【发布时间】:2021-08-21 15:33:33
【问题描述】:
from fastai import *
from fastai.tabular import *
from fastai.tabular.all import *
import pandas as pd
# set seed for reproducibility
custom_set_seed(42)
df = pd.read_csv('credit_card_default.csv', index_col=0, na_values='')
df.head()
DEP_VAR = 'default_payment_next_month'
num_features = list(df.select_dtypes('number').columns)
num_features.remove(DEP_VAR)
cat_features = list(df.select_dtypes('object').columns)
preprocessing = [FillMissing, Categorify, Normalize]
data = (TabularList.from_df(df, cat_names=cat_features, cont_names=num_features, procs=preprocessing).split_by_rand_pct(valid_pct=0.2, seed=42).label_from_df(cols=DEP_VAR).databunch())
我一直在尝试运行这段代码,但一直遇到这个错误:
NameError Traceback (most recent call last)
<ipython-input-42-5ca7e57a8e36> in <module>
1 # Create a TabularDataBunch from the DataFrame
2
----> 3 data = (TabularList.from_df(df, cat_names=cat_features, cont_names=num_features, procs=preprocessing).split_by_rand_pct(valid_pct=0.2, seed=42).label_from_df(cols=DEP_VAR).databunch())
NameError: name 'TabularList' is not defined
我相信我已经导入了所有需要的模块。有人可以为此提出解决方案吗?
【问题讨论】:
标签: python deep-learning fast-ai