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
Hi, I'M Trying To Set Storage Manager To Use Our Internal Miniio Installation But I Ran Into This Issue With This Testing Code:

Hi,
I'm trying to set storage manager to use our internal MiniIO installation but I ran into this issue with this testing code:
` from trains import Task, StorageManager

task = Task.init(project_name="tests", task_name="storage test")
local_iris_pkl = StorageManager.get_local_copy(remote_url="s3://localhost:9000/trains/dataset_test") the result is: trains.storage - ERROR - Could not download s3://localhost:9000/trains/dataset_test , err: SSL validation failed for [SSL: WRONG_VERSION_NUMBER] wrong version number (_ssl.c:1108) I think this is related the fact s3 config set by the injector in the code is always setting secure=True `

just to be sure the issue is not on minio itself I did
` import boto3

s3 = boto3.resource('s3',
endpoint_url=' ',
aws_access_key_id='DEMOaccessKey',
aws_secret_access_key='DEMOsecretKey')
s3.Bucket('trains').download_file('dataset_test.csv', '/tmp/dataset_test.csv') `
and this is working like a charm.

How can I fix this?

  
  
Posted 3 years ago
Votes Newest

Answers 23


JuicyFox94
NICE!!! this is exactly what I had in mind.
BTW: you do not need to put the default values there, basically it reads the defaults from the package itself trains-agent/trains and uses the conf file as overrides, so this section can only contain the parts that are important (like cache location credentials etc)

  
  
Posted 3 years ago

an implementation of this kind is interesting for you or do you suggest to fork

You mean adding a config map storing a default trains.conf for the agent?

  
  
Posted 3 years ago

It is way too much to pass on env variable 😞

  
  
Posted 3 years ago

I think this is great! That said, it only applies when you are spining agents (the default helm is for the server). So maybe we need another one? or an option?

  
  
Posted 3 years ago

an implementation of this kind is interesting for you or do you suggest to fork? I mean, I don't want to impact your time reviewing

  
  
Posted 3 years ago

ok so it' time to create a configmap with the entire file πŸ˜„

  
  
Posted 3 years ago

Nice!
is trainsConfig pure text blob ?

  
  
Posted 3 years ago

basically a new helm chart πŸ˜„

  
  
Posted 3 years ago

Yes πŸ™‚
BTW: do you guys do remote machine development (i.e. Jupyter / vscode-server) ?

  
  
Posted 3 years ago

at that point we define a queue and the agents will take care of training πŸ˜„

  
  
Posted 3 years ago

as usual it starts small and after 5 mins discussion is getting challenging πŸ˜„ I love this stuff... let me think a bit about it I will get back to you asap on this.

  
  
Posted 3 years ago

our data engineer directly write code in pycharm and test it on the fly with brakpoints. when good we simply commit in git and we set a tag "prod ready"

  
  
Posted 3 years ago

an implementation of this kind is interesting for you or do you suggest to fork

You mean adding a config map storing a default trains.conf for the agent?

  
  
Posted 3 years ago

AgitatedDove14 trainsConfig is totally optional and you can put the config file itself in it.. e.g.:
trainsConfig: |- sdk { aws { s3 { key: "" secret: "" region: "" credentials: [ { host: "minio.minio:9000" key: "DEMOaccessKey" secret: "DEMOsecretKey" multipart: false secure: false region: "" } ] } boto3 { pool_connections: 512 max_multipart_concurrency: 16 } } development { default_output_uri: "s3://minio.minio:9000/trains/" } }

  
  
Posted 3 years ago

about helm chart, yes, I mean adding capability of managing a configmap qith config file. If it's interesting I can raise a PR otherwise I need to fork πŸ˜„

  
  
Posted 3 years ago

at that point we define a queue and the agents will take care of trainingΒ 

This is my preferred way as well :)

  
  
Posted 3 years ago

The easiest is to pass an entire trains.conf file

  
  
Posted 3 years ago

(probably it's not possible)

  
  
Posted 3 years ago

Hi JuicyFox94
you pointed to exactly the issue πŸ™‚
In your trains.conf
https://github.com/allegroai/trains/blob/f27aed767cb3aa3ea83d8f273e48460dd79a90df/docs/trains.conf#L94

  
  
Posted 3 years ago

Hi Martin πŸ˜„ ok got it but now the question: how I can pass this to the train-agent deployed with Helm chart?

  
  
Posted 3 years ago

https://github.com/allegroai/trains-server-k8s/pull/13 i think you will like it πŸ˜„

  
  
Posted 3 years ago

Sounds good to me πŸ™‚

  
  
Posted 3 years ago
489 Views
23 Answers
3 years ago
one year ago
Tags