-
Notifications
You must be signed in to change notification settings - Fork 0
/
evaluate.py
90 lines (72 loc) · 2.88 KB
/
evaluate.py
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
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
import time
import os
from typing import List
import numpy as np
import PySimpleGUI as sg
import utility
from config import *
class MLData:
"""Example MLData class for data compositiing."""
def __init__(self, path):
self.path = path
self.image = utility.load_image(path)
self.result = None
class StubModel:
"""Stub model that sleeps and returns some stub data."""
def __init__(self) -> None:
time.sleep(4)
print("Model 'initialization' finished")
def load_weights(self, path: str) -> None:
"""Example weight loading that can take some time."""
time.sleep(3)
print(f"Stub model from {path} 'loaded'")
def detect(self, images:List[np.ndarray]):
"""Example detection function that gets multiple images,
and returns multiple results(ex. bounding box).
"""
time.sleep(2)
print("'detection' complete")
return [{"key": "values"}] * len(images)
class Evaluator:
"""Example Evaluator that runs on different thread.
change or inherit as you like for your model.
"""
def __init__(self, window: sg.Window) -> None:
self.fixed_data = "Example data"
self.model = StubModel() # Change to your model.
self.model.load_weights("DATASET_PATH")
self.window = window
self.window.write_event_value(THREAD_EVENT, EVAL_READY) # Send EVAL_READY to main thread.
def try_evaluate(self, image_paths: List[str], results: List):
"""이미지의 경로들을 받아, 경로에 대해 이미지를 불러오고 evaluate하여 Wood 클래스들을 반환하는 메소드
evaluate 시작시와 종료시에 THREAD_EVENT를 통해 window에 메시지를 전달한다.
멀티쓰레딩/프로세싱에 활용할 수 있다.
"""
print("Loading files...")
results.clear()
for path in image_paths:
try:
# Load image
ml_data = MLData(path)
results.append(ml_data)
except FileNotFoundError as e:
print(e)
print("image not found at ", path)
except ValueError as e:
print(e)
print("Invalid file at ", path)
print("Starting evaluation...")
self.window.write_event_value(THREAD_EVENT, EVAL_START)
for ml_data in results:
ml_data.result = self.evaluate(ml_data.image)
self.window.write_event_value(THREAD_EVENT, EVAL_ONE_COMPLETE)
self.window.write_event_value(THREAD_EVENT, EVAL_COMPLETE)
def evaluate(self, image) -> List:
"""example evaluation function for single image.
If you want to use it for multiple images, feel free to change
or create a new version for multiple images
"""
print(image.shape)
r = self.model.detect([image])[0]
print(r)
return r