Found array with 0 sample(s) (shape=(0, 1)) while a minimum of 1 is required

Question

This error keeps coming when I try to find MI values. My code is as follows

X_new = X.copy()
X_new = X_new.fillna(0)
y = data.SalePrice


def make_mi_scores(X, y):
    X = X.copy()
    for colname in X.select_dtypes(["object", "category"]):
        X[colname], _ = X[colname].factorize()       
    discrete_features = [pd.api.types.is_integer_dtype(t) for t in X.dtypes]   
    mi_scores = mutual_info_regression(X, y, discrete_features=discrete_features, random_state=0)
    mi_scores = pd.Series(mi_scores, name="MI Scores", index=X.columns)
    mi_scores = mi_scores.sort_values(ascending=False)
    return mi_scores

def plot_mi_scores(scores):
    scores = scores.sort_values(ascending=True)
    width = np.arange(len(scores))
    ticks = list(scores.index)
    plt.barh(width, scores)
    plt.yticks(width, ticks)
    plt.title("Mutual Information Scores")
    
plt.figure(dpi=100, figsize=(8, 5))
plot_mi_scores(make_mi_scores(X_new,y))

if you want the full notebook here is a link https://www.kaggle.com/code/snigdhkarki/house-price-competition

The error is as follows

    ValueError                                Traceback (most recent call last)
/tmp/ipykernel_19/1575243112.py in <module>
     42 
     43 plt.figure(dpi=100, figsize=(8, 5))
---> 44 plot_mi_scores(make_mi_scores(X_new,y))

/tmp/ipykernel_19/1575243112.py in make_mi_scores(X, y)
     28     print(X.isnull().any().any())
     29     print(y.isnull().any().any())
---> 30     mi_scores = mutual_info_regression(X, y, discrete_features=discrete_features, random_state=0)
     31     mi_scores = pd.Series(mi_scores, name="MI Scores", index=X.columns)
     32     mi_scores = mi_scores.sort_values(ascending=False)

/opt/conda/lib/python3.7/site-packages/sklearn/feature_selection/_mutual_info.py in mutual_info_regression(X, y, discrete_features, n_neighbors, copy, random_state)
    382            of a Random Vector", Probl. Peredachi Inf., 23:2 (1987), 9-16
    383     """
--> 384     return _estimate_mi(X, y, discrete_features, False, n_neighbors, copy, random_state)
    385 
    386 

/opt/conda/lib/python3.7/site-packages/sklearn/feature_selection/_mutual_info.py in _estimate_mi(X, y, discrete_features, discrete_target, n_neighbors, copy, random_state)
    300     mi = [
    301         _compute_mi(x, y, discrete_feature, discrete_target, n_neighbors)
--> 302         for x, discrete_feature in zip(_iterate_columns(X), discrete_mask)
    303     ]
    304 

/opt/conda/lib/python3.7/site-packages/sklearn/feature_selection/_mutual_info.py in <listcomp>(.0)
    300     mi = [
    301         _compute_mi(x, y, discrete_feature, discrete_target, n_neighbors)
--> 302         for x, discrete_feature in zip(_iterate_columns(X), discrete_mask)
    303     ]
    304 

/opt/conda/lib/python3.7/site-packages/sklearn/feature_selection/_mutual_info.py in _compute_mi(x, y, x_discrete, y_discrete, n_neighbors)
    160         return mutual_info_score(x, y)
    161     elif x_discrete and not y_discrete:
--> 162         return _compute_mi_cd(y, x, n_neighbors)
    163     elif not x_discrete and y_discrete:
    164         return _compute_mi_cd(x, y, n_neighbors)

/opt/conda/lib/python3.7/site-packages/sklearn/feature_selection/_mutual_info.py in _compute_mi_cd(c, d, n_neighbors)
    137     radius = radius[mask]
    138 
--> 139     kd = KDTree(c)
    140     m_all = kd.query_radius(c, radius, count_only=True, return_distance=False)
    141     m_all = np.array(m_all) - 1.0

sklearn/neighbors/_binary_tree.pxi in sklearn.neighbors._kd_tree.BinaryTree.__init__()

/opt/conda/lib/python3.7/site-packages/sklearn/utils/validation.py in check_array(array, accept_sparse, accept_large_sparse, dtype, order, copy, force_all_finite, ensure_2d, allow_nd, ensure_min_samples, ensure_min_features, estimator)
    806                 "Found array with %d sample(s) (shape=%s) while a"
    807                 " minimum of %d is required%s."
--> 808                 % (n_samples, array.shape, ensure_min_samples, context)
    809             )
    810 

ValueError: Found array with 0 sample(s) (shape=(0, 1)) while a minimum of 1 is required.

There were only a few places where this question was asked and even in those places I was unable to find any answers to my question

Hi there Please edit the question to include the *exact* error message, we need to know where something went wrong. This question might be improved with a [minimal reproducible example](https://stackoverflow.com/help/minimal-reproducible-example). — Alexander L. Hayes, Dec 24 '22 at 13:58

score 0 · Accepted Answer · answered Dec 24 '22 at 16:32

0

The issue is arising where you are calling mutual_info_regression over here -

mi_scores = mutual_info_regression(X, y, discrete_features=discrete_features, random_state=0)

As per sklearn's documentation, the parameter discrete_features should be a boolean mask that has True for discrete variables and False otherwise.

I checked your Kaggle code and it seems like your technique for identifying discrete and continuous features in your data frame is wrong.

A simple hack to get the code running would be to identify all features as continuous using the following code -

discrete_features = [False]*73
# 73 is the number of columns X has

However, the result might be wrong if the mutual_info_regression algorithm requires you to accurately identify discrete and continuous features.

answered Dec 24 '22 at 16:32

desert_ranger

1,096
3
13
26

wow, it worked wow just wow, I got stuck for a whole day because of this. I have so many questions like why? cuz it worked in other notebook but thanks man you saved me another day. I want to upvote you but I dont have enough upvotes myself to upvote :( – Tottle Dec 24 '22 at 16:37
I did get the upvote :) It just doesn't show. – desert_ranger Dec 24 '22 at 16:45
can you tell why having dicrete features as all false worked? Cuz shouldnt it have true for columns with int datatype and false for columns with other datatypes? My previous code (without putting discrete_features = [False]*73) did work for other previous datasets like https://www.kaggle.com/code/snigdhkarki/use-of-xgb-and-randomforest-includes-feature-eng why did it not work this time? – Tottle Dec 24 '22 at 17:50
It's a little hard to say, especially since there are 73 features. For instance, pandas can assign integer dtypes to columns with missing values - https://stackoverflow.com/questions/21287624/convert-pandas-column-containing-nans-to-dtype-int – desert_ranger Dec 24 '22 at 18:12
A follow up would be, you are discerning a discrete column using Pandas and then feeding it into Sklearn. However, Sklearn and Pandas may have different definitions of a discrete variable. – desert_ranger Dec 24 '22 at 18:20
Why does this manifest a an error about zero _rows_ though?? – Ben Reiniger Dec 25 '22 at 01:54
This could be attributed to another method/function within sklearn failing because mutual_info_regression wasn't called correctly. I am not sure though. – desert_ranger Dec 25 '22 at 02:53

Found array with 0 sample(s) (shape=(0, 1)) while a minimum of 1 is required

1 Answers1