Python Pandas 将列表中的多个 SAS 文件读取到单独的数据帧中

Question

我正在阅读一堆这样的 SAS 文件：

demography = pd.read_sas("demography.sas7bdat", encoding = 'latin-1') adverse_event_ds = pd.read_sas("adverse_event_ds.sas7bdat", encoding = 'latin-1') rpt10344 = pd.read_sas("rpt10344.sas7bdat", encoding = 'latin-1') vaccine_administration = pd.read_sas("vaccine_administration.sas7bdat", encoding = 'latin-1') lab_tests_blood_chemistry_ds = pd.read_sas("lab_tests_blood_chemistry_ds.sas7bdat", encoding = 'latin-1') lab_tests_hematology_ds = pd.read_sas("lab_tests_hematology_ds.sas7bdat", encoding = 'latin-1') lab_tests_miscellaneous_ds = pd.read_sas("lab_tests_miscellaneous_ds.sas7bdat", encoding = 'latin-1') vital_signs = pd.read_sas("vital_signs.sas7bdat", encoding = 'latin-1')

我希望能够用这样的东西替换它：

datasets = ["demography", "adverse_event_ds", "rpt10344", "vaccine_administration", "lab_tests_blood_chemistry_ds", "lab_tests_hematology_ds", "lab_tests_miscellaneous_ds", "vital_signs"]

for dataset in datasets: dataset = pd.read_sas(dataset+".sas7bdat", encoding = 'latin-1')

但是当我做类似的事情时： demography.info()

我得到： NameError: name 'demography' is not defined

幕后发生了什么，我该如何解决？

Answer 1

这是在每次迭代时分配给dataset，而不是创建新变量（例如demography、rpt10344 等）。

我会使用如下数据集字典：

dsd = {}
for dataset in datasets:
    dsd[dataset] = pd.read_sas(dataset+".sas7bdat", encoding = 'latin-1')

或者更pythonic的路线：

dsd = { d : pd.read_sas(d + ".sas7bdat", encoding = 'latin-1') for d in datasets }

出于here 和here 解释的原因，我强烈建议不要分配给单个变量名，但如果您绝对必须使用，则可以使用

for d in datasets:
    globals()[d] = pd.read_sas(d + ".sas7bdat", encoding = 'latin-1')