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
Profile picture
PompousParrot44
Moderator
32 Questions, 84 Answers
  Active since 10 January 2023
  Last activity 8 months ago

Reputation

0

Badges 1

84 × Eureka!
0 Votes
12 Answers
276 Views
0 Votes 12 Answers 276 Views
for the frameworks which are supported in built, trains stores the trained model as output model e.g. for xgboost here https://github.com/allegroai/trains/bl...
3 years ago
0 Votes
0 Answers
221 Views
0 Votes 0 Answers 221 Views
is is in planning or already supported
3 years ago
0 Votes
0 Answers
231 Views
0 Votes 0 Answers 231 Views
currently when i run script manually.. script runs outside of trains-agent although it reports the progress back to server
3 years ago
0 Votes
2 Answers
223 Views
0 Votes 2 Answers 223 Views
is there a document which describes what kind of data is stored in elasticsearch, mongodb and redis.. just trying to understand the architecture of trains an...
3 years ago
0 Votes
0 Answers
191 Views
0 Votes 0 Answers 191 Views
i have tried quoting it like this python -m 'test.scripts.script' but that doesn't work either
2 years ago
0 Votes
12 Answers
205 Views
0 Votes 12 Answers 205 Views
in UI under execution tab, i see that the trains has script path listed as -m test.scripts
2 years ago
0 Votes
0 Answers
191 Views
0 Votes 0 Answers 191 Views
hmm it seems the plot are ending up under debug samples
3 years ago
0 Votes
7 Answers
182 Views
0 Votes 7 Answers 182 Views
3 years ago
0 Votes
2 Answers
191 Views
0 Votes 2 Answers 191 Views
i have a script in which i added Task.init() hook, so when i run the script, it creates the experiment and i can its logs there etc.. but init() function doe...
3 years ago
0 Votes
6 Answers
186 Views
0 Votes 6 Answers 186 Views
2 years ago
0 Votes
3 Answers
234 Views
0 Votes 3 Answers 234 Views
Hi.. wondering if it is possible to schedule experiments run at a given time e.g. cron schedule or it needs to be triggered from UI/API
3 years ago
0 Votes
0 Answers
184 Views
0 Votes 0 Answers 184 Views
and the above error is caused by routine update check being blocked...
3 years ago
0 Votes
0 Answers
193 Views
0 Votes 0 Answers 193 Views
so wondering why trains is not picking up the whole path
2 years ago
0 Votes
14 Answers
201 Views
0 Votes 14 Answers 201 Views
when running in cpu-only mode, is it possible to restrict the amount of cpu given to an agent or it takes all the available cpus on the machine... similarly ...
3 years ago
0 Votes
25 Answers
200 Views
0 Votes 25 Answers 200 Views
colors of CM reporting are strange... is it possible to adjust the default ones
3 years ago
0 Votes
0 Answers
219 Views
0 Votes 0 Answers 219 Views
as i can not read the numbers at all
3 years ago
0 Votes
0 Answers
219 Views
0 Votes 0 Answers 219 Views
i am using 0.16.1 version of server and 0.16.2 version of agent
2 years ago
0 Votes
0 Answers
222 Views
0 Votes 0 Answers 222 Views
when i am trying to see the "Plots" under result section.. i am seeing this error in browser console ERROR TypeError: e['trains-server'] is undefined and the...
3 years ago
0 Votes
8 Answers
193 Views
0 Votes 8 Answers 193 Views
another question is if i have a conda env available on my workers systemwide.. can i use that env directly when running tasks with trains-agent or trains-age...
3 years ago
0 Votes
8 Answers
261 Views
0 Votes 8 Answers 261 Views
looking at the docs.. i couldn't find a way to cleanup the experiments... only archive them... i also noticed AgitatedDove14 mentioned the https://github.com...
3 years ago
0 Votes
0 Answers
191 Views
0 Votes 0 Answers 191 Views
whereas plots are on fileserver and i can see them by opening the files manually in the editor
3 years ago
0 Votes
0 Answers
218 Views
0 Votes 0 Answers 218 Views
i am thinking that when i run my training script, on the terminal it should say that the task has been created in the server and put it in X queue.. you can ...
3 years ago
0 Votes
3 Answers
197 Views
0 Votes 3 Answers 197 Views
so if there a way to make it run in the trains-agent and not in the terminal directly when executed manually through script
3 years ago
0 Votes
2 Answers
202 Views
0 Votes 2 Answers 202 Views
i am seeing issue when running a script with command as python -m test.scripts.script to be scheduled as task in trains. Trains doesn't pickup the last scrip...
2 years ago
0 Votes
11 Answers
207 Views
0 Votes 11 Answers 207 Views
when trains-agent is configured with system_site_packages: true why agent tries to create a new env and install packages ?
3 years ago
0 Votes
2 Answers
229 Views
0 Votes 2 Answers 229 Views
given these are settled.. another question i have is about the job scheduling based on cron style.. e.g. run training every night at 2 AM etc.
3 years ago
0 Votes
8 Answers
202 Views
0 Votes 8 Answers 202 Views
i know i can run this manually in step by step but wondering if this can be automated as scheduled tasks
3 years ago
0 Votes
3 Answers
200 Views
0 Votes 3 Answers 200 Views
is there a link which describes the differences in community and enterprise versions
3 years ago
0 Votes
0 Answers
207 Views
0 Votes 0 Answers 207 Views
is it possible to create workflow using trains... e.g. nightly fetch data from data warehouse.. run Task.TaskTypes.data_processing to do ETL one this task is...
3 years ago
0 Votes
23 Answers
213 Views
0 Votes 23 Answers 213 Views
wondering why my plots are not in plot section but under debug section
3 years ago
Show more results questions
3 years ago
0 Colors Of Cm Reporting Are Strange... Is It Possible To Adjust The Default Ones

i am simply proxying it using ssh port forwarding

3 years ago
0 Colors Of Cm Reporting Are Strange... Is It Possible To Adjust The Default Ones

also one thing i noticed.. when i report confusion matrix and some other plots e.g. seaborn with matplotlib.. on server side i can the plots are there but not visible at all

3 years ago
0 Colors Of Cm Reporting Are Strange... Is It Possible To Adjust The Default Ones

trains is run using docker-compose allegroai/trains-agent-services:latest and allegroai/trains:latest

3 years ago
0 Colors Of Cm Reporting Are Strange... Is It Possible To Adjust The Default Ones

this is when executed from directly with task.init()

3 years ago
3 years ago
0 For The Frameworks Which Are Supported In Built, Trains Stores The Trained Model As Output Model E.G. For Xgboost Here

AgitatedDove14 when using OutputModel(task, name='LightGBM model', framework='LightGBM').update_weights(f"{args.out}/model.pkl") i am seeing this in the logs No output storage destination defined, registering local model /tmp/model.pkl when i got to trains UI.. i see the model name and details but when i try to download it point to the path file:///tmp/model.pkl which is incorrect wondering how to fix it

3 years ago
0 For The Frameworks Which Are Supported In Built, Trains Stores The Trained Model As Output Model E.G. For Xgboost Here

AgitatedDove14 it seems uploading artifact and uploading models are two different things when it comes to treating fileserver... as when i upload artifact it works as expected but when uploading model using outputmodel class, it wants output_uri path.. wondering how can i as it to store it under the fileserver like artifacts LightGBM.1104445eca4749f89962669200481397/artifacts/Model%20object/model.pkl

3 years ago
0 I Know I Can Run This Manually In Step By Step But Wondering If This Can Be Automated As Scheduled Tasks

looking at the above link, it seems i might be able to create it with some boilerplate as it has concept of parent and child... but not sure how status checks and dependency get sorted out

3 years ago
0 Wondering Why My Plots Are Not In Plot Section But Under Debug Section

yeah i still see it.. but that seems to be due to dns address being blocked by our datacenter

3 years ago
0 Wondering Why My Plots Are Not In Plot Section But Under Debug Section

as i am seeing now my plots but they are lending into metrics section not plot section.

3 years ago
0 Wondering Why My Plots Are Not In Plot Section But Under Debug Section

so as you say.. i don't think the issue i am seeing is due to this error

3 years ago
3 years ago
3 years ago
0 Wondering Why My Plots Are Not In Plot Section But Under Debug Section

simply changing to show doesn't work in my case as i am displaying CM.. what about if i use matshow

3 years ago
0 Wondering Why My Plots Are Not In Plot Section But Under Debug Section

it may be that i am new to trains but in my normal notebook flow they both are images and i as trains user expected it to be under the Plot section as i think this is an image.. as in nutshell all matplotlib plots display data as an image 🙂

3 years ago
0 Wondering Why My Plots Are Not In Plot Section But Under Debug Section

TimelyPenguin76 also is there any reason for trating show and imshow differently

3 years ago
3 years ago
0 Colors Of Cm Reporting Are Strange... Is It Possible To Adjust The Default Ones

its not that they are blank.. whole page is blank including plotly plots

3 years ago
3 years ago
0 Wondering Why My Plots Are Not In Plot Section But Under Debug Section

TimelyPenguin76 yeah when i run matplotlib with show plots does land under Plot section... so its mainly then the imshow part.. i am wondering why the distinction and what is the usual way to emit plots to debug samples

3 years ago
Show more results compactanswers