Work through most of the cspell warnings in python (#13794)

This commit is contained in:
gtsiam
2024-09-17 10:41:46 -05:00
committed by GitHub
parent 350abda21a
commit edababa88e
20 changed files with 208 additions and 66 deletions
+4 -4
View File
@@ -83,11 +83,11 @@ class HailoDetector(DetectionApi):
self.network_group_params = self.network_group.create_params()
# Create input and output virtual stream parameters
self.input_vstreams_params = InputVStreamParams.make(
self.input_vstream_params = InputVStreamParams.make(
self.network_group,
format_type=self.hef.get_input_vstream_infos()[0].format.type,
)
self.output_vstreams_params = OutputVStreamParams.make(
self.output_vstream_params = OutputVStreamParams.make(
self.network_group, format_type=getattr(FormatType, output_type)
)
@@ -162,8 +162,8 @@ class HailoDetector(DetectionApi):
try:
with InferVStreams(
self.network_group,
self.input_vstreams_params,
self.output_vstreams_params,
self.input_vstream_params,
self.output_vstream_params,
) as infer_pipeline:
input_dict = {}
if isinstance(input_data, dict):
+8 -6
View File
@@ -129,10 +129,10 @@ class OvDetector(DetectionApi):
strides = [8, 16, 32]
hsizes = [self.h // stride for stride in strides]
wsizes = [self.w // stride for stride in strides]
hsize_list = [self.h // stride for stride in strides]
wsize_list = [self.w // stride for stride in strides]
for hsize, wsize, stride in zip(hsizes, wsizes, strides):
for hsize, wsize, stride in zip(hsize_list, wsize_list, strides):
xv, yv = np.meshgrid(np.arange(wsize), np.arange(hsize))
grid = np.stack((xv, yv), 2).reshape(1, -1, 2)
grids.append(grid)
@@ -216,10 +216,12 @@ class OvDetector(DetectionApi):
conf_mask = (image_pred[:, 4] * class_conf.squeeze() >= 0.3).squeeze()
# Detections ordered as (x1, y1, x2, y2, obj_conf, class_conf, class_pred)
dets = np.concatenate((image_pred[:, :5], class_conf, class_pred), axis=1)
dets = dets[conf_mask]
detections = np.concatenate(
(image_pred[:, :5], class_conf, class_pred), axis=1
)
detections = detections[conf_mask]
ordered = dets[dets[:, 5].argsort()[::-1]][:20]
ordered = detections[detections[:, 5].argsort()[::-1]][:20]
for i, object_detected in enumerate(ordered):
detections[i] = self.process_yolo(
+5 -5
View File
@@ -17,7 +17,7 @@ supported_socs = ["rk3562", "rk3566", "rk3568", "rk3576", "rk3588"]
supported_models = {ModelTypeEnum.yolonas: "^deci-fp16-yolonas_[sml]$"}
model_chache_dir = "/config/model_cache/rknn_cache/"
model_cache_dir = "/config/model_cache/rknn_cache/"
class RknnDetectorConfig(BaseDetectorConfig):
@@ -110,7 +110,7 @@ class Rknn(DetectionApi):
if model_matched:
model_props["filename"] = model_path + f"-{soc}-v2.0.0-1.rknn"
model_props["path"] = model_chache_dir + model_props["filename"]
model_props["path"] = model_cache_dir + model_props["filename"]
if not os.path.isfile(model_props["path"]):
self.download_model(model_props["filename"])
@@ -125,12 +125,12 @@ class Rknn(DetectionApi):
return model_props
def download_model(self, filename):
if not os.path.isdir(model_chache_dir):
os.mkdir(model_chache_dir)
if not os.path.isdir(model_cache_dir):
os.mkdir(model_cache_dir)
urllib.request.urlretrieve(
f"https://github.com/MarcA711/rknn-models/releases/download/v2.0.0/{filename}",
model_chache_dir + filename,
model_cache_dir + filename,
)
def check_config(self, config):
+6 -6
View File
@@ -285,14 +285,14 @@ class TensorRtDetector(DetectionApi):
boxes, scores, classes
"""
# filter low-conf detections and concatenate results of all yolo layers
detections = []
detection_list = []
for o in trt_outputs:
dets = o.reshape((-1, 7))
dets = dets[dets[:, 4] * dets[:, 6] >= conf_th]
detections.append(dets)
detections = np.concatenate(detections, axis=0)
detections = o.reshape((-1, 7))
detections = detections[detections[:, 4] * detections[:, 6] >= conf_th]
detection_list.append(detections)
detection_list = np.concatenate(detection_list, axis=0)
return detections
return detection_list
def detect_raw(self, tensor_input):
# Input tensor has the shape of the [height, width, 3]
+1 -1
View File
@@ -26,7 +26,7 @@ def preprocess(tensor_input, model_input_shape, model_input_element_type):
logger.warn(
f"preprocess: tensor_input.shape {tensor_input.shape} and model_input_shape {model_input_shape} do not match!"
)
# cv2.dnn.blobFromImage is faster than numpying it
# cv2.dnn.blobFromImage is faster than running it through numpy
return cv2.dnn.blobFromImage(
tensor_input[0],
1.0 / 255,