MCPcopy Create free account
hub / github.com/TencentARC/Track4World / _evaluate_model

Function _evaluate_model

evaluation/flow/eval.py:381–490  ·  view source on GitHub ↗

Generic evaluation loop shared across datasets. Args: model: The model to evaluate. loader: The dataloader. batch_processor_fn: A function that takes (batch, model) and returns: - rgb_images: Input tensor for the model. - gt_data: Dict co

(model, loader, batch_processor_fn, alignment_exp=10, aligned_scene_flow=True)

Source from the content-addressed store, hash-verified

source not stored for this graph (policy: none)

Callers 3

test_kittiFunction · 0.85
test_kubricFunction · 0.85
test_blinkvisionFunction · 0.85

Calls 7

compute_pc_metricsFunction · 0.85
infoMethod · 0.80
infer_pairMethod · 0.80

Tested by 3

test_kittiFunction · 0.68
test_kubricFunction · 0.68
test_blinkvisionFunction · 0.68