Quantcast
Channel: OpenCV Q&A Forum - RSS feed
Viewing all articles
Browse latest Browse all 2088

How can I define "scopesToKeep" and "scopesToIgnore" nodes of a frozen model?

$
0
0
I run successfully the following files for the model, which was created using TensorFlow API with Inception v2 for MaskRCNN: https://github.com/opencv/opencv/blob/master/samples/dnn/tf_text_graph_mask_rcnn.py https://github.com/opencv/opencv/blob/master/samples/dnn/tf_text_graph_common.py However, now I created frozen model from keras model with resnet101. Now I want to modify the above code for my model. How can I know or define the "scopesToKeep" and "scopesToIgnore" nodes? In the above file case, you use only one input node e.g., "image_tensor", such as: "graph_def = TransformGraph(graph_def, ['image_tensor'], outNodes, ['sort_by_execution_order'])" But in my model has more than one input node, How can I use here? Will I use all input node such 'input_image', 'input_image_meta', and 'input_anchors'. Anyone help me please? Thanks

Viewing all articles
Browse latest Browse all 2088

Trending Articles