Tutorial 0: Overview of MMFewShot Detection Design of data flow Tutorial 1: Learn about Configs Modify a config through script arguments Config file naming convention An example of TFA FAQ Use intermediate variables in configs Tutorial 2: Adding New Dataset Customize dataset Load annotations from file Load annotations from predefined benchmark Load annotations from another dataset during runtime Use predefined class splits Customize the number of annotations Customize the organization of annotations Customize pipeline Customize a new dataset wrapper Create a new dataset wrapper Update dataset builder Update dataloader builder Update the arguments in model using customize dataset wrapper in config Customize a dataloader wrapper Create a new dataloader wrapper Update dataloader builder Tutorial 3: Customize Models Develop new components Add a new detector Add a new backbone 1. Define a new backbone (e.g. MobileNet) 2. Import the module 3. Use the backbone in your config file Add new necks 1. Define a neck (e.g. PAFPN) 2. Import the module 3. Modify the config file Add new heads Add new loss Customize frozen parameters Customize a query-support based detector 1. Define a new detector 2. Import the module 3. Use the detector in your config file Customize an aggregation layer 1. Define a new aggregator 2. Use the aggregator in your config file 3. Use the aggregator in your model Tutorial 4: Customize Runtime Settings Customize optimization settings Customize optimizer supported by Pytorch Customize self-implemented optimizer 1. Define a new optimizer 2. Add the optimizer to registry 3. Specify the optimizer in the config file Customize optimizer constructor Additional settings Customize training schedules Customize workflow Customize hooks Customize self-implemented hooks 1. Implement a new hook 2. Register the new hook 3. Modify the config Use hooks implemented in MMCV Modify default runtime hooks Checkpoint config Log config Evaluation config