| import glob | |
| import os | |
| list_of_files = glob.glob('/path/to/folder/filename_*.csv') | |
| latest_file = max(list_of_files, key=os.path.getmtime) | |
| print latest_file |
| # listings$lat <- 0 | |
| # listings$lng <- 0 | |
| # listings$address <- 0 | |
| # | |
| # for (i in 1:nrow(listings)) { | |
| # row <- listings[i,] | |
| # company_name <- gsub(' ', '\\+', row$company) | |
| # location <- gsub(' ', '\\+', row$location) | |
| # | |
| # url <- sprintf('https://maps.googleapis.com/maps/api/place/findplacefromtext/json?input=%s+%s&inputtype=textquery&fields=geometry,formatted_address&key=<API_key>', company_name, location) |
Multicollinearity in regression must be addressed — variables should be removed until the multicollinearity is gone.
Multicollinearity is not such a problem for trees, clustering and nearest-neighbours methods. In these methods, it may be advisable to retain p dummy variables. However, even in these methods, non-redundancy in predictor variables is still desired.
| ## Import | |
| from sklearn.grid_search import GridSearchCV | |
| ## Define grid | |
| param_grid = {‘polynomialfeatures__degree’: np.arrange(21), | |
| ‘linearregression__fit_intercept’: [True, False], | |
| ‘linearregression_normalize’: [True, False]} | |
| ## Grid search | |
| grid = GridSearchCV(PolynomialRegression(), param_grid, cv=7) |
| ## Import libraries | |
| from googleapiclient.discovery import build | |
| ## Set credentials | |
| my_api_key = "API_key" | |
| my_cse_id = "CSE_ID" | |
| ## Define function | |
| def google_search(search_term, api_key, cse_id, **kwargs): | |
| service = build("customsearch", "v1", developerKey=api_key) |