indian-food-101"},{"metadata":{"_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","trusted":true},"cell_type":"code","source":"import numpy as np\nimport pandas as pd\nimport geopandas as gpd\nimport matplotlib.pyplot as plt\n%matplotlib inline\nimport seaborn as sns\nimport random\nfrom plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot\nimport plotly.express as px\nimport plotly.graph_objects as go\nimport plotly.figure_factory as ff\nfrom plotly.colors import n_colors\nfrom plotly.subplots import make_subplots\ninit_notebook_mode(connected=True)\nimport cufflinks as cf\ncf.go_offline()\nfrom wordcloud import WordCloud , ImageColorGenerator\nfrom PIL import Image\n\nfrom sklearn.preprocessing import LabelEncoder, StandardScaler\nfrom sklearn.model_selection import train_test_split\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Loadthe dataset"},{"metadata":{"trusted":true},"cell_type":"code","source":"df = pd.read_csv('../input/indian-food-101/indian_food.csv')\ndf=df.replace(-1,np.nan)\ndf=df.replace('-1',np.nan)\ndf","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df.shape","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"The dataset consists of about 255 Indian dishes and 9 columns associated with each of them.\n\nThe 9 columns are as follows:-\n\nname : name of the dish\n\ningredients : main ingredients used\n\ndiet : type of diet - either vegetarian or non vegetarian\n\nprep_time : preparation time\n\ncook_time : cooking time\n\nflavor_profile : flavor profile includes whether the dish is spicy, sweet, bitter, etc\n\ncourse : course of meal - starter, main course, dessert, etc\n\nstate : state where the dish is famous or is originated\n\nregion : region where the state belongs"},{"metadata":{},"cell_type":"markdown","source":"All the observations in this notebook will be based on these 255 dishes. There are many more dishes in Indian Cuisine!"},{"metadata":{"trusted":true},"cell_type":"code","source":"df.describe()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Proportion of Vegetarian and Non-Vegetarian dishes"},{"metadata":{"trusted":true},"cell_type":"code","source":"pie_df = df.diet.value_counts().reset_index()\npie_df.columns = ['diet','count']\nfig = px.pie(pie_df, values='count', names='diet', title='Proportion of Vegetarian and Non-Vegetarian dishes',\n             color_discrete_sequence=['green', 'red'])\nfig.show()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"#### Observation :\nVegetarian dishes consists 88.6% out of all the diets(Ofcourse higly imbalanced dependent variable).\n\n#### Fun fact : India is ranked top in the world with 38% of the total population being vegetarians"},{"metadata":{},"cell_type":"markdown","source":"### Number of dishes based on regions"},{"metadata":{"trusted":true},"cell_type":"code","source":"reg_df = df.region.value_counts().reset_index()\nreg_df.columns = ['region','count']\nreg_df = reg_df.sample(frac=1)\nfig = px.bar(reg_df,x='region',y='count',title='Number of dishes based on regions',\n             color_discrete_sequence=['#416219'])\nfig.show()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Number of dishes based on courses of meal "},{"metadata":{"trusted":true},"cell_type":"code","source":"course_df = df.course.value_counts().reset_index()\ncourse_df.columns = ['course','count']\ncourse_df = course_df.sample(frac=1)\nfig = px.bar(course_df,x='course',y='count',title='Number of dishes based on courses of meal',\n             color_discrete_sequence=['#AB63FA'])\nfig.show()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"Observation :\nAbout 129 dishes are preferably eaten as a main course\n\n#### Fun fact : The very favourite Chicken Tikka Masala, a popular dish in India, is not Indian. It was invented in Glasgow, Scotland!"},{"metadata":{},"cell_type":"markdown","source":"### Proportion of Flavor Profiles\n"},{"metadata":{"trusted":true},"cell_type":"code","source":"pie_df = df.flavor_profile.value_counts().reset_index()\npie_df.columns = ['flavor','count']\nfig = px.pie(pie_df, values='count', names='flavor', title='Proportion of Flavor Profiles',\n             color_discrete_sequence=['#FF7F0E', '#00B5F7','#AB63FA','#00CC96'])\nfig.show()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"#### Observation :\nMore than 50% of the dishes are spicy in flavor\n\n#### Fun fact : India is the largest producer of spices in the world. No country in the world produces as many varieties of spices as India.\n"},{"metadata":{},"cell_type":"markdown","source":"### Ingredients used in Indian desserts"},{"metadata":{"trusted":true},"cell_type":"code","source":"dessert_df  = df[df['course']=='dessert'].reset_index()\n\ningredients = []\nfor i in range(0,len(dessert_df)):\n    text = dessert_df['ingredients'][i].split(',')\n    text = ','.join(text)\n    ingredients.append(text)\n    text = ' '.join(ingredients)\n\nwordcloud = WordCloud(width = 500, height = 500, colormap = 'seismic'\n                      ,background_color ='white', \n                min_font_size = 10).generate(text)                  \nplt.figure(figsize = (8, 8), facecolor = None) \nplt.imshow(wordcloud) \nplt.axis('off') \nplt.show()\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"#### Observation :\nMost commonly used ingredients in making Indian sweets are Sugar, Ghee, Milk, Rice"},{"metadata":{},"cell_type":"markdown","source":"### Ingredients used in North-Indian cuisine"},{"metadata":{"trusted":true},"cell_type":"code","source":"north_df = df[df['region']=='North'].reset_index()\n\ningredients = []\nfor i in range(0,len(north_df)):\n    text = north_df['ingredients'][i].split(',')\n    text = ','.join(text)\n    ingredients.append(text)\n    text = ' '.join(ingredients)\n\nwordcloud = WordCloud(width = 400, height = 400, colormap = 'winter',\n                      background_color ='white', \n                min_font_size = 10).generate(text)                  \nplt.figure(figsize = (8, 8), facecolor = None) \nplt.imshow(wordcloud) \nplt.axis('off') \nplt.show()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Dishes"},{"metadata":{"trusted":true},"cell_type":"code","source":"wordCloud = WordCloud(\n    background_color='White',colormap = 'seismic',\n    max_font_size = 50).generate(' '.join(df['name']))\nplt.figure(figsize=(15,7))\nplt.axis('off')\nplt.imshow(wordCloud)\nplt.show()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def state_infograph(statename, title) : \n    new_df = df[df['state']== statename]\n\n    total_dishes = new_df.shape[0]\n\n    course_df = new_df['course'].value_counts().reset_index()\n    course_df.columns = ['course','num']\n\n    diet_df = new_df['diet'].value_counts().reset_index()\n    diet_df.columns = ['diet','num']\n    \n#     flavor_df = new_df['prep_time'].value_counts().reset_index()\n#     flavor_df.columns = ['Flavor','num']\n\n    prep_time_df = new_df['prep_time'].value_counts().reset_index()\n    prep_time_df.columns = ['prep_time', 'num']\n\n    fig = make_subplots(\n        rows=2, cols=3,subplot_titles=('Total Dishes','Dishes by Courses','Dishes by Preparation time', '',''),\n        specs=[[{'type': 'indicator'},{'type': 'bar','rowspan': 2},{'type': 'bar','rowspan': 2}],\n              [ {'type': 'pie'} , {'type': 'pie'}, {'type': 'pie'}]])\n\n    fig.add_trace(go.Indicator(\n        mode = 'number',\n        value = int(total_dishes),\n        number={'font':{'color': '#270082','size':50}},\n    ),row=1, col=1)\n\n\n    fig.add_trace(go.Bar(x=course_df['course'],y=course_df['num'], marker={'color': 'blue'}, \n                         text=course_df['num'],name='dishes by courses',textposition ='auto'),row=1, col=2)\n\n    fig.add_trace(go.Pie(labels=diet_df['diet'], values=diet_df['num'],textinfo='percent',\n                         marker= dict(colors=['#00bd0d','#fc0303'])),row=2, col=1)\n\n    fig.add_trace(go.Bar(\n        x=prep_time_df['prep_time'],y=course_df['num'],marker={'color': '#fc0335'}, text=course_df['num'],\n        name='flavors by courses',textposition ='auto'),row=1, col=3)\n\n    fig.update_layout(title_text= title,template='plotly',title_x=0.5)\n\n    return fig.show()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### West Bengal Food - Mini Infograph"},{"metadata":{"trusted":true},"cell_type":"code","source":"state_infograph('West Bengal', 'West Bengal food infograph')","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Punjab Infograph"},{"metadata":{"trusted":true},"cell_type":"code","source":"state_infograph('Punjab', 'Punjab food infograph')","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Histogram for preperation time"},{"metadata":{"trusted":true},"cell_type":"code","source":"plt.figure(figsize=(14,7))\nsns.distplot(df['prep_time'],color='red')\nplt.title(\"Histogram for preparation time\",fontsize=24)\nplt.show()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### In the above histogram we can see some outliers, lets handle them before moving ahead."},{"metadata":{"trusted":true},"cell_type":"code","source":"# Handling outliers for prep_time \n\n# as our data is skewed so we will compute the Interquantile range to calculate the boundaries \nIQR=df['prep_time'].quantile(0.75)-df['prep_time'].quantile(0.25)\nlower_bridge=df['prep_time'].quantile(0.25)-(IQR*1.5)\nupper_bridge=df['prep_time'].quantile(0.75)+(IQR*1.5)\nprint(lower_bridge), print(upper_bridge)\n\n#### Extreme outliers\nlower_bridge=df['prep_time'].quantile(0.25)-(IQR*3)\nupper_bridge=df['prep_time'].quantile(0.75)+(IQR*3)\nprint(lower_bridge), print(upper_bridge)\n\ndf.loc[df['prep_time']>=50,'prep_time']=50\n\n\nplt.figure(figsize=(14,8))\nsns.distplot(df['prep_time'],color='red')\nplt.title(\"Histogram for prep time after handling outliers\",fontsize=24)\nplt.show()\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Histogram for cook time"},{"metadata":{"trusted":true},"cell_type":"code","source":"plt.figure(figsize=(14,8))\nsns.distplot(df['cook_time'],color='seagreen')\nplt.title(\"Histogram for cook time\",fontsize=24)\nplt.show()\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Lets start building Model"},{"metadata":{},"cell_type":"markdown","source":"### Data Preprocessing"},{"metadata":{"trusted":true},"cell_type":"code","source":"df.isna().sum()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"food_vocab = set()\n\nfor ingredients in df['ingredients']:\n    for food in ingredients.split(','):\n        if food.strip().lower() not in food_vocab:\n            food_vocab.add(food.strip().lower())","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_kg_hide-output":true},"cell_type":"code","source":"food_vocab","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"food_columns = pd.DataFrame()\n\nfor i, ingredients in enumerate(df['ingredients']):\n    for food in ingredients.split(','):\n        if food.strip().lower() in food_vocab:\n            food_columns.loc[i, food.strip().lower()] = 1\n\nfood_columns = food_columns.fillna(0)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"food_columns","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df = df.drop(['name', 'ingredients'], axis=1)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"{column: list(df[column].unique()) for column in df.columns if df.dtypes[column] == 'object'}","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df[['flavor_profile', 'state', 'region']] = df[['flavor_profile', 'state', 'region']].replace('-1', np.NaN)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def onehot_encode(df, columns, prefixes):\n    df = df.copy()\n    for column, prefix in zip(columns, prefixes):\n        dummies = pd.get_dummies(df[column], prefix=prefix)\n        df = pd.concat([df, dummies], axis=1)\n        df = df.drop(column, axis=1)\n    return df\n\ndf = onehot_encode(\n    df,\n    ['flavor_profile', 'course', 'state', 'region'],\n    ['f', 'c', 's', 'r']\n)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df[['prep_time', 'cook_time']] = df[['prep_time', 'cook_time']].replace(-1, np.NaN)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df['prep_time'] = df['prep_time'].fillna(df['prep_time'].mean())\ndf['cook_time'] = df['cook_time'].fillna(df['cook_time'].mean())","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"label_encoder = LabelEncoder()\n\ndf['diet'] = label_encoder.fit_transform(df['diet'])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"{index: label for index, label in enumerate(label_encoder.classes_)}","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"y = df['diet']\n\nX = df.drop('diet', axis=1)\nX_food = pd.concat([X, food_columns], axis=1)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"scaler = StandardScaler()\n\nX = scaler.fit_transform(X)\nX_food = scaler.fit_transform(X_food)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"X_food","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Spliting the data into test and train set."},{"metadata":{"trusted":true},"cell_type":"code","source":"X_train, X_test, y_train, y_test = train_test_split(X, y, train_size=0.7, random_state=42)\nX_food_train, X_food_test, y_food_train, y_food_test = train_test_split(X_food, y, train_size=0.7, random_state=42)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_kg_hide-output":true},"cell_type":"code","source":"#Dependencies\nimport keras\nfrom keras.models import Sequential\nfrom keras.layers import Dense\n# Neural network\nmodel = Sequential()\nmodel.add(Dense(64, input_dim=40, activation='relu'))\nmodel.add(Dense(64, activation='relu'))\nmodel.add(Dense(2, activation='softmax'))\n\nmodel.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n\nhistory = model.fit(X_train, y_train, validation_split=0.2, epochs=5000, batch_size=64)\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import matplotlib.pyplot as plt\nplt.plot(history.history['accuracy'])\nplt.plot(history.history['val_accuracy'])\nplt.title('Model accuracy')\nplt.ylabel('Accuracy')\nplt.xlabel('Epoch')\nplt.legend(['Train', 'Test'], loc='upper left')\nplt.show()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.evaluate(X_test, y_test)\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"y_pred = model.predict(X_test)\ny_pred","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"plt.plot(history.history['loss']) \nplt.plot(history.history['val_loss']) \nplt.title('Model loss') \nplt.ylabel('Loss') \nplt.xlabel('Epoch') \nplt.legend(['Train', 'Test'], loc='upper left') \nplt.show()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]}],"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":4,"nbformat_minor":4}