def generate(self):
model_path = os.path.expanduser(self.model_path)
assert model_path.endswith(
'.h5'), 'Keras model or weights must be a .h5 file.'
# 计算总的种类
self.num_classes = len(self.class_names) + 1
# 载入模型，如果原来的模型里已经包括了模型结构则直接载入。
# 否则先构建模型再载入
self.model_rpn, self.model_classifier = frcnn.get_predict_model(
self.config, self.num_classes)
self.model_rpn.load_weights(self.model_path, by_name=True)
self.model_classifier.load_weights(self.model_path,
by_name=True,
skip_mismatch=True)
print('{} model, anchors, and classes loaded.'.format(model_path))
# 画框设置不同的颜色
hsv_tuples = [(x / len(self.class_names), 1., 1.)
for x in range(len(self.class_names))]
self.colors = list(map(lambda x: colorsys.hsv_to_rgb(*x), hsv_tuples))
self.colors = list(
map(lambda x: (int(x[0] * 255), int(x[1] * 255), int(x[2] * 255)),
self.colors))
def load_rcnn_model(self):
'''
加载分开预测的载模型
Returns
-------
'''
model_path = os.path.expanduser(self.model_path)
assert model_path.endswith(
'.h5'), 'Keras model or weights must be a .h5 file.'
try:
# 若是完整的模型,直接载入,不需网络结构,也不需要编译.
model_rpn = load_model(model_path, compile=False)
model_classifier = load_model(model_path, compile=False)
except Exception as v:
print(v)
# 包含背景,所以要+1
num_classes = len(self.class_names) + 1
try:
model_rpn, model_classifier = get_predict_model(
self.config, num_classes)
model_rpn.load_weights(self.model_path, by_name=True)
model_classifier.load_weights(self.model_path, by_name=True)
except Exception as e:
print('模型加载失败:', e)
raise
return model_rpn, model_classifier
def generate(self):
model_path = os.path.expanduser(self.model_path)
assert model_path.endswith(
'.h5'), 'Keras model or weights must be a .h5 file.'
#-------------------------------#
# Calculate the total number of classes
#-------------------------------#
self.num_classes = len(self.class_names) + 1
#-------------------------------#
# Loading model and weight
#-------------------------------#
self.model_rpn, self.model_classifier = frcnn.get_predict_model(
self.config, self.num_classes)
self.model_rpn.load_weights(self.model_path, by_name=True)
self.model_classifier.load_weights(self.model_path, by_name=True)
print('{} model, anchors, and classes loaded.'.format(model_path))
# Set different colors for the frame
hsv_tuples = [(x / len(self.class_names), 1., 1.)
for x in range(len(self.class_names))]
self.colors = list(map(lambda x: colorsys.hsv_to_rgb(*x), hsv_tuples))
self.colors = list(
map(lambda x: (int(x[0] * 255), int(x[1] * 255), int(x[2] * 255)),
self.colors))
def generate(self):
model_path = os.path.expanduser(self.model_path)
assert model_path.endswith('.h5'), 'Keras model or weights must be a .h5 file.'
#-------------------------------#
# 载入模型与权值
#-------------------------------#
self.model_rpn, self.model_classifier = frcnn.get_predict_model(self.num_classes, self.backbone)
self.model_rpn.load_weights(self.model_path, by_name=True)
self.model_classifier.load_weights(self.model_path, by_name=True)
print('{} model, anchors, and classes loaded.'.format(model_path))