lupejuares lupejuares - 1 month ago 18
Python Question

predicting crime in san francisco, ValueError

i ran into this error while trying to do a project.
ValueError: Found arrays with inconsistent numbers of samples: [878049 884262]. i get it when i try to run my knn classifier at the bottom. ive been reading about it and i know its because my X and y are not the same. the shape for X is (878049, 2) and y is (884262,). how can i fix this error so that they match?

# drop features that we wont be using
#train.head()
df = train.drop(['Descript', 'Resolution', 'Address'],axis=1)

df2 = test.drop(['Address'],axis=1)

# trying to see the times during a day a particualr crime occurs, for example
# rapes occur more from 12am-4am during the weekend.example below
dow = {
'Monday':0,
'Tuesday':1,
'Wednesday':2,
'Thursday':3,
'Friday':4,
'Saturday':5,
'Sunday':6
}
df['DOW'] = df.DayOfWeek.map(dow)

# Add column containing time of day
df['Hour'] = pd.to_datetime(df.Dates).dt.hour

# making my feature column
feature_cols = ['DOW','Hour']
X = df[feature_cols]

df2['DOW'] = df2.DayOfWeek.map(dow)


y=df2['DOW']

# columns in X and y dont match
print(X.shape)
print(y.shape)
print(y.head())
print(X.head())

# Knn classifier
k = 5
my_knn_for_cs4661 = KNeighborsClassifier(n_neighbors=k)
my_knn_for_cs4661.fit(X, y)
#KNN (with k=5), Decision Tree accuracy
y_predict = my_knn_for_cs4661.predict(X)
print('\n')
score = accuracy_score(y, y_predict)
print("K=",k,"Has ",score, "Accuracy")
results = pd.DataFrame()
results['actual'] = y
results['prediction'] = y_predict
print(results.head(10))



---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-11-5a002c1fd668> in <module>()
7 k = 5
8 my_knn_for_cs4661 = KNeighborsClassifier(n_neighbors=k)
----> 9 my_knn_for_cs4661.fit(X, y)
10 #KNN (with k=5), Decision Tree accuracy
11 y_predict = my_knn_for_cs4661.predict(X)

C:\Users\Michael\Anaconda3\lib\site-packages\sklearn\neighbors\base.py in fit(self, X, y)
776 """
777 if not isinstance(X, (KDTree, BallTree)):
--> 778 X, y = check_X_y(X, y, "csr", multi_output=True)
779
780 if y.ndim == 1 or y.ndim == 2 and y.shape[1] == 1:

C:\Users\Michael\Anaconda3\lib\site-packages\sklearn\utils\validation.py in check_X_y(X, y, accept_sparse, dtype, order, copy, force_all_finite, ensure_2d, allow_nd, multi_output, ensure_min_samples, ensure_min_features, y_numeric, warn_on_dtype, estimator)
518 y = y.astype(np.float64)
519
--> 520 check_consistent_length(X, y)
521
522 return X, y

C:\Users\Michael\Anaconda3\lib\site-packages\sklearn\utils\validation.py in check_consistent_length(*arrays)
174 if len(uniques) > 1:
175 raise ValueError("Found arrays with inconsistent numbers of samples: "
--> 176 "%s" % str(uniques))
177
178

ValueError: Found arrays with inconsistent numbers of samples: [878049 884262]

Answer

Check shape of X and y by using X.shape. Stack trace says you have different no of instances(no of samples) in X and y. This is why fit function is throwing ValueError.

Refer documentation it states:

"""Fit the model using X as training data and y as target values
        Parameters
        ----------
        X : {array-like, sparse matrix, BallTree, KDTree}
            Training data. If array or matrix, shape [n_samples, n_features],
            or [n_samples, n_samples] if metric='precomputed'.
        y : {array-like, sparse matrix}
            Target values, array of float values, shape = [n_samples]
             or [n_samples, n_outputs]
        """

In simple words,

X is (878049, 2) -> n_samples  = 878049 and n_features = 2
y is (884262,)  -> Here, n_samples = 884262

You are passing extra target values. Reduce no of target values in y. As your n_samples for X is 878049, you must pass same number of target values(878049).

You can try:

my_knn_for_cs4661.fit(X, y[:878049])