You are viewing a plain text version of this content. The canonical link for it is here.

Posted to discuss-archive@tvm.apache.org by Siyuan Li via Apache TVM Discuss <no...@discuss.tvm.ai> on 2020/09/27 07:24:28 UTC

[Apache TVM Discuss] [Questions] Supporting Faster RCNN and Mask RCNN models


Hi:

I am also trying to compile a SSD model from [TensorFlow 1 Detection Model Zoo](https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf1_detection_zoo.md), but the `relay.frontend.from_tensorflow` fails because some input tensor of `Select` has dynamic shape, which will cause `TypeSolver::Solve error` from `WhereRel` 

Would you mind share more details about how you guys compile Tensorflow Object Detection models, such as Tensorflow version, how to overcome the dynamic shape of input tensor of Select operator, graph_def modification and so on.

Thanks a lot





---
[Visit Topic](https://discuss.tvm.apache.org/t/supporting-faster-rcnn-and-mask-rcnn-models/2064/38) to respond.

You are receiving this because you enabled mailing list mode.

To unsubscribe from these emails, [click here](https://discuss.tvm.apache.org/email/unsubscribe/df333d406d343e365833d672f847000147ac04b04d494d71c4da2c2735b51d10).