python 3.x – I have dictionaries in one of the columns and I need to fit this into a Logistic Regression model

This is one of the columns of my training data. The training data has 3 more columns. But I only show one here because this column is taking up too much space.

the word count
1 {& # 39; it & # 39 ;: 2, & # 39 ;: 1, & # 39; early & # 39 ;: 1, & # 39 ;: 3, & # 39; wa ...
2 {& # 39; Very & # 39 ;: 1, & # 39; smooth & # 39 ;: 1, & # 39 ;: 2, & # 39; comfortable & # 39; ...
3 {& # 39; This & # 39 ;: 1, & # 39 ;: 4, & # 39 ;: 2, & # 39; product & # 39 ;: 2, & # 39; we ...
4 {& # 39; All & # 39 ;: 1, & # 39 ;: 1, & # 39 ;: 1, & # 39; kids & # 39 ;: 2, & # 39; have & # 39; ...
5 {& # 39; When & # 39 ;: 1, & # 39 ;: 5, & # 39; Binky & # 39 ;: 3, & # 39; Hada & # 39 ;: 3, ...  

Now I am trying to adjust this data frame and the following error message:

from sklearn.linear_model import LogisticRegression
model = LogisticRegression (). fit (X_train, y_train)


The TypeError: float () argument must be a string or a number, not & dict & # 39;

How to solve this? Can someone please help?