CODEWITHSUNDEEP
pythonpytorchvalueerror
sempersax
sempersax

Reputation: 31

test_train_split ValueError: Found input variables with inconsistent numbers of samples: [200000, 6]

I've looked at a couple of other posts with this issue, and I cannot figure out what I'm getting wrong here.

I have X_data, and Y_data, and they both have the shape (200000,6). Sample data output from them looks like this:

X_data:
(200000, 6)
[[ 0.00237987  0.00237987 -0.00075756 -0.00221595 -0.00368199  0.00019625]
 [ 0.00171481  0.00171481  0.00176989  0.00125255  0.00275689 -0.00111833]
 [ 0.00190234  0.00190234  0.00333571  0.00127516  0.00146631 -0.00240469]
 ...
 [ 0.00211437  0.00211437  0.00221987  0.0002214   0.00273094 -0.00114419]
 [ 0.00185682  0.00185682  0.00352099  0.00064055 -0.00051575  0.00335213]
 [ 0.00155133  0.00155133 -0.00368774 -0.00200935  0.00225988 -0.00161371]]

Y_data:
(200000, 6)
[[1.         0.14713856 0.04063819 0.03123633 0.00239176 0.01674091]
 [1.         0.35532772 0.09834969 0.19631962 0.0153588  0.10071312]
 [1.         0.17015225 0.04700213 0.04208244 0.00322773 0.02244747]
 ...
 [1.         0.14534398 0.04014234 0.03046259 0.0023313  0.01633189]
 [1.         0.18606737 0.05138638 0.0368341  0.00281708 0.01979553]
 [1.         0.31199003 0.0863072  0.14879644 0.01157114 0.07705023]]

As soon as I do test_train_split, as follows:

    ts1 = 0.2
    rs1 = 42
    X_train, X_test, Y_train, Y_test = train_test_split(X_data, Y_data[0], test_size = ts1, random_state = rs1)

My code crashes with the value error. I have no idea where I'm going wrong.

Upvotes: 0

Views: 116

Answers (1)

Alex Li
Alex Li

Reputation: 274

It seems like the first column of your Y_data matrix is the label for your x data, (I'm not sure what the other 5 columns in your Y_train represent). You are currently getting the first row, which isn't correct (note the size is 6 but you would like one y-value for each x input). So the code I think you want is

X_train, X_test, Y_train, Y_test = train_test_split(X_data,\
    Y_data[:, 0], test_size = ts1, random_state = rs1)

Upvotes: 1

Related Questions