def predownload(nnDir):
import os
import sys
# change this property
NOMEROFF_NET_DIR = os.path.abspath(nnDir)
sys.path.append(NOMEROFF_NET_DIR)
from NomeroffNet.YoloV5Detector import Detector
detector = Detector()
detector.load()
from NomeroffNet.BBoxNpPoints import NpPointsCraft, getCvZoneRGB, convertCvZonesRGBtoBGR, reshapePoints
npPointsCraft = NpPointsCraft()
npPointsCraft.load()
from NomeroffNet.OptionsDetector import OptionsDetector
from NomeroffNet.TextDetector import TextDetector
from NomeroffNet.TextPostprocessing import textPostprocessing
# load models
optionsDetector = OptionsDetector()
optionsDetector.load("latest")
textDetector = TextDetector.get_static_module("eu")()
textDetector.load("latest")
return npPointsCraft, optionsDetector, textDetector, detector
# Initialize text detector.
textDetector = TextDetector({
"eu_ua_2004_2015": {
"for_regions": ["eu_ua_2015", "eu_ua_2004"],
"model_path": "latest"
},
"eu_ua_1995": {
"for_regions": ["eu_ua_1995"],
"model_path": "latest"
},
"eu": {
"for_regions": ["eu"],
"model_path": "latest"
},
"ru": {
"for_regions": ["ru", "eu-ua-ordlo-lpr", "eu-ua-ordlo-dpr"],
"model_path": "latest"
},
"kz": {
"for_regions": ["kz"],
"model_path": "latest"
},
"ge": {
"for_regions": ["ge"],
"model_path": "latest"
},
"su": {
"for_regions": ["su"],
"model_path": "latest"
}
})
# Initialize text detector.
textDetector = TextDetector({
"eu_ua_2004_2015": {
"for_regions": ["eu_ua_2015", "eu_ua_2004"],
"model_path": "latest"
},
"eu_ua_1995": {
"for_regions": ["eu_ua_1995"],
"model_path": "latest"
},
"eu": {
"for_regions": ["eu"],
"model_path": "latest"
},
"ru": {
"for_regions": ["ru", "eu-ua-ordlo-lpr", "eu-ua-ordlo-dpr"],
"model_path": "latest"
},
"kz": {
"for_regions": ["kz"],
"model_path": "latest"
},
"ge": {
"for_regions": ["ge"],
"model_path": "latest"
},
"su": {
"for_regions": ["su"],
"model_path": "latest"
}
})
detector = Detector()
detector.load()
from NomeroffNet.BBoxNpPoints import NpPointsCraft, getCvZoneRGB, convertCvZonesRGBtoBGR, reshapePoints
npPointsCraft = NpPointsCraft()
npPointsCraft.load()
from NomeroffNet.OptionsDetector import OptionsDetector
from NomeroffNet.TextDetector import TextDetector
from NomeroffNet.TextPostprocessing import textPostprocessing
# load models
optionsDetector = OptionsDetector()
optionsDetector.load("latest")
textDetector = TextDetector.get_static_module("eu")()
textDetector.load("latest")
# Detect numberplate
img_path = 'images/example2.jpeg'
img = cv2.imread(img_path)
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
targetBoxes = detector.detect_bbox(img)
all_points = npPointsCraft.detect(img, targetBoxes,[5,2,0])
# cut zones
zones = convertCvZonesRGBtoBGR([getCvZoneRGB(img, reshapePoints(rect, 1)) for rect in all_points])
# predict zones attributes
regionIds, countLines = optionsDetector.predict(zones)
def main():
args = parse_args()
filepath = args["filepath"]
batch_size = args["batch_size"]
n = args["number_tests"]
# get models
# Initialize text detector.
textDetector = TextDetector({
"eu_ua_2004_2015": {
"for_regions": ["eu_ua_2015", "eu_ua_2004"],
"model_path": "latest"
},
"eu_ua_1995": {
"for_regions": ["eu_ua_1995"],
"model_path": "latest"
},
"eu": {
"for_regions": ["eu"],
"model_path": "latest"
},
"ru": {
"for_regions": ["ru", "eu-ua-ordlo-lpr", "eu-ua-ordlo-dpr"],
"model_path": "latest"
},
"kz": {
"for_regions": ["kz"],
"model_path": "latest"
},
"ge": {
"for_regions": ["ge"],
"model_path": "latest"
},
"su": {
"for_regions": ["su"],
"model_path": "latest"
}
})
# get device
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
print(f"[INFO] device", device)
model_repository_dir = "./model_repository"
for detector, name in zip(textDetector.detectors,
textDetector.detectors_names):
print(f"\n\n[INFO] detector name", name)
model_filepath = filepath.replace("{ocr_name}", name)
# get model and model inputs
model = detector.model
model = model.to(device)
x = torch.randn(batch_size,
detector.color_channels,
detector.height,
detector.width,
requires_grad=True)
x = x.to(device)
# make dirs
p = pathlib.Path(os.path.dirname(model_filepath))
p.mkdir(parents=True, exist_ok=True)
# Export the model
model.to_onnx(
model_filepath,
x,
export_params=
True, # store the trained parameter weights inside the model file
opset_version=10, # the ONNX version to export the model to
do_constant_folding=
True, # whether to execute constant folding for optimization
input_names=[f'inp_{name}'], # the model's input names
output_names=[f'out_{name}'], # the model's output names
dynamic_axes={
f'inp_{name}': {
0: 'batch_size'
}, # variable length axes
f'out_{name}': {
1: 'batch_size'
},
})
# Test torch model
outs = model(x)
start_time = time.time()
for _ in range(n):
outs = model(x)
print(
f"[INFO] torch time {(time.time() - start_time)/n * 1000}ms torch")
# Load onnx model
ort_session = onnxruntime.InferenceSession(model_filepath)
input_name = ort_session.get_inputs()[0].name
ort_inputs = {
input_name:
np.random.randn(batch_size, detector.color_channels,
detector.height, detector.width).astype(np.float32)
}
# run onnx model
print(f"[INFO] available_providers",
onnxruntime.get_available_providers())
ort_outs = ort_session.run(None, ort_inputs)
start_time = time.time()
for _ in range(n):
ort_outs = ort_session.run(None, ort_inputs)
print(
f"[INFO] onnx time {(time.time() - start_time)/n * 1000}ms tensorrt"
)