How Do I Reshape A List Into An Array?

Question

Hello, I have been trying to format a list for my predictive algorithm. However, when I try to predict I get the error:

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.

So, when I try to reshape the array using .reshape(1,-1) [As told to me by the error] I get that the 'list' object has no attribute 'reshape'. However, according to this post, my list IS an array, and should be able to do this.

Furthermore, I have tried using numpy to force it to be an array (or transpose it) I got the error: ValueError: setting an array element with a sequence.

My code is such:

for i in range(len(best.indi)):
    data.append(best.features[best.indi[i]])
for i in data:
  try:
    value = i[-1:]
    prediction_data.append(value[0])
  except:
    prediction_data.append(i)

from sklearn.neighbors import KNeighborsClassifier
from sklearn.model_selection import train_test_split as tts
import numpy as np

knn = KNeighborsClassifier(n_neighbors=best.neighbors)

knn.fit(np.transpose(data), best.y)

prediction = knn.predict(np.transpose(prediction_data))


print(prediction)

The try catch is going through the data (which is a combo of 1 itemed lists and numbers) and creating a list that is only a collection of numbers.

https://repl.it/@JacksonEnnis/KNN-Final

So, to reiterate, how do you reshape the data to a format that scikit will recognize for predictions?

When you say you "tried using numpy to force it to be an array", how did you try? It would be helpful to see the code. Generally this can be accomplished with `my_array=np.array(my_list)` — G. Anderson, Jul 03 '19 at 18:54

score 1 · Accepted Answer · answered Jul 03 '19 at 19:14

In the example you showed you are appending the data to a list. To use reshape you have to convert them to an numpy array. Make sure that you check the type of your variable before calling reshape.

import numpy as np

data = []
for i in range(10):
    data.append(i)

print(type(data))

Output: <class 'list'>

data = np.array(data) # Convert the list to numpy array
print(type(data))

Output: <class 'numpy.ndarray'>

Now you can reshape as you like.

print(data.reshape(1,-1))

Output: [[0 1 2 3 4 5 6 7 8 9]]

How Do I Reshape A List Into An Array?

1 Answers1