Tian Jin

Any update for this matter? I have implemented something similar in Python and would be glad to work on a pull request for this feature.

As you can see the verbosity is set to 1, it's the same value when I trained Resnet50 on ImageNet.

@reedwm thanks, I could look into this one, so the graph definition is also written to the tfevent logs? My concern is whether this solves huge tfevent file problem...

@jsimsa can you shed some light into this situation? Since your name appeared in Cifar10ImagePreprocessor class...

@jsimsa , I wonder if doing what @reedwm suggests could also solve the problem for huge tfevents file?

@reedwm thanks a lot for looking into this issue, I'll see what I can do.

Can we document the usage of onnx_op and tf_op and what these information will be used? I find it quite important for future developers.

@AlexandreEichenberger there seems to be a really easy fix, if you just put all the unoptimized loops in a single paranthesis. ```mlir module { func @test_gemm_big(%arg0: memref, %arg1: memref, %arg2:...

You might find this tutorial helpful: http://onnx.ai/onnx-mlir//doxygen_html/index.html

@CHUNERnubi i'm not sure what's stopping you from seeing the intermediate results of gradual lowering. Do you mean that you want to "edit" the intermediate results and then continue? (which...