數(shù)據(jù)科學(xué)的面試的一些基本問題總結(jié)

來源：Deephub Imba
本文約5000字，建議閱讀10分鐘
本文將介紹如何為成功的面試做準(zhǔn)備的，以及可以幫助我們面試的一些資源。

代碼開發(fā)基礎(chǔ)

• 從表中選擇某些列
• 連接兩個(gè)表（內(nèi)連接、左連接、右連接和外連接）
• 匯總結(jié)果（總和、平均值、最大值、最小值）
• 在 SQL 中使用窗口函數(shù)
• 日期處理

• 處理df（pandas），例如讀取、加入、合并、過濾
• 操作日期和格式化日期
• 操作字符串，例如使用正則表達(dá)式、搜索字符串包含的內(nèi)容
• 有效地使用循環(huán)
• 使用列表和字典
• 在 Python 中創(chuàng)建函數(shù)和類

了解數(shù)據(jù)結(jié)構(gòu)和算法

• 大O符號(hào)
• 二進(jìn)制搜索
• 數(shù)組和鏈表
• 選擇排序
• 快速排序
• 冒泡排序
• 合并排序
• 哈希表

線性回歸

Logistic 回歸

聚類

隨機(jī)森林和提升樹

自編碼器

梯度下降

獨(dú)熱編碼與標(biāo)簽編碼

• 標(biāo)簽編碼
• One-Hot 編碼

# Import label encoderfrom sklearn import preprocessing# label_encoder object knows how to understand word labels.label_encoder = preprocessing.LabelEncoder()# Encode labels in column ‘Country’.data[‘Country’]= label_encoder.fit_transform(data[‘Country’])print(data.head())

# importing one hot encoderfrom sklearn from sklearn.preprocessing import OneHotEncoder# creating one hot encoder objectonehotencoder = OneHotEncoder()#reshape the 1-D country array to 2-D as fit_transform expects 2-D and finally fit the objectX = onehotencoder.fit_transform(data.Country.values.reshape(-1,1)).toarray()#To add this back into the original dataframedfOneHot = pd.DataFrame(X, columns = [“Country_”+str(int(i)) for i in range(data.shape[1])])df = pd.concat([data, dfOneHot], axis=1)#droping the country columndf= df.drop([‘Country’], axis=1)#printing to verifyprint(df.head())

• VIF=1，多重共線性非常少
• VIF<5，中度多重共線性
• VIF>5，極端多重共線性（這是我們必須避免的）

超參數(shù)調(diào)優(yōu)

from sklearn.model_selection import RandomizedSearchCV# Number of trees in random forestn_estimators = [int(x) for x in np.linspace(start = 200, stop = 2000, num = 10)]# Number of features to consider at every splitmax_features = [‘a(chǎn)uto’, ‘sqrt’]# Maximum number of levels in treemax_depth = [int(x) for x in np.linspace(10, 110, num = 11)]max_depth.append(None)# Minimum number of samples required to split a nodemin_samples_split = [2, 5, 10]# Minimum number of samples required at each leaf nodemin_samples_leaf = [1, 2, 4]# Method of selecting samples for training each treebootstrap = [True, False]# Create the random gridrandom_grid = {‘n_estimators’: n_estimators,‘max_features’: max_features,‘max_depth’: max_depth,‘min_samples_split’: min_samples_split,‘min_samples_leaf’: min_samples_leaf,‘bootstrap’: bootstrap}pprint(random_grid){‘bootstrap’: [True, False],‘max_depth’: [10, 20, 30, 40, 50, 60, 70, 80, 90, 100, None],‘max_features’: [‘a(chǎn)uto’, ‘sqrt’],‘min_samples_leaf’: [1, 2, 4],‘min_samples_split’: [2, 5, 10],‘n_estimators’: [200, 400, 600, 800, 1000, 1200, 1400, 1600, 1800, 2000]}# Use the random grid to search for best hyperparameters# First create the base model to tunerf = RandomForestRegressor()# Random search of parameters, using 3 fold cross validation,# search across 100 different combinations, and use all available coresrf_random = RandomizedSearchCV(estimator = rf, param_distributions = random_grid, n_iter = 100, cv = 3, verbose=2, random_state=42, n_jobs = -1)# Fit the random search modelrf_random.fit(train_features, train_labels)

rf_random.best_params_{‘bootstrap’: True,‘max_depth’: 70,‘max_features’: ‘a(chǎn)uto’,‘min_samples_leaf’: 4,‘min_samples_split’: 10,‘n_estimators’: 400}

from sklearn.model_selection import GridSearchCV
# Create the parameter grid based on the results of random searchparam_grid = {‘bootstrap’: [True],‘max_depth’: [80, 90, 100, 110],‘max_features’: [2, 3],‘min_samples_leaf’: [3, 4, 5],‘min_samples_split’: [8, 10, 12],‘n_estimators’: [100, 200, 300, 1000]}
# Create a based modelrf = RandomForestRegressor()
# Instantiate the grid search modelgrid_search = GridSearchCV(estimator = rf, param_grid = param_grid,cv = 3, n_jobs = -1, verbose = 2)

精度和召回

損失函數(shù)

最后總結(jié)

編輯：王菁