Edit model card
YAML Metadata Warning: empty or missing yaml metadata in repo card (https://huggingface.co/docs/hub/model-cards#model-card-metadata)

Yolov7-pose Benchmark results

Benchmark, comparing CPU/iGPU with 3 different processors:

  • i5-1250pe (alder lake)
  • i5-1145G7E (tiger lake)
  • i7-1185G7E (tiger lake)

3 different models:

  • FP16
  • Partial QAT (model is quantized but the head is preserved for accuracy)
  • int8 QAT (model is fully quantized)
Device - Model Precision Average Latency (ms) FPS
i5-1250pe-CPU int8 QAT 11.63 85.10
i7-1185G7E CPU int8 QAT 13.36 74.50
i7-1185G7E iGPU Partial QAT 13.51 73.32
i7-1185G7E iGPU int8 QAT 13.66 72.53
i5-1250pe-CPU Partial QAT 14.14 68.75
i5-1145G7E CPU int8 QAT 16.32 61.01
i7-1185G7E CPU Partial QAT 17.30 60.39
i7-1185G7E iGPU FP16 17.52 56.63
i5-1145G7E iGPU Partial QAT 17.91 55.78
i5-1145G7E iGPU int8 QAT 17.95 55.66
i5-1250pe-iGPU Partial QAT 18.99 51.04
i5-1250pe-iGPU int8 QAT 20.12 49.29
i5-1145G7E CPU Partial QAT 20.28 49.14
i5-1250pe-iGPU FP16 21.25 46.71
i5-1145G7E iGPU FP16 23.86 42.73
i7-1185G7E CPU FP16 37.03 26.96
i5-1250pe-CPU FP16 37.83 26.38
i5-1145G7E CPU FP16 44.27 22.56
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference API
Unable to determine this model's library. Check the docs .