Problem with "ValueError: Expected 2D array, got 1D array instead"

Question

I need to run a SVR (supported vector regression). I have a CSV data frame.I had no problems to run the OLS regression, with one target variable and multiple regressors. But I have a problem with this part of the code.

So, here is my code:


import matplotlib.pyplot as plt
import pandas as pd
from sklearn.preprocessing import StandardScaler
from sklearn.svm import SVR

sc_X = StandardScaler()
sc_y = StandardScaler()
X = sc_X.fit_transform(X)
y = sc_y.fit_transform(y)

y_pred = sc_y.inverse_transform ((regressor.predict (sc_X.transform(np.array([[6.5]])))))
plt.scatter(X, y, color = 'magenta')
plt.plot(X, regressor.predict(X))
plt.title('SVR')
plt.xlabel('X')
plt.ylabel('VF')
plt.show()

X_grid = np.arange(min(X), max(X), 0.1)
X_grid = X_grid.reshape((len(X_grid), 1))
plt.scatter(X, y)
plt.plot(X_grid, regressor.predict(X_grid))
plt.title('SVR')
plt.xlabel('X')
plt.ylabel('VF')
plt.show()

I have the following error message:"ValueError: Expected 2D array, got 1D array instead Reshape your data either using array.reshape(-1, 1) if your data has a single feature or array.reshape(1, -1) if it contains a single sample."

It's the first time I encounter this problem. I saw in other topic that is not scarce, but in fact i don't understand where to reshape the data in my code. When i tried to do it, it says that DataFrame has no reshape function.

Here is a pic of my dataset. The target is VF, all the other variables are the regressors.

Thanks,

I put a picture of my dataset in order to show you the structure. It's a big one, with more than 25.000 values. — mmcr_economics
– mmcr_economics, Commented Nov 22, 2019 at 23:14
It returns (14085, 8). I made a mistake on the comment above, I have 14085 observations. — mmcr_economics
– mmcr_economics, Commented Nov 22, 2019 at 23:18
np.arange(min(X), max(X), 0.1) returns a 1D array, not a 2D one. — Anis R.
– Anis R., Commented Nov 22, 2019 at 23:20

seralouk · Accepted Answer · 2019-11-22 23:19:52Z

1

It seems that when you do:

 X = sc_X.fit_transform(X)

X contains more than one variables. 8 to be specific

Next you do:

 regressor.predict(sc_X.transform(np.array([[6.5]])))

Now you try to transform a new sample that has only one variable but the sc model was trained on data with more than one variable.

answered Nov 22, 2019 at 23:19

seralouk

33.6k10 gold badges127 silver badges141 bronze badges

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

3 Comments

mmcr_economics Over a year ago

Yes, that's exactly the point. How can I fix the problem then ?

seralouk Over a year ago

you have to define a sample that has also 8 features, or fit the sc model using only one variable from X

mmcr_economics Over a year ago

After modifying the code, I have this new error message : non-broadcastable output operand with shape (1,1) doesn't match the broadcast shape (1,2)

Collectives™ on Stack Overflow

Problem with "ValueError: Expected 2D array, got 1D array instead"

1 Answer 1

3 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

3 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related