Concrete Compressive Strength#

Compressive strength of 1030 concrete samples, to be predicted from the amounts of seven ingredients of the mixture and the age of the concrete. The source describes the compressive strength as a highly nonlinear function of age and ingredients. It is a small, clean regression dataset: all features are numbers and nothing is missing.

Overview#

Loader

load_concrete_strength()

Task

Regression

Samples

1030

Features

8

Targets

1 continuous target, concrete_compressive_strength

Missing values

None

License

CC BY 4.0

Features#

The features are stored in this order: cement, blast_furnace_slag, fly_ash, water, superplasticizer, coarse_aggregate, fine_aggregate and age. The first seven are the amounts of the ingredients in kg in a m3 mixture, and age is the age of the concrete in days, from 1 to 365. All of them are stored as float64. The names are the names of the source written in lower case with underscores.

Target#

y is the concrete compressive strength in MPa (megapascals), stored as float64.

Source and citation#

Please cite the dataset as:

Yeh, I. (1998). Concrete Compressive Strength [Dataset]. UCI Machine Learning Repository. https://doi.org/10.24432/C5PK67

See also:

  • Yeh, I. (1998). Modeling of strength of high-performance concrete using artificial neural networks. Cement and Concrete Research, Vol. 28, No. 12.

License#

CC BY 4.0, as stated by the source. The license was checked at the source on 2026-10-24. See License and citation.

Changes made#

  • Converted to the NumPy NPZ format. Features and the target are stored as float64.

  • The names of the columns are written in lower case with underscores, for example Blast Furnace Slag is stored as blast_furnace_slag. The features are stored in the order of the data description of the source, the target is the concrete compressive strength.

Note

The seven ingredients are given in kg in a m3 mixture, the age in days (1 to 365) and the compressive strength in MPa.

Note

The source describes the compressive strength of concrete as a highly nonlinear function of age and ingredients.

Note

The data file that UCI offers for download (Concrete_Data.xls) has more decimals than the table that ucimlrepo returns and that is stored here: the amounts of the ingredients differ by at most 0.05 and the strength by less than 0.005.

Example#

>>> from pyml_datasets import load_concrete_strength
>>> data = load_concrete_strength()
>>> data.X.shape, data.y.shape
((1030, 8), (1030,))
>>> data.feature_names[-1]
'age'
>>> data.has_missing
False
>>> round(float(data.y.min()), 2), round(float(data.y.max()), 2)
(2.33, 82.6)