Zach Dwiel

remove method GraphManager.emulate_act_on_trainer

4

A comment from the code: ``` # TODO-remove - this is a temporary flow, used by the trainer worker, duplicated from observe() - need to create # an external trainer...

how to parse logged .json files?

3

The `json` files stored in the `experiments` directory are not valid `json` objects. Is there anything preventing us from using valid json? Is there a method available for loading this...

doesn't work at all

3

This used to work all the time and I really enjoyed it, but for some reason it doesn't work anymore. is there anything I can do to debug?

constraints on experiment names

1

``` metadata.name: Invalid value: \"hierarchical_goals\": a DNS-1123 subdomain must consist of lower case alphanumeric characters, '-' or '.', and must start and end with an alphanumeric character (e.g. 'example.com', regex...

keep job_name and output log directory names consistent

1

I'm not sure the best way to do this, but it would be nice to be able to easily keep the name of the job and the name of the...

This might not be the best interface/name, but it would be nice to have a simple method for checking to see if an experiment already exists and adding to it...

For example: ``` client.create_job(experiment, { 'master': { 'noise_eps': 0.5, }, 'sub': { 'noise_eps': 0.1, }, }) ``` These could be serialized as: `MASTER__NOISE_EPS` with double under: `__` separating levels.

Doesn't work: ``` experiment = exp.Experiment(name=name, job_template=job['spec']) client.create_experiment(experiment) job = client.create_job(experiment, parameters) ``` Does: ``` experiment = exp.Experiment(name=name, job_template=job['spec']) experiment = client.create_experiment(experiment) job = client.create_job(experiment, parameters) ``` The error message...

atom: multiline `trail` and `crew`

1

these commands currently only select words on the current line. It would be nice if they spanned the entire file.