File size: 1,776 Bytes
759bd70
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
"""Next Federation: load the released global model (FedProx) or rerun the federated simulation.

Usage:
    python inference.py --selftest
    from inference import load, predict
"""
import json
import os
import sys

HERE = os.path.dirname(os.path.abspath(__file__))
sys.path.insert(0, HERE)

import numpy as np  # noqa: E402

import fed_core as fc  # noqa: E402
from nlrt import forward, load_model  # noqa: E402


def load(path=os.path.join(HERE, "model.json")):
    return load_model(path)


def predict(spec, pixels):
    """pixels: (N, 64) ints 0..16. Returns class probabilities (N, 10) from the global model."""
    z = forward(spec["nets"]["global"], np.asarray(pixels, dtype=float) / 16.0)
    z = z - z.max(axis=1, keepdims=True)
    e = np.exp(z)
    return e / e.sum(axis=1, keepdims=True)


def _selftest():
    spec = load()
    with open(os.path.join(HERE, "selftest.json"), encoding="utf-8") as f:
        t = json.load(f)
    P = predict(spec, t["pixels"])
    diff = float(np.abs(P - np.asarray(t["expected_proba"])).max())
    # rerun a tiny federated simulation and compare with stored output
    X = np.asarray(t["sim"]["pixels"], dtype=float) / 16.0
    y = np.asarray(t["sim"]["y"])
    parts = fc.partition(y.tolist(), 4, 0.5, 2)
    gt = list(range(len(y)))
    r = fc.run(X, y, parts, X[gt], y[gt], "fedprox", rounds=2, seed=2, **spec["meta"]["hyper"])
    sim_diff = float(np.abs(np.asarray(r["curve"]) - np.asarray(t["sim"]["expected_curve"])).max())
    ok = diff < 1e-6 and sim_diff < 1e-9
    print(json.dumps({"passed": ok, "max_abs_diff": diff, "sim_curve_diff": sim_diff, "n_samples": len(t["pixels"])}))
    return ok


if __name__ == "__main__":
    if "--selftest" in sys.argv:
        sys.exit(0 if _selftest() else 1)
    print(__doc__)