Galen Han

Using the [Deep Learning Base AMI](https://aws.amazon.com/marketplace/pp/B077GCZ4GR?qid=1517156289784&sr=0-1&ref_=srh_res_product_title), which does not have Caffe2 pre-installed, I was able to build from source without issue.

Thanks for the reply! I'm unfamilar with protocol buffers right now, but briefly read about them just now at https://developers.google.com/protocol-buffers/?hl=en How would I then use tensorboardX to log an instance...

I see, thanks! Unfortunately, I'm developing it for internal company use at the moment. I'll need to see if I can get permission to open-source it once it's more mature.

This is still an issue in v3.4.8

Strange, using the workflow level example I am still seeing the hooks stuck in the `Pending` state. ## Workflow YAML ```yaml apiVersion: argoproj.io/v1alpha1 kind: Workflow metadata: generateName: lifecycle-hook- spec: entrypoint:...