This dataset was obtained from the StatLib repository
(https://www.dcc.fc.up.pt/~ltorgo/Regression/cal_housing.html).
It's a continuous regression dataset with 20,640 samples with 8 features each.
The target variable is a scalar: the median house value for California districts, in dollars.
The 8 input features are the following:
MedInc: median income in block group
HouseAge: median house age in block group
AveRooms: average number of rooms per household
AveBedrms: average number of bedrooms per household
Population: block group population
AveOccup: average number of household members
Latitude: block group latitude
Longitude: block group longitude
This dataset was derived from the 1990 U.S. census, using one row per census block group. A block group is the smallest geographical unit for which the U.S. Census Bureau publishes sample data (a block group typically has a population of 600 to 3,000 people).
A household is a group of people residing within a home. Since the average number of rooms and bedrooms in this dataset are provided per household, these columns may take surprisingly large values for block groups with few households and many empty houses, such as vacation resorts.
dataset_california_housing(
version = "large",
path = "california_housing.npz",
test_split = 0.2,
seed = 113L,
convert = TRUE
)Nested list of arrays: (x_train, y_train), (x_test, y_test).
str(dataset_california_housing())
## List of 2
## $ train:List of 2
## ..$ x: num [1:16512, 1:8] -118 -118 -122 -118 -123 ...
## ..$ y: num [1:16512(1d)] 252300 146900 290900 141300 500001 ...
## $ test :List of 2
## ..$ x: num [1:4128, 1:8] -118 -120 -121 -122 -117 ...
## ..$ y: num [1:4128(1d)] 397900 227900 172100 186500 148900 ...str(dataset_california_housing(convert = FALSE))
## List of 2
## $ train:List of 2
## ..$ x: <numpy.ndarray shape(16512,8), dtype=float32>
## ..$ y: <numpy.ndarray shape(16512), dtype=float32>
## $ test :List of 2
## ..$ x: <numpy.ndarray shape(4128,8), dtype=float32>
## ..$ y: <numpy.ndarray shape(4128), dtype=float32>"small" or "large". The small version
contains 600 samples, the large version contains
20,640 samples. The purpose of the small version is
to serve as an approximate replacement for the
deprecated boston_housing dataset.
path where to cache the dataset locally
(relative to Sys.getenv("KERAS_HOME")).
fraction of the data to reserve as test set.
Random seed for shuffling the data before computing the test split.
When TRUE (default) the datasets are returned as R arrays.
If FALSE, objects are returned as NumPy arrays.
Other datasets:
dataset_boston_housing()
dataset_cifar10()
dataset_cifar100()
dataset_fashion_mnist()
dataset_imdb()
dataset_mnist()
dataset_reuters()