mirror of
https://github.com/dmlc/dgl.git
synced 2026-06-05 19:54:25 +08:00
0fb13f7b9d
* Add serialization * add serialization * add serialization * lalalalalalalala * lalalalalalalala * serialize * serialize * nnn * WIP: import tvm runtime node system * WIP: object system * containers * tested basic container composition * tested custom object * tmp * fix setattr bug * tested object container return * fix lint * some comments about get/set state * fix lint * fix lint * update cython * fix cython * ffi doc * fix doc * WIP: using object system for graph * c++ side refactoring done; compiled * remove stale apis * fix bug in DGLGraphCreate; passed test_graph.py * fix bug in python modify; passed utest for pytorch/cpu * fix lint * Add serialization * Add serialization * fix * fix typo * serialize with new ffi * commit * commit * commit * save * save * save * save * commit * clean * Delete tt2.py * fix lint * Add serialization * fix lint 2 * fix lint * fix lint * fix lint * fix lint * Fix Lint * Add serialization * Change to Macro * fix * fix * fix bugs * refactor * refactor * updating dmlc-core to include force flag * trying tempfile * delete leaked pointer * Fix assert * fix assert * add comment and test case * add graph labels * add load labels * lint * lint * add graph labels * lint * fix windows * fix * update dmlc-core to latest * fix * fix camel naming
118 lines
2.6 KiB
ReStructuredText
118 lines
2.6 KiB
ReStructuredText
.. _apidata:
|
|
|
|
Dataset
|
|
=======
|
|
|
|
.. currentmodule:: dgl.data
|
|
|
|
Utils
|
|
-----
|
|
|
|
.. autosummary::
|
|
:toctree: ../../generated/
|
|
|
|
utils.get_download_dir
|
|
utils.download
|
|
utils.check_sha1
|
|
utils.extract_archive
|
|
utils.split_dataset
|
|
utils.save_graphs
|
|
utils.load_graphs
|
|
utils.load_labels
|
|
|
|
.. autoclass:: dgl.data.utils.Subset
|
|
:members: __getitem__, __len__
|
|
|
|
Dataset Classes
|
|
---------------
|
|
|
|
Stanford sentiment treebank dataset
|
|
```````````````````````````````````
|
|
|
|
For more information about the dataset, see `Sentiment Analysis <https://nlp.stanford.edu/sentiment/index.html>`__.
|
|
|
|
.. autoclass:: SST
|
|
:members: __getitem__, __len__
|
|
|
|
Mini graph classification dataset
|
|
`````````````````````````````````
|
|
|
|
.. autoclass:: MiniGCDataset
|
|
:members: __getitem__, __len__, num_classes
|
|
|
|
|
|
Graph kernel dataset
|
|
````````````````````
|
|
|
|
For more information about the dataset, see `Benchmark Data Sets for Graph Kernels <https://ls11-www.cs.tu-dortmund.de/staff/morris/graphkerneldatasets>`__.
|
|
|
|
.. autoclass:: TUDataset
|
|
:members: __getitem__, __len__
|
|
|
|
|
|
Graph isomorphism network dataset
|
|
```````````````````````````````````
|
|
|
|
A compact subset of graph kernel dataset
|
|
|
|
.. autoclass:: GINDataset
|
|
:members: __getitem__, __len__
|
|
|
|
|
|
Protein-Protein Interaction dataset
|
|
```````````````````````````````````
|
|
|
|
.. autoclass:: PPIDataset
|
|
:members: __getitem__, __len__
|
|
|
|
Molecular Graphs
|
|
----------------
|
|
|
|
To work on molecular graphs, make sure you have installed `RDKit 2018.09.3 <https://www.rdkit.org/docs/Install.html>`__.
|
|
|
|
Featurization
|
|
`````````````
|
|
|
|
For the use of graph neural networks, we need to featurize nodes (atoms) and edges (bonds). Below we list some
|
|
featurization methods/utilities:
|
|
|
|
.. autosummary::
|
|
:toctree: ../../generated/
|
|
|
|
chem.one_hot_encoding
|
|
chem.BaseAtomFeaturizer
|
|
chem.CanonicalAtomFeaturizer
|
|
|
|
Graph Construction
|
|
``````````````````
|
|
|
|
Several methods for constructing DGLGraphs from SMILES/RDKit molecule objects are listed below:
|
|
|
|
.. autosummary::
|
|
:toctree: ../../generated/
|
|
|
|
chem.mol_to_graph
|
|
chem.smile_to_bigraph
|
|
chem.mol_to_bigraph
|
|
chem.smile_to_complete_graph
|
|
chem.mol_to_complete_graph
|
|
|
|
Dataset Classes
|
|
```````````````
|
|
|
|
If your dataset is stored in a ``.csv`` file, you may find it helpful to use
|
|
|
|
.. autoclass:: dgl.data.chem.CSVDataset
|
|
:members: __getitem__, __len__
|
|
|
|
Currently two datasets are supported:
|
|
|
|
* Tox21
|
|
* TencentAlchemyDataset
|
|
|
|
.. autoclass:: dgl.data.chem.Tox21
|
|
:members: __getitem__, __len__, task_pos_weights
|
|
|
|
.. autoclass:: dgl.data.chem.TencentAlchemyDataset
|
|
:members: __getitem__, __len__, set_mean_and_std
|