BashirEmran commited on
Commit
405244c
·
verified ·
1 Parent(s): ce59138

scorevision: push artifact

Browse files
Files changed (1) hide show
  1. README.md +3 -2
README.md CHANGED
@@ -3,8 +3,9 @@
3
  - `weights.onnx`: single-scale (stride 8) YOLO-style detector, input 960x960 RGB, output [1, 7, N]
4
  (cx, cy, w, h in input pixels, then per-class sigmoid scores for cup, can, bottle).
5
  - `miner.py`: CPU onnxruntime with 2 intra-op threads, letterbox to 960, per-class NMS, then the set of boxes
6
- with the highest expected score is sent (the numbers are in the `esm` block of `config.json`).
 
7
  Deterministic numpy / onnxruntime only.
8
  - `chute_config.yml`: Chutes runtime with the package versions the miner was tested with.
9
 
10
- p95 latency on 2 pinned CPU cores (AMD EPYC 7542): 91 to 95 ms per 1024x1024 image, PNG decoding included.
 
3
  - `weights.onnx`: single-scale (stride 8) YOLO-style detector, input 960x960 RGB, output [1, 7, N]
4
  (cx, cy, w, h in input pixels, then per-class sigmoid scores for cup, can, bottle).
5
  - `miner.py`: CPU onnxruntime with 2 intra-op threads, letterbox to 960, per-class NMS, then the set of boxes
6
+ with the highest expected score is sent (the numbers are in the `esm` block of `config.json`). Nothing is sent
7
+ when the best candidate scores below `gate` (an image without cups, cans or bottles).
8
  Deterministic numpy / onnxruntime only.
9
  - `chute_config.yml`: Chutes runtime with the package versions the miner was tested with.
10
 
11
+ p95 latency on 2 pinned CPU cores (AMD EPYC 7542): 88 to 92 ms per 1024x1024 image, PNG decoding included.