Reputation
Badges 1
57 × Eureka!Interesting... the code looks similar... i really don't know why it didn't showed the models in my trainings... I will try to add it manually for now
AgitatedDove14 In my train I see under this section " NO ARTIFACTS RECORDED"
so I didn't have much time to upgrade all the packs because I have some issues with that but it is on my todo list
TF 1.15 (yes I know... i work on an old model of maskrcnn)
yes, I can see the model for his training in the web
we decided to buy the full version but I can't find the documentation on the sections that are only included in the full version, I suppose that the full version includes more options and i want to read about them and see the documantation
If the link is visible only after I buy it then I can't see it right now (iI wanted to read about it before) if it is visible on the free trains allegro then I can't find it :/
@ https://app.slack.com/team/UT8T0V3NE ok so I see the packages that are listed are the ones that are imported on the python script that runs but it doesn't include the ones it needs from imports from other inner classes... I guess it is the way it works, so i need to specify all the packages that are not on the py script i run? (all the packages that the inner classes are using)
OK I guess you ment the installed packages on the execution tab and i looked at the log.. so there i see the following:
Python 3.7.5 (default, Nov 7 2019, 10:50:52) [GCC 8.3.0]
argparse_utils == 1.3.0
trains == 0.14.3
Detailed import analysis
**************************
IMPORT PACKAGE argparse_utils
simple_pipe2.py: 10
IMPORT PACKAGE trains
simple_pipe2.py: 9
@ https://app.slack.com/team/UT8T0V3NE any idea for what is the reason they are not listed there?
so yea... the packages are not there...
(with the right branch, commit and all...)
in the clone I see "Environment setup completed successfully" and nothing about timeout
in the "original one"'s log i see:
TRAINS Task: created new task id=2eec56cb60e9441897a4af9c10f656c0
TRAINS new package available: UPGRADE to v0.15.0 is recommended!
Release Notes:
Features
- Add automation support including hyper-parameters optimization (see example here )
Task.init()
auto_connect_arg_parser
argument can accept a dictionary disabling specific keys from the argparser (Tr...
I ran it till the end first time and it ended successfully but i put the trains commend line on comment so i didn't see it on trains,,, so i ran it again and stoped it at the middle...
if i stopped the running of the Original one before it ended, can it be the problem?
and it continues... but nothing about time out
ok, thank you 🙂 i will check it
*I see there is 0.16.1