Examples: query, "exact match", wildcard*, wild?ard, wild*rd
Fuzzy search: cake~ (finds cakes, bake)
Term boost: "red velvet"^4, chocolate^2
Field grouping: tags:(+work -"fun-stuff")
Escaping: Escape characters +-&|!(){}[]^"~*?:\ with \, e.g. \+
Range search: properties.timestamp:[1587729413488 TO *] (inclusive), properties.title:{A TO Z}(excluding A and Z)
Combinations: chocolate AND vanilla, chocolate OR vanilla, (chocolate OR vanilla) NOT "vanilla pudding"
Field search: properties.title:"The Title" AND text
Answered
Hey Trains Riders, This Must Be Something Simple I Am Missing, But Still I Couldn'T Realize What The Problem Is. I Am Trying To Run Trains-Agent On My Experiments. Setup Of The Server And The Agent Is Fine, But I Am Struggling To Run Real Experiments (Not

Hey Trains riders,
This must be something simple I am missing, but still I couldn't realize what the problem is.
I am trying to run trains-agent on my experiments.
Setup of the server and the agent is fine, but I am struggling to run real experiments (not samples).
The current problem I am facing is that when a task is executed by the agent, I get the following error:
import Xxx.Yyy as Yyy
ModuleNotFoundError: No module named Xxx

Please note:
Xxx is the module where my main experiment script resides.
In my requirements file I pip install Xxx as: "pip install -e ."

Any ideas?

  
  
Posted 3 years ago
Votes Newest

Answers 9


just doubles the setup time...

  
  
Posted 3 years ago

Is it possible, instead of reinstalling the packages, to simply switch to a different existing virtualenv, at the beginning of the task execution?

  
  
Posted 3 years ago

Hi ColossalDeer61 ,

Xxx is the module where my main experiment script resides.

So I think there are two options,
Assuming you have a similar folder structure-main_folder
--package_folder
--script_folder
---script.py
Then if you set the "working directory" in the execution section to "." and the entry point to "script_folder/script.py", then your code could do:
from package_folder import ABC
2. After cloning the original experiment, you can edit the "installed packages", and add a new line with:
git+ https://github.com/user/repo/repo.git
This is the equivalent of doing pip install git+ https://github.com/user/repo/repo.git

What do you think?

  
  
Posted 3 years ago

Manually editing the requirements is indeed a step forward, thanls.
The current challenge is installing with "--no-binary" flag

For example:
h5py==2.10.0 --no-binary=h5py

  
  
Posted 3 years ago

hmm interesting use case, why do you need to add the "--no-binary"

  
  
Posted 3 years ago

Another (minor) issue is that all the packages that are installed using git+https are cloned and installed twice, immediately one after the other

  
  
Posted 3 years ago

In this case local changes are made that can impact these libraries.
I am not sure it is a real problem in the trains scenario

  
  
Posted 3 years ago

Hi ColossalDeer61 ,
the next trains-agent RC (solving the #196 issue) will also solve the double install issue 🙂

  
  
Posted 3 years ago

Another (minor) issue is that all the packages that are installed using git+https are cloned and installed twice, immediately one after the other

Yes this is so that we can better log the installed package name, not a major issue, but we just fixed a bug with derivative packages from git packages.
https://github.com/allegroai/trains/issues/196

  
  
Posted 3 years ago