skchem.data.datasets.base.Dataset(**kwargs)[source]¶Bases: fuel.datasets.hdf5.H5PYDataset
Abstract base class providing an interface to the skchem data format.
download(output_directory=None, download_directory=None)[source]¶Download the dataset and convert it.
| Parameters: |
|
|---|---|
| Returns: | The path of the downloaded and processed dataset. |
| Return type: | str |
load_data(sets=(), sources=())[source]¶Load a set of sources.
| Parameters: |
|
|---|
Example
(X_train, y_train), (X_test, y_test) = Dataset.load_data(sets=(‘train’, ‘test’), sources=(‘X’, ‘y’))
load_set(set_name, sources=())[source]¶Load the sources for a single set.
| Parameters: |
|
|---|---|
| Returns: |
|
read_frame(key, *args, **kwargs)[source]¶Load a set of features from the dataset as a pandas object.
| Parameters: | key (str) – The HDF5 key for required data. Typically, this will be one of
|
|---|---|
| Returns: |
|
## skchem.data.datasets
Module defining skchem datasets.
skchem.data.datasets.Diversity(**kwargs)[source]¶Bases: skchem.data.datasets.base.Dataset
Example dataset, the NCI DTP Diversity Set III.
converter¶alias of DiversityConverter
downloader¶alias of DiversityDownloader
filename = 'diversity.h5'¶skchem.data.datasets.BursiAmes(**kwargs)[source]¶Bases: skchem.data.datasets.base.Dataset
converter¶alias of BursiAmesConverter
downloader¶alias of BursiAmesDownloader
filename = 'bursi_ames.h5'¶skchem.data.datasets.MullerAmes(**kwargs)[source]¶Bases: skchem.data.datasets.base.Dataset
converter¶alias of MullerAmesConverter
downloader¶alias of MullerAmesDownloader
filename = 'muller_ames.h5'¶skchem.data.datasets.PhysProp(**kwargs)[source]¶Bases: skchem.data.datasets.base.Dataset
converter¶alias of PhysPropConverter
downloader¶alias of PhysPropDownloader
filename = 'physprop.h5'¶skchem.data.datasets.BradleyOpenMP(**kwargs)[source]¶Bases: skchem.data.datasets.base.Dataset
converter¶alias of BradleyOpenMPConverter
downloader¶alias of BradleyOpenMPDownloader
filename = 'bradley_open_mp.h5'¶