CogDL v0.5.3

Release 0.5.3

The CogDL v0.5.3 release supports mixed-precision training by setting fp16=True and provides a basic example written by 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 #323
• [Feature] Support fp16 by @cenyk1230 in #325
• [Bugfix] Fix copying args by @cenyk1230 in #326
• [Example] Add GAT for ogbn-arxiv dataset by @cenyk1230 in #327
• [Enhancement] Merge parallel training by @cenyk1230 in #332
• [Bugfix] Fix dgk/graph2vec/gdc/grace by @cenyk1230 in #335
• [Dependency] Fix numpy version by @cenyk1230 in #338
• [Dataset] Update download links by @cenyk1230 in #346
• [Doc] Update doc tutorials by @cenyk1230 in #348
• [Bugfix] Fix edge softmax by @cenyk1230 in #349
• [Feature] Jittor gcn example by @cenyk1230 in #350
• [Doc] Prepare v0.5.3 release by @cenyk1230 in #351

Full Changelog: v0.5.2...v0.5.3