4432 proj.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"4432 proj.ipynb","provenance":[],"collapsed_sections":[],"authorship_tag":"ABX9TyOSgvXNOblB5axR6D+SEaey"},"kernelspec":{"name":"python3","display_name":"Python 3"},"accelerator":"TPU"},"cells":[{"cell_type":"code","metadata":{"id":"-PvkbdgYNZeZ","executionInfo":{"status":"ok","timestamp":1607831662873,"user_tz":-480,"elapsed":1873,"user":{"displayName":"Gordan Yeung","photoUrl":"","userId":"16744045141325193456"}}},"source":["import numpy as np\r\n","import pandas as pd\r\n","import json\r\n","import matplotlib.pyplot as plt\r\n","\r\n","\r\n","from sklearn.preprocessing import MultiLabelBinarizer\r\n","from sklearn.linear_model import LinearRegression\r\n","from sklearn.linear_model import Ridge\r\n","from sklearn.ensemble import GradientBoostingRegressor"],"execution_count":2,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"oHYlmaXRNgQ0","executionInfo":{"status":"ok","timestamp":1607831665205,"user_tz":-480,"elapsed":1464,"user":{"displayName":"Gordan Yeung","photoUrl":"","userId":"16744045141325193456"}},"outputId":"7db8f739-da90-4abc-f6cf-d8c089af4fde"},"source":["from google.colab import drive\r\n","drive.mount('/content/gdrive')\r\n","\r\n","X_removed = pd.read_csv('/content/gdrive/My Drive/data_final.csv')"],"execution_count":3,"outputs":[{"output_type":"stream","text":["Drive already mounted at /content/gdrive; to attempt to forcibly remount, call drive.mount(\"/content/gdrive\", force_remount=True).\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"ugrsSHyqNm66","executionInfo":{"status":"ok","timestamp":1607831668404,"user_tz":-480,"elapsed":544,"user":{"displayName":"Gordan Yeung","photoUrl":"","userId":"16744045141325193456"}}},"source":["y_revenue_removed = np.log(X_removed['revenue'])\r\n","\r\n","X_budget = X_removed['budget'].values.reshape(-1, 1)\r\n","X_budget = np.log(X_budget + 1.0)\r\n","\r\n","X_removed['budget_zero'] = np.where(X_removed['budget'] == 0, 1, 0)\r\n","X_budget_zero = X_removed['budget_zero'].values.reshape(-1, 1)"],"execution_count":4,"outputs":[]},{"cell_type":"code","metadata":{"id":"m-kN8rJ6NhkM","executionInfo":{"status":"ok","timestamp":1607831671137,"user_tz":-480,"elapsed":1558,"user":{"displayName":"Gordan Yeung","photoUrl":"","userId":"16744045141325193456"}}},"source":["# Processing Genres data\r\n","genres_prop = np.array([])\r\n","for i in range(len(X_removed)):\r\n","    temp = json.loads(X_removed['genres'][i])\r\n","    temp_2 = np.array([])\r\n","    for j in range(len(temp)):\r\n","        temp_2 = np.append(temp_2, temp[j]['name'])\r\n","    genres_prop = np.append(genres_prop, temp_2)\r\n","\r\n","# Get set of unique elements in movie's genres\r\n","genres_prop = list(set(genres_prop))\r\n","\r\n","X_genres = np.array([])\r\n","for i in range(len(X_removed)):\r\n","    null_genres = [0]*len(genres_prop)\r\n","    temp = json.loads(X_removed['genres'][i])\r\n","    temp_2 = np.array([])\r\n","\r\n","    for j in range(len(temp)):\r\n","        temp_2 = np.append(temp_2, temp[j]['name'])\r\n","\r\n","    for j in range(len(temp_2)):\r\n","        if temp_2[j] in genres_prop:\r\n","            null_genres[genres_prop.index(temp_2[j])] = 1\r\n","        else:\r\n","            null_genres[genres_prop.index(temp_2[j])] = 0\r\n","    X_genres = np.append(X_genres, null_genres)\r\n","\r\n","# Genres data, 1 indicate the movie belongs to that genres' type\r\n","X_genres = X_genres.reshape(-1, len(genres_prop))"],"execution_count":5,"outputs":[]},{"cell_type":"code","metadata":{"id":"itjSkQLQNi8f","executionInfo":{"status":"ok","timestamp":1607831825136,"user_tz":-480,"elapsed":151529,"user":{"displayName":"Gordan Yeung","photoUrl":"","userId":"16744045141325193456"}}},"source":["# Processing Keywords data\r\n","keywords_prop = np.array([])\r\n","for i in range(len(X_removed)):\r\n","    temp = json.loads(X_removed['keywords'][i])\r\n","    temp_2 = np.array([])\r\n","    for j in range(len(temp)):\r\n","        temp_2 = np.append(temp_2, temp[j]['name'])\r\n","    keywords_prop = np.append(keywords_prop, temp_2)\r\n","\r\n","# Get set of unique elements in movie's keywords\r\n","keywords_prop = list(set(keywords_prop))\r\n","\r\n","X_keywords = np.array([])\r\n","for i in range(len(X_removed)):\r\n","    null_keywords = [0]*len(keywords_prop)\r\n","    temp = json.loads(X_removed['keywords'][i])\r\n","    temp_2 = np.array([])\r\n","\r\n","    for j in range(len(temp)):\r\n","        temp_2 = np.append(temp_2, temp[j]['name'])\r\n","\r\n","    for j in range(len(temp_2)):\r\n","        if temp_2[j] in keywords_prop:\r\n","            null_keywords[keywords_prop.index(temp_2[j])] = 1\r\n","        else:\r\n","            null_keywords[keywords_prop.index(temp_2[j])] = 0\r\n","    X_keywords = np.append(X_keywords, null_keywords)\r\n","\r\n","# Genres data, 1 indicate the movie belongs to that genres' type\r\n","X_keywords = X_keywords.reshape(-1, len(keywords_prop))"],"execution_count":6,"outputs":[]},{"cell_type":"code","metadata":{"id":"ZHwSQfZrNlCw","executionInfo":{"status":"ok","timestamp":1607831884793,"user_tz":-480,"elapsed":59653,"user":{"displayName":"Gordan Yeung","photoUrl":"","userId":"16744045141325193456"}}},"source":["# Processing production_companies data\r\n","pro_company_prop = np.array([])\r\n","for i in range(len(X_removed)):\r\n","    temp = json.loads(X_removed['production_companies'][i])\r\n","    temp_2 = np.array([])\r\n","    for j in range(len(temp)):\r\n","        temp_2 = np.append(temp_2, temp[j]['name'])\r\n","    pro_company_prop = np.append(pro_company_prop, temp_2)\r\n","\r\n","# Get set of unique elements in movie's keywords\r\n","pro_company_prop = list(set(pro_company_prop))\r\n","\r\n","X_prop_company = np.array([])\r\n","for i in range(len(X_removed)):\r\n","    null_keywords = [0]*len(pro_company_prop)\r\n","    temp = json.loads(X_removed['production_companies'][i])\r\n","    temp_2 = np.array([])\r\n","\r\n","    for j in range(len(temp)):\r\n","        temp_2 = np.append(temp_2, temp[j]['name'])\r\n","\r\n","    for j in range(len(temp_2)):\r\n","        if temp_2[j] in pro_company_prop:\r\n","            null_keywords[pro_company_prop.index(temp_2[j])] = 1\r\n","        else:\r\n","            null_keywords[pro_company_prop.index(temp_2[j])] = 0\r\n","    X_prop_company = np.append(X_prop_company, null_keywords)\r\n","\r\n","# Genres data, 1 indicate the movie belongs to that genres' type\r\n","X_prop_company = X_prop_company.reshape(-1, len(pro_company_prop))"],"execution_count":7,"outputs":[]},{"cell_type":"code","metadata":{"id":"k8omSwpBNsEj","executionInfo":{"status":"ok","timestamp":1607831884794,"user_tz":-480,"elapsed":59650,"user":{"displayName":"Gordan Yeung","photoUrl":"","userId":"16744045141325193456"}}},"source":["reg = LinearRegression()"],"execution_count":8,"outputs":[]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"KL6CFhxnNuSP","executionInfo":{"status":"ok","timestamp":1607831884794,"user_tz":-480,"elapsed":59646,"user":{"displayName":"Gordan Yeung","photoUrl":"","userId":"16744045141325193456"}},"outputId":"b7e532af-21ed-4c42-8eea-c76eacca2bd5"},"source":["X = np.concatenate((X_budget, X_budget_zero), axis = 1)\r\n","reg.fit(X, y_revenue_removed)\r\n","reg.score(X, y_revenue_removed)"],"execution_count":9,"outputs":[{"output_type":"execute_result","data":{"text/plain":["0.41966934086335517"]},"metadata":{"tags":[]},"execution_count":9}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"biOnlfYeNwqv","executionInfo":{"status":"ok","timestamp":1607831884795,"user_tz":-480,"elapsed":59644,"user":{"displayName":"Gordan Yeung","photoUrl":"","userId":"16744045141325193456"}},"outputId":"44b6eb00-1881-4440-dd7d-982eafdac04b"},"source":["X = np.concatenate((X_budget, X_budget_zero, X_genres), axis = 1)\r\n","reg.fit(X, y_revenue_removed)\r\n","reg.score(X, y_revenue_removed)"],"execution_count":10,"outputs":[{"output_type":"execute_result","data":{"text/plain":["0.4350566638988216"]},"metadata":{"tags":[]},"execution_count":10}]},{"cell_type":"code","metadata":{"id":"mgCJn-HLNyKW"},"source":["X = np.concatenate((X_budget, X_budget_zero, X_genres, X_keywords), axis = 1)\r\n","reg.fit(X, y_revenue_removed)\r\n","reg.score(X, y_revenue_removed)"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"dq_TxhrjNy53"},"source":["X = np.concatenate((X_budget, X_budget_zero, X_genres, X_keywords, X_prop_company), axis = 1)\r\n","reg.fit(X, y_revenue_removed)\r\n","reg.score(X, y_revenue_removed)"],"execution_count":null,"outputs":[]}]}