OC_AI_08/CLI_server.py at main · Valkea/OC_AI_08 · GitHub

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
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
#! /usr/bin/env python3
# coding: utf-8

import os
import io
import pathlib
import requests

from flask import Flask, request, url_for, Response

import numpy as np
import matplotlib.pyplot as plt
from matplotlib.backends.backend_agg import FigureCanvasAgg as FigureCanvas
from PIL import Image


# --- API Flask app ---

app = Flask(__name__)
app.secret_key = "super secret key"

base_W = 512
base_H = 256
base_resolution = f"{base_W}x{base_H}"


# ########## DEMO FRONTEND ##########
# This could be a different Flask script totally independant from the API!

# API_URL = "http://0.0.0.0:5000"
API_URL = "http://cityscape-segmentation.herokuapp.com"


@app.route("/")
def index():
    return "The 'CityScape Semantic-segmentation CLIENT' is up."


def get_ids(path):
    ids = []
    for x in path.glob("*_labels.png"):
        path = str(x)
        file = path[path.rfind('/')+1:-11]
        ids.append(file)
    return ids


@app.route("/list/")
def file_list():
    files_path = pathlib.Path('data', 'preprocessed', base_resolution, "val")
    ids = get_ids(files_path)

    fileslist = "".join([f"<p><a href='{url_for('display', pic_id=x)}'>{x}</a></p>" for x in ids])

    global API_URL
    if API_URL is None:
        API_URL = request.url_root
    if API_URL is None:
        API_URL = request.url_root

    return f"""
    <!doctype html>
    <html>
        <head>
            <title>List of available ids</title>
        </head>
        <body>
            <h1>Upload new File</h1>
            <form action={API_URL}/predict/ method=post enctype=multipart/form-data>
                <input type=file name=file>
                <input type=submit value=Upload>
            </form>
            <h1>List of available ids</h1>
            {fileslist}
        </body>
    </html>
    """


def compare_segmentations(img_source, mask_source, mask, iou, dice):

    # mask = np.argmax(predictions, axis=3)[0]

    fig = plt.figure(figsize=(19, 5))
    plt.subplot(1, 3, 1)
    plt.imshow(img_source)
    plt.axis('off')
    plt.title("Source")

    plt.subplot(1, 3, 2)
    plt.imshow(mask)
    plt.axis('off')
    plt.title(f"Predicted mask (IoU={iou:.4f} | Dice:{dice:.4f})")

    plt.subplot(1, 3, 3)
    plt.imshow(mask_source)
    plt.axis('off')
    plt.title("Expected mask")

    plt.tight_layout()
    # plt.show()
    return fig


def mIOU(label, pred, num_classes=8):

    iou_list = list()
    present_iou_list = list()

    for sem_class in range(num_classes):
        pred_inds = (pred == sem_class)
        target_inds = (label == sem_class)
        if target_inds.sum().item() == 0:
            iou_now = float('nan')
        else:
            intersection_now = (pred_inds[target_inds]).sum().item()
            union_now = pred_inds.sum().item() + target_inds.sum().item() - intersection_now
            iou_now = float(intersection_now) / float(union_now)
            present_iou_list.append(iou_now)
        iou_list.append(iou_now)
    return np.mean(present_iou_list)


@app.route("/display/<pic_id>", methods=["GET", "POST"])
def display(pic_id):

    # load base & mask images from the given 'pic_id'
    source_img = Image.open(str(pathlib.Path('data', 'preprocessed', base_resolution, 'val', f"{pic_id}.png")))
    source_mask = Image.open(str(pathlib.Path('data', 'preprocessed', base_resolution, 'val', f"{pic_id}_labels.png")))

    # convert img to bytes for POST action
    img_byte_arr = io.BytesIO()
    source_img.save(img_byte_arr, format='PNG')
    img_byte_arr = img_byte_arr.getvalue()

    # POST the source_img (as a byte array)
    global API_URL
    if API_URL is None:
        API_URL = request.url_root
    files = {'file': img_byte_arr}
    res = requests.post(f'{API_URL}/predict/', files=files)

    # process API response
    predict = np.array(res.json())
    print("CLIENT: response type:", type(predict))
    print("CLIENT: responde shape:", predict.shape)

    # display the 3 images side by side for comparison (with scores)
    # y_true = np.eye(8)[source_mask]
    # y_pred = np.eye(8)[predict]
    y_true2 = np.asarray(source_mask)
    y_pred2 = predict
    iou = float(mIOU(y_true2, y_pred2, 8))
    dice = (2*iou)/(iou+1)
    # print(iou, sm.metrics.iou_score(y_true, y_pred))
    # print(dice, sm.metrics.f1_score(y_true, y_pred))

    fig = compare_segmentations(
            source_img,
            source_mask,
            predict,
            iou,
            dice,
            )

    # return the mask as an image
    output = io.BytesIO()
    FigureCanvas(fig).print_png(output)
    return Response(output.getvalue(), mimetype='image/png')

    return f"""
    <!doctype html>
    <title>Predict display</title>
    <h1>Display result</h1>
    {pic_id}
    """


# ########## START BOTH API & FRONTEND ##########


if __name__ == "__main__":
    current_port = int(os.environ.get("PORT") or 5001)
    app.run(debug=True, host="0.0.0.0", port=current_port)