QueensGambit commented on issue #17827: [Numpy] Bugfix of slice operator export (MXNet to ONNX) URL: https://github.com/apache/incubator-mxnet/pull/17827#issuecomment-609433068 Hi, there is actually an additional problem: The function `create_onnx_graph_proto()` assumes that `in_shape` remains fixed throughout the whole network: https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/contrib/onnx/mx2onnx/export_onnx.py#L244 ```python converted = MXNetGraph.convert_layer( node, is_input=False, mx_graph=mx_graph, weights=weights, in_shape=in_shape, in_type=in_type, proc_nodes=all_processed_nodes, initializer=initializer, index_lookup=index_lookup, idx=idx ) ``` I suggest to infer the current shape before each layer and to refactor `create_onnx_graph_proto()`.
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
