executing-plans
executing-plans
Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available
configuration-model-zooChoose, inspect, modify, and validate MMDetection 3.3.0 configs and model-zoo entries without training or inference side effects.
Independent GitHub snapshot · Automated classification · Original documentation governs
Use this sub-skill when an agent needs to select an MMDetection model family, inspect a config, resolve _base_ inheritance, apply safe config overrides, or decide between MIM model names and config paths.
model, dataloaders, pipelines, evaluator, runtime hooks, default_scope, or auto_scale_lr settings before running anything expensive.--cfg-options or equivalent Config.merge_from_dict semantics._base_, config import/registry-scope, checkpoint/config pairing, or missing auto_scale_lr.base_batch_size failures.training-testing.DetInferencer, init_detector, inference_detector, visualization outputs, and CPU/GPU inference execution: use inference-visualization.customization-extension.python sub-skills/configuration-model-zoo/scripts/inspect_config.py CONFIG.py --summary --cfg-options model.test_cfg.rcnn.score_thr=0.3
The script reads configs with MMEngine, supports --cfg-options, prints a concise summary by default, and can dump the merged config to .py, .json, or .yml.