gerbil
gerbil copied to clipboard
Some errors while startup.sh
Hi, Thanks for your help!
I have tried a lot of times run ./startup.sh, and I have some errors here.
on mac os + maven 2.2.1 + java 1.8.0_144
when I run ./startup.sh, it stops here:
I don't why, is it caused by macos?
Thanks for using GERBIL :smiley:
On my systems, there follows a line with something like:
INFO: Starting ProtocolHandler ["http-bio-1234"]
However, please note that the start.sh
script simply starts a Tomcat Server and the console is attached to the log of the tomcat. So it will not show a prompt or something like that, until you stop the server with ctrl+c
(or whatever you use on macos :wink: ). So if nothing is happening, it is possible that the tomcat server is already running and just waits for a user. You could try to open http://localhost:1234/gerbil/ and check whether you can see the UI.
Thanks for your help!
And I have some other problem
The first : when I try the system on github :https://github.com/dalab/end2end_neural_el
it shows
can you help my about this?
The second how should I use gerbil, when I use gerbil on a linux server without the UI? Thanks a lot for your help! 😉
Hey, As i have no mac os on my hands right now i am not sure if i can reproduce the start.sh error but i will check what it might be.
I will test the system locally and check what happens and will let you know whats wrong.
Do you mean with using gerbil without the UI on how to start a experiment and receive its results?
You do not happen to have an online demo of the system? Would spare me a lot of time :)
@TortugaAttack Thanks for your help!
the mac os can run start.sh but it didn't show INFO: Starting ProtocolHandler ["http-bio-1234"]
Fortunately mac os can open http://localhost:1234/gerbil/ and gerbil can be used.
Yes, could I use gerbil without the UI on how to start a experiment and receive its results?
I ‘m sorry that I havn't modify the system so there isn’t a online demo😢
Thanks for all your help! It's very kind of you to help my for these troubles😄
So i debugged it and the answer is parsed correctly but do not contain any markings which resolves into the 0 results with 0 errors. REST services: http://localhost:1234/gerbil/exexcute?experimentData={expData} whereas expData is a JSON object which looks like the following:
{
"type": "D2KB",
"matching": "STRONG_ANNOTATION_MATCH",
"annotator": [
"NIFWS_try1(http://localhost:1235/gerbil-spotWrapNifWS4Test/myalgorithm)"
],
"dataset": [
"ACE2004",
"DBpediaSpotlight"
]
}
http://localhost:1234/gerbil/experiment?id={ID} whereas {ID} is your experiment ID
if this does not work please let me know :)
Thanks for your help!
But i still have the problem as following
Could you tell me which function of end2endel did you use, el or ed ?
Thanks a lot!
EL but i accidentally shut down the EL server. Now it have markings, will look further into why its got 0 as results . There were some correct markings as far as i can tell.
I think it is rather strange that the experiments finished within 2 minutes in your screenshots, where as it takes up to a few hours for me (still not finished) The EL python command uses 65% of my memory (out of 16GB RAM) without doing anything, is that normal?
Thanks~ Which datasets did you use, and which of them has correct markings? Could you show me your markings? My markings are all 0 😢 I used mac os where python3 used 10.65GB(out of 16GB) same as yours.