Resize and normalize images before model execution. In a measured on-device pipeline, resize took 0.809 ms and normalization took 0.705 ms per image; together with decoding, these stages accounted for 82.3% of the per-image work. section 12.2
EfficientNet-Lite0 on M3
The pipeline decodes JPEG bytes to RGB, resizes to 224×224 with bilinear interpolation, then normalizes to 1CHW using ImageNet mean and standard deviation. Decode averaged 4.205 ms, compared with 0.809 ms for resize and 0.705 ms for normalization. section 12.2
Listing 12.4 Timing each stage and the end-to-end path
import time
import numpy as np
from dataclasses import dataclass
from _common import percentile # shared helper, lives next to the script
@dataclass
class StageTimes:
decode: float = 0.0
resize: float = 0.0
normalize: float = 0.0
infer: float = 0.0
post: float = 0.0
def total(self) -> float:
return self.decode + self.resize + self.normalize + self.infer + self.post
def run_once(predict, path):
t = StageTimes()
t0 = time.perf_counter()
img = _decode(path); t1 = time.perf_counter() # JPEG bytes -> RGB
img = _resize(img); t2 = time.perf_counter() # 224x224 bilinear
x = _normalize(img); t3 = time.perf_counter() # 1CHW, ImageNet mean/std
logits = predict(x); t4 = time.perf_counter() # Core ML forward pass
_ = _post(logits); t5 = time.perf_counter() # argmax
t.decode = (t1 - t0) * 1000.0
t.resize = (t2 - t1) * 1000.0
t.normalize = (t3 - t2) * 1000.0
t.infer = (t4 - t3) * 1000.0
t.post = (t5 - t4) * 1000.0
return t #A
predict = _build_predictor("coreml", model_path) #B
for p in paths[:warmup]: #C
run_once(predict, p)
samples = [run_once(predict, p) for p in paths[warmup:warmup + n]]
stage_mean = {f: float(np.mean([getattr(s, f) for s in samples]))
for f in ("decode", "resize", "normalize", "infer", "post")}
e2e_p95 = percentile([s.total() for s in samples], 0.95) #DSee how the pipeline times decode, resize, normalization, inference, and postprocessing separately, then aggregates stage timings. section 12.2
























