MNN
MNN copied to clipboard
alibaba
模型调用segmentation fault
你好，我这边模型转换成功没有报错 但是调用的时候，用python API V2版本通过createSession的方式初始化时报 segmentation fault 用python API v3 版本通过MNN.expr调用结果正常 请问这是什么情况呢，模型是pytorch 转onnx再转mnn的，里面有self-attention 谢谢，我可以提供模型和调用代码
现在已经定位到是torch.matmul的问题，加了这个后，原本正常的模型也报segmentation fault了。调用代码如下： def inference_v2(): """ inference mobilenet_v1 using a specific picture """ interpreter = MNN.Interpreter("lstr.mnn") interpreter.setCacheFile('.tempcache') config = {} config['precision'] = 'low' print ('create sess') session = interpreter.createSession() print ('create end') input_tensor = interpreter.getSessionInput(session, 'image') mask_tensor = interpreter.getSessionInput(session, 'mask') print ('create input') image = cv2.imread('cuts_1205_000273.jpg') input_size = (360, 640) height, width = image.shape[0:2]
images = np.zeros((1, 3, input_size[0], input_size[1]), dtype=np.float32)
masks = np.ones((1, 1, input_size[0], input_size[1]), dtype=np.float32)
pad_image = image.copy()
pad_mask = np.zeros((height, width, 1), dtype=np.float32)
resized_image = cv2.resize(pad_image, (input_size[1], input_size[0]))
resized_mask = cv2.resize(pad_mask, (input_size[1], input_size[0]))
masks[0][0] = resized_mask.squeeze()
resized_image = resized_image / 255.
resized_image -= np.array([0.40789654, 0.44719302, 0.47026115], dtype=np.float32)
resized_image /= np.array([0.28863828, 0.27408164, 0.27809835], dtype=np.float32)
resized_image = resized_image.transpose(2, 0, 1)
images[0] = resized_image
tmp_input = MNN.Tensor((1, 3, input_size[0], input_size[1]), MNN.Halide_Type_Float,\
images, MNN.Tensor_DimensionType_Caffe)
tmp_mask = MNN.Tensor((1, 1, input_size[0], input_size[1]), MNN.Halide_Type_Float,\
masks, MNN.Tensor_DimensionType_Caffe)
input_tensor.copyFrom(tmp_input)
mask_tensor.copyFrom(tmp_mask)
print ('start infer')
interpreter.runSession(session)
output_tensor = interpreter.getSessionOutput(session,'aux_outputs')
curves_tensor = interpreter.getSessionOutput(session,'aux_curves')
outputs = np.array(output_tensor.getData())
curves = np.array(curves_tensor.getData())
outputs = outputs.reshape(-1, 3)
curves = curves.reshape(-1, 8)
result = post_process(outputs, curves)
pred = result
img = pad_image
img_h, img_w, _ = img.shape
pred = pred[pred[:, 0].astype(int) == 1]
overlay = img.copy()
color = (0, 255, 0)
for i, lane in enumerate(pred):
lane = lane[1:] # remove conf
lower, upper = lane[0], lane[1]
lane = lane[2:] # remove upper, lower positions
# generate points from the polynomial
ys = np.linspace(lower, upper, num=100)
points = np.zeros((len(ys), 2), dtype=np.int32)
points[:, 1] = (ys * img_h).astype(int)
points[:, 0] = ((lane[0] / (ys - lane[1]) ** 2 + lane[2] / (ys - lane[1]) + lane[3] + lane[4] * ys -
lane[5]) * img_w).astype(int)
points = points[(points[:, 0] > 0) & (points[:, 0] < img_w)]
# draw lane with a polyline on the overlay
for current_point, next_point in zip(points[:-1], points[1:]):
overlay = cv2.line(overlay, tuple(current_point), tuple(next_point), color=color, thickness=5)
# draw lane ID
if len(points) > 0:
cv2.putText(img, str(i)+'_%.2f'%1.0, tuple(points[-1]), fontFace=cv2.FONT_HERSHEY_SIMPLEX, fontScale=1,
color=color,
thickness=3)
# Add lanes overlay
w = 0.6
img = ((1. - w) * img + w * overlay).astype(np.uint8)
cv2.imshow('test',img)
cv2.waitKey(0)
