i have been asking a lot of questions, i hope its ok and not too troublesome AgitatedDove14 and TimelyPenguin76 I like the product, just trying to setup for ...

I am wondering is it possible to schedule a task to run at certain time in periodic fashion aka. cron style... thinking of having a monitoring task to be run...

is it possible to link independent training experiments.. for example.. i have an ensemble of 2 models (A & B) each models are trained under their own traini...

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...

is there a link which describes the differences in community and enterprise versions

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...

so if there a way to make it run in the trains-agent and not in the terminal directly when executed manually through script

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 ...

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...

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...

colors of CM reporting are strange... is it possible to adjust the default ones

when trains-agent is configured with system_site_packages: true why agent tries to create a new env and install packages ?

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

i know i can run this manually in step by step but wondering if this can be automated as scheduled tasks

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...

whereas plots are on fileserver and i can see them by opening the files manually in the editor

i have tried quoting it like this python -m 'test.scripts.script' but that doesn't work either

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...

so wondering why trains is not picking up the whole path

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...

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.

is is in planning or already supported

i am using 0.16.1 version of server and 0.16.2 version of agent

wondering why my plots are not in plot section but under debug section

hmm it seems the plot are ending up under debug samples

as i can not read the numbers at all

currently when i run script manually.. script runs outside of trains-agent although it reports the progress back to server

wondering what is causing it

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...

and the above error is caused by routine update check being blocked...