Ian Quah

> even if it means duplicating code from TF Unless Edward duplicates large chunks of TF I don't know if that's the best practice here. While I was looking through...

I'm swamped with stuff but can probably look into that this weekend / early next week. I'll keep you updated on what I find when I look into ```tf.contrib.graph_editor```

@CMoebus I'm not familiar with Anaconda so I can't help you there but if it's possible to specify versions, you want to use Tensorflow 1.6 or prior with Edward. @dustinvtran...

#891 and #882 The ```set_shapes_for_outputs``` function seems to have been deprecated as of 1.7

Sorry to necro an old thread but I tried looking through the road map and could not find it. What's the status of this? Canceled, on hold, or completed? Thank...

I think it does show how one can do it? Check out this [colab](https://colab.research.google.com/drive/1aZJUjQ9wZkNFwJxu5Jf9ba8_C5hrF7Yr?usp=sharing) link that I made. Basically, `forward.apply` does the inference for you

Works just fine on my Linux machine

Nope, it hangs :( It stalls on the instantiation line, `gp_model = gpb.GPModel(group_data=group, likelihood=likelihood)`

Ahh, gotcha! I'm running into ``` Exception: Please install CMake and all required dependencies first The full version of error log was saved into /Users/ianquah/GPBoost_compilation.log ``` when doing an `Installation...

No worries! I ended up reading through the examples and found that most things just revolved around ``` Foo = jnp.ndarray ```