Cogdl

The new v0.6 release updates the tutorials and adds more examples, such as [GraphMAE](https://github.com/THUDM/cogdl/tree/master/examples/graphmae), [GraphMAE2](https://github.com/THUDM/cogdl/tree/master/examples/graphmae2), and [BGRL](https://github.com/THUDM/cogdl/tree/master/examples/bgrl).

What's Changed
* Update doc tutorials by QingFei1 in https://github.com/THUDM/cogdl/pull/352, https://github.com/THUDM/cogdl/pull/355
* Integrate GRB by xll2001 in https://github.com/THUDM/cogdl/pull/347
* Add dgraph-cogdl in examples by Kinseys in https://github.com/THUDM/cogdl/pull/357
* Update README by cenyk1230 in https://github.com/THUDM/cogdl/pull/364
* update ogbl datasets by Diego0511 in https://github.com/THUDM/cogdl/pull/358
* update APIs for gensim 4.x by Saltsmart in https://github.com/THUDM/cogdl/pull/361
* Update Triple_Link_Prediction by QingFei1 in https://github.com/THUDM/cogdl/pull/371
* Small Changes by QingFei1 in https://github.com/THUDM/cogdl/pull/374, https://github.com/THUDM/cogdl/pull/381
* Update Graphsage/Unsup_Graphsage by QingFei1 in https://github.com/THUDM/cogdl/pull/379, https://github.com/THUDM/cogdl/pull/384, https://github.com/THUDM/cogdl/pull/425
* Fix bugs in oagbert.encode_paper by THINK2TRY in https://github.com/THUDM/cogdl/pull/385
* Revise for GCC by hwangyeong in https://github.com/THUDM/cogdl/pull/392
* Update GRB by cenyk1230 in https://github.com/THUDM/cogdl/pull/406
* Add stgcn code for traffic prediction task by Renxs177 in https://github.com/THUDM/cogdl/pull/407
* BGRL with CogDL by hwangyeong in https://github.com/THUDM/cogdl/pull/408
* GCC update by hwangyeong in https://github.com/THUDM/cogdl/pull/409
* Add GraphMAE by cenyk1230 in https://github.com/THUDM/cogdl/pull/428
* Add GraphMAE2 by cenyk1230 in https://github.com/THUDM/cogdl/pull/429

New Contributors
* xll2001 made their first contribution in https://github.com/THUDM/cogdl/pull/347
* Kinseys made their first contribution in https://github.com/THUDM/cogdl/pull/357
* Diego0511 made their first contribution in https://github.com/THUDM/cogdl/pull/358
* Saltsmart made their first contribution in https://github.com/THUDM/cogdl/pull/361
* hwangyeong made their first contribution in https://github.com/THUDM/cogdl/pull/392
* Renxs177 made their first contribution in https://github.com/THUDM/cogdl/pull/407

**Full Changelog**: https://github.com/THUDM/cogdl/compare/v0.5.3...v0.6

The CogDL v0.5.3 release supports mixed-precision training by setting *fp16=True* and provides a basic [example](https://github.com/THUDM/cogdl/blob/master/examples/jittor/gcn.py) written by [Jittor](https://github.com/Jittor/jittor). It also updates the tutorial in the document, fixes downloading links of some datasets, and fixes potential bugs of operators.

What's Changed
* [Dataset] Update rd2cd datasets by cenyk1230 in https://github.com/THUDM/cogdl/pull/323
* [Feature] Support fp16 by cenyk1230 in https://github.com/THUDM/cogdl/pull/325
* [Bugfix] Fix copying args by cenyk1230 in https://github.com/THUDM/cogdl/pull/326
* [Example] Add GAT for ogbn-arxiv dataset by cenyk1230 in https://github.com/THUDM/cogdl/pull/327
* [Enhancement] Merge parallel training by cenyk1230 in https://github.com/THUDM/cogdl/pull/332
* [Bugfix] Fix dgk/graph2vec/gdc/grace by cenyk1230 in https://github.com/THUDM/cogdl/pull/335
* [Dependency] Fix numpy version by cenyk1230 in https://github.com/THUDM/cogdl/pull/338
* [Dataset] Update download links by cenyk1230 in https://github.com/THUDM/cogdl/pull/346
* [Doc] Update doc tutorials by cenyk1230 in https://github.com/THUDM/cogdl/pull/348
* [Bugfix] Fix edge softmax by cenyk1230 in https://github.com/THUDM/cogdl/pull/349
* [Feature] Jittor gcn example by cenyk1230 in https://github.com/THUDM/cogdl/pull/350
* [Doc] Prepare v0.5.3 release by cenyk1230 in https://github.com/THUDM/cogdl/pull/351

**Full Changelog**: https://github.com/THUDM/cogdl/compare/v0.5.2...v0.5.3

The CogDL 0.5.2 release adds a GNN example for ogbn-products and updates geom datasets. It also fixes some potential bugs including setting devices, using cpu for inference, etc.

What's Changed
* [Bugfix] Fix packing operator files by cenyk1230 in https://github.com/THUDM/cogdl/pull/314
* [Dataset] Update geom datasets by cenyk1230 in https://github.com/THUDM/cogdl/pull/315
* [Bugfix] Fix set device by cenyk1230 in https://github.com/THUDM/cogdl/pull/316
* [Bugfix] Fix data memory by cenyk1230 in https://github.com/THUDM/cogdl/pull/317
* [Example] Add clustergcn for ogbn by cenyk1230 in https://github.com/THUDM/cogdl/pull/320
* [Doc] Prepare v0.5.2 release by cenyk1230 in https://github.com/THUDM/cogdl/pull/322


**Full Changelog**: https://github.com/THUDM/cogdl/compare/v0.5.1...v0.5.2

The CogDL 0.5.1 release adds fast operators including SpMM (cpu version) and scatter_max (cuda version). It also adds lots of datasets for node classification which can be found in this link.

What's Changed
* [Feature] Add fast spmm (cpu) by cenyk1230 in https://github.com/THUDM/cogdl/pull/312
* [Operator] new scatter_max by fishmingyu in https://github.com/THUDM/cogdl/pull/308
* [Dataset] Add more datasets by cenyk1230 in https://github.com/THUDM/cogdl/pull/313


**Full Changelog**: https://github.com/THUDM/cogdl/compare/v0.5.0...v0.5.1

The **CogDL 0.5.0 release** focuses on **modular design** and **ease of use**. It designs and implements a unified training loop for GNN, which introduces `DataWrapper` to help prepare the training/validation/test data and `ModelWrapper` to define the training/validation/test steps.

What's Changed
* [Bugfix] Fix MoEGCN & actnn import by cenyk1230 in https://github.com/THUDM/cogdl/pull/271
* [Notebook] Add notebooks by cenyk1230 in https://github.com/THUDM/cogdl/pull/276
* [Paperlist] 100 GNN papers by cenyk1230 in https://github.com/THUDM/cogdl/pull/277
* [Framework] Unify the GNN training loop by cenyk1230 in https://github.com/THUDM/cogdl/pull/285
* [Framework] Remove register models/datasets/wrappers by cenyk1230 in https://github.com/THUDM/cogdl/pull/286
* [Pipeline] Fix pipeline by cenyk1230 in https://github.com/THUDM/cogdl/pull/289
* [Custom] Fix model name by cenyk1230 in https://github.com/THUDM/cogdl/pull/290
* [Docs] Update docs & examples by cenyk1230 in https://github.com/THUDM/cogdl/pull/292
* [Docs] Fix building docs by cenyk1230 in https://github.com/THUDM/cogdl/pull/293
* [Dataset] Update ogb arxiv & Fix epochs by cenyk1230 in https://github.com/THUDM/cogdl/pull/294
* [Custom] Fix custom wrappers by cenyk1230 in https://github.com/THUDM/cogdl/pull/295
* [Dataset] Add geom datasets by cenyk1230 in https://github.com/THUDM/cogdl/pull/296
* [Model] Add fused GAT by cenyk1230 in https://github.com/THUDM/cogdl/pull/297
* [Submodule] Add FastMoE as third-party library by cenyk1230 in https://github.com/THUDM/cogdl/pull/298
* [Model] Move pyg models to examples by cenyk1230 in https://github.com/THUDM/cogdl/pull/299
* [Bugfix] Fix sample adj by cenyk1230 in https://github.com/THUDM/cogdl/pull/301
* [DATASET] Add description for datasets by THINK2TRY in https://github.com/THUDM/cogdl/pull/304
* [Utility] Update spmm utils by cenyk1230 in https://github.com/THUDM/cogdl/pull/303
* [Model] VRGCN example by huangtinglin in https://github.com/THUDM/cogdl/pull/305
* [Utility] Update spmm utils by cenyk1230 in https://github.com/THUDM/cogdl/pull/306
* [Bugfix] Update loading datasets by cenyk1230 in https://github.com/THUDM/cogdl/pull/307
* [Feature] Support AutoGNN by jasmine-yu in https://github.com/THUDM/cogdl/pull/309
* [Bugfix] Fix GAT's NaN by cenyk1230 in https://github.com/THUDM/cogdl/pull/310

New Contributors
* huangtinglin made their first contribution in https://github.com/THUDM/cogdl/pull/305
* jasmine-yu made their first contribution in https://github.com/THUDM/cogdl/pull/309

**Full Changelog**: https://github.com/THUDM/cogdl/compare/0.4.1...v0.5.0

The **CogDL 0.5.0 release** focuses on **modular design** and **ease of use**. It designs and implements a unified training loop for GNN, which introduces `DataWrapper` to help prepare the training/validation/test data and `ModelWrapper` to define the training/validation/test steps.