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 Everyone - I’M Trying To Run Trains On An Aws Lambda Function. To Purpose Is Just To Query My Trains Agent For Some Stats [# Of Experiments, Workers, Etc] Using The Backend Api. The Problem Is That Aws Lambda Have A Limitation Of 250Mb For A Codebas

Hey everyone - i’m trying to run trains on an AWS Lambda function. to purpose is just to query my trains agent for some stats [# of experiments, workers, etc] using the backend API. the problem is that AWS Lambda have a limitation of 250MB for a codebase and trains includes very heavy dependencies [like plotly for example]. i tried to package it without plotly but trains checking for it and the process fails. is there any way to use trains in a leaner setting?

  
  
Posted 3 years ago
pic
More
SlipperyDove40
0 × 1
Report
Votes Newest

Answers 6

Hi SlipperyDove40
plotly is about 4Mb... trains about 0.5MB what'd the breakdown of the packages ? This seems far away from 250Mb limit

  
  
Posted 3 years ago
pic
More
AgitatedDove14
0 × 1
Report

the limit is for post install/build by pip.

  
  
Posted 3 years ago
pic
More
SlipperyDove40
0 × 1
Report

SlipperyDove40 I just installed a fresh copy py3.6 and plotly on ubuntu. the entire venv dir is ~86MB

  
  
Posted 3 years ago
pic
More
AgitatedDove14
0 × 1
Report

This is the numbers i’m getting:
du -sh ./plotly
125M ./plotly

du -sh ./numpy
50M ./numpy

  
  
Posted 3 years ago
pic
More
SlipperyDove40
0 × 1
Report

thanks. I’ll try it this way

  
  
Posted 3 years ago
pic
More
SlipperyDove40
0 × 1
Report

BTW: trains-agent is leaner, and does not need plotly. And you can use the APIClient to basically query the entire system, would that be a better solution? See https://github.com/allegroai/trains-agent/blob/master/examples/archive_experiments.py

  
  
Posted 3 years ago
pic
More
AgitatedDove14
0 × 1
Report
Write your answer
491 Views
6 Answers
3 years ago
one year ago
Tags
mlops