How to fix this error: ValueError: could not convert string to float: 'A'

Question

I'm not sure how to get rid of this error. Below is my example datasets. Is there another step that I'm missing?

Code below: 
from sklearn.model_selection import train_test_split 
from sklearn.ensemble import RandomForestClassifier 
models = RandomForestClassifier(n_estimators=100) 
np.random.seed(42)

X = re_arrange.drop('Gender',axis=1) 
y = re_arrange['Gender']

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)

models.fit(X_train,y_train)
models.score(X_test, y_test)

RandomForestClassifier requires features to be numerical (float or int). Branch is string and cannot be converted to float. You should look at either categorical or one-hot-encoder. — Quang Hoang
– Quang Hoang, Commented May 26, 2020 at 3:55
@QuangHoang appreicate the fast response. will look into this — dhan004
– dhan004, Commented May 26, 2020 at 4:06

sam · Accepted Answer · 2020-05-26 04:00:28Z

2

Your column "Branch" has letters whereas the RandomForestClassifier expects numbers. I believe it is of categorical type. So you can encode the column "Branch" using some categorical encoding as shown below before you do train test split

X["Branch"] = pd.get_dummies(X["Branch"])

It will map letters 'A', 'B' etc in numbers. It does not change your data but just converts them in computational-friendly state

answered May 26, 2020 at 4:00

sam

2,31725 silver badges34 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

dhan004 Over a year ago

appreicate the fast response. never knew what get_dummies was. going to play around with this as well.

Mehul Gupta · Accepted Answer · 2020-05-26 04:07:34Z

0

RandomForestClassifier can handle only numerical values in any of its features. As you can see, you have text/object data in almost all your features. So 1st of all: do X.info() to know the data type of your features. If you find 'string' & 'object', encode all those features in numbers using One-Hot-Encoder or LabelEncoding.

One-Hot-Encoding

LabelEncoding

answered May 26, 2020 at 4:07

Mehul Gupta

1,9393 gold badges21 silver badges39 bronze badges

Collectives™ on Stack Overflow

How to fix this error: ValueError: could not convert string to float: 'A'

2 Answers 2

1 Comment

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

Comments

Your Answer

Sign up or log in

Post as a guest

Related