Frank Liu
Frank Liu
Change-Id: I7e3f53efd35e7971d12b4bfca7677d9ef383620d ## Description ## Brief description of what this PR is about - If this change is a backward incompatible change, why must this change be made? - Interesting...
## Description To make it easy for user install DJL serving as a product, we should provide installation package for user to use directly: [x] brew install support for macOS...
## Description DJL serving itself is a web container, it's capable to serve static web page and customized REST API calls. We'd like to allow user to add customized feature...
## Description DJL central is a centralized location to find DJL pre-trained models. It keeps a registry of know ModelZoos and their metadata. It also has a web based UI...
## Description DJL serving support serving arbitrary model by provide tensor in tensor out fashion already. To provide better user experience, DJL provide default pre-processing and post-processing for some of...
## Description In current DJL serving implementation, the registered model information only kept in memory. When DJL serving process is restarted, the last model serving stat is lost. User must...
## Description DJL serving should provide a web base management portal to manage models [ ] Manage models - register a model - un-register a model - update model scaling...
## Description DJL serving documentation should includes: [ ] DJL serving architecture [ ] DJL serving quickstart [ ].DJL serving management API specification [ ] DJL serving inference API specification...
Add a magic that allow user to execute external program in jupyter notebook. Example: %system ls -la
This fixes #80. This might not be the best solution, it assumes DependencyResolver is an instanceof IBiblioResolver (this is not future proof).