Closed robinvanemden closed 3 years ago
Thanks for the PR.
I noticed that in the generated test case the concat node lacks the output tensor given as a parameter. The tests still pass, since there is a global reserved tensor of the same name (it is intended that a pointer to this is passed in as a parameter).
There are some pytorch-generated tests in the backend test suite too, one of which tests concat - and this one fails:
ONNX_type_test(operator_concat2 ${ONNX_BACKEND_TEST_DATA_DIR}/pytorch-operator/test_operator_concat2 ONNX_backend_pytorch_conv 0.00001 0)
Also, the generated code lacks some indentations, which make it a bit difficult to read. Tabs are preferred, since that allows the reader to adjust the indentation depth to their preference, and keeps the indentation consistent across nodes.
Thanks for the feedback. I will make the suggested changes asap.
I made the requested changes and cleaned up the code some more. Thanks again for the review!
All ONNX backend tests pass. First op that contains variadic inputs.