import time import pandas as pd import advance,run app_labels = pd.read_csv(r'data\app_labels.csv') label_categories= pd.read_csv('data\label_categories.csv') new_app_lables=run.allAdvance(app_labels) app_cate={} new_app_lables = app_labels['genter'].tolist() for i in new_app_lables: app_cate[i] = app_cate.get(i,0) + 1 items=list(app_cate.items()) items.sort(key=lambda x:x[1],reverse=True) new_app_10_lables=pd.DataFrame(items,columns=['male', 'count']) print(new_app_10_lables) # result = pd.merge(new_app_10_lables, label_categories, on='label_id') # result.to_csv("lable_trans.csv")
import time import pandas as pd import advance, run phone_brand_device_model = pd.read_csv(r'data\phone_brand_device_model.csv') events = pd.read_csv(r'data\events.csv') new_phone_brand_device_model = run.allAdvance(phone_brand_device_model) events = run.allAdvance(events) appPingPai_cate = {} new_app_brand = pd.merge(events, phone_brand_device_model, on='device_id') new_app_brand = new_app_brand['phone_brand'].tolist() for i in new_app_brand: appPingPai_cate[i] = appPingPai_cate.get(i, 0) + 1 items = list(appPingPai_cate.items()) items.sort(key=lambda x: x[1], reverse=True) new_app_brand_phone = pd.DataFrame(items[:10], columns=['pingpai', 'count']) new_app_brand_phone.to_csv("pingpai.csv")
import time import pandas as pd import advance, run from sklearn.preprocessing import LabelEncoder, OneHotEncoder phone_brand_device_model = pd.read_csv('data\phone_brand_device_model.csv') phone_brand_device_model['newColumn'] = phone_brand_device_model[ 'phone_brand'] + phone_brand_device_model['device_model'] advance.drop_col(phone_brand_device_model, ['phone_brand', 'device_model']) # 使用LabelEncoder将类别转换为数字 model_label_encoder = LabelEncoder() model_label_encoder.fit(phone_brand_device_model['newColumn'].values) phone_brand_device_model[ 'brand_model_label_code'] = model_label_encoder.transform( phone_brand_device_model['newColumn'].values) advance.drop_col(phone_brand_device_model, ['newColumn']) phone_brand_device_model = run.allAdvance(phone_brand_device_model) phone_brand_device_model.to_csv('new_phone_brand_device_model.csv')