Trying to run PL

I’m trying to get PL running, I’m seeing a few different errors, and I’m not sure if they’re related to version issues or if it’s because I’m on Apple M1.

The terminal prints this many times over:

PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001

No matter which data collection I pick, when I try to start a new model I get the error modal saying:

Error: Couldn't infer data types due to an exception: Error: Network Error

My system specs:
macOS Big Sir version 11.0.1 (20B29)
MacBook Air (M1, 2020) Apple M1
Python 3.8.12 (also tried 3.7)
PerceptiLabs 0.13.5
scipy version 1.4.1

I tried installing scipy 1.4.1 because someone recommended it as a solution, but this didn’t seem to change anything. Not sure what else to try now.

Hi @rmrfetc

Nice to welcome another new face! I think you’ll find the PL guys very responsive and I expect they’ll get you up and running soon. I thought I’d just get in first with a simple “Hi!”, “Don’t Panic” and some basic you’re-not-alone-here: I’ve had great help from @robertl and there’s a lot of ongoing chat and to-and-fro about the tool and specific models. I hope you’ll find what you’re looking for here :slight_smile:

I just use PL and don’t know all the technical stuff @robertl does but I do recall this question in the forum relating to M1 issues around Apple’s TensorFlow fork - but network issues aren’t necessarily covered by that!

That said, my terminal only prints this when it (successfully) starts up.

PerceptiLabs: Starting PerceptiLabs: Waiting for services to listen on these ports: PerceptiLabs: rendering-kernel on port 5001 PerceptiLabs: PerceptiLabs Started PerceptiLabs: PerceptiLabs is running at http://localhost:8080/?token=gVl2D_oMzVh29SWLO2LDktWibgnxh7At-tkeMaQvW3U PerceptiLabs: Use Control-C to stop this server and shut down all PerceptiLabs processes.

I see you also tried python 3.7, which rules out my suggestion to back off to 3.8.10 (which I run, though I can’t recall why that specific version I have an idea I tried later and did not succeed with PL at that time)

My only practical suggestion is to restart with the command line perceptilabs -v =3 to get more verbose output you can post here to help the PL guys with a bit more info.

Once you are up and running, what are you planning to do with the tool? Got any special projects in mind or just checking out the whole ML thing?

1 Like

Hi @JulianSMoore . Thanks very much for the warm welcome. Can’t wait to get started building my own models :slight_smile:.

I had run PL with -v=3 but I didn’t see anything of concern and disregarded it. But maybe I missed something so here’s that output:

(pyenv3-7) MacBook-Air:~ rob$ perceptilabs -v=3
Operations to perform:
  Apply all migrations: admin, api, auth, contenttypes, sessions
Running migrations:
  No migrations to apply.
  Your models in app(s): 'api' have changes that are not yet reflected in a migration, and so won't be applied.
  Run 'manage.py makemigrations' to make new migrations, and then re-run 'manage.py migrate' to apply them.
PerceptiLabs: Starting
Performing system checks...

System check identified no issues (0 silenced).
December 19, 2021 - 21:51:50
Django version 3.2, using settings 'static_file_server.settings'
Starting development server at http://127.0.0.1:8080/
Quit the server with CONTROL-C.
2021-12-19 13:51:50,651 - INFO - perceptilabs.utils:4 - Initialized sentry for environment 'prod' and release '0.13.5'
Performing system checks...

System check identified no issues (0 silenced).
December 19, 2021 - 21:51:51
Django version 3.2, using settings 'rygg.settings'
Starting development server at http://127.0.0.1:8000/
Quit the server with CONTROL-C.
[19/Dec/2021 21:51:52] "GET /?token=83rAJF4MKzmqhCRnlbAK33i8qp7HcIkefh532Ww_DMw HTTP/1.1" 200 483
[19/Dec/2021 21:51:52] "GET /static/styles/vendors.0ba335490269725c4fb6.css HTTP/1.1" 304 0
[19/Dec/2021 21:51:52] "GET /static/js/0.0b51ad40931d9a95af76.js HTTP/1.1" 304 0
[19/Dec/2021 21:51:52] "GET /static/js/app.c6237d467944e052fe7d.js HTTP/1.1" 304 0
[19/Dec/2021 21:51:52] "GET /rygg_url HTTP/1.1" 200 0
[19/Dec/2021 21:51:52] "GET /is_enterprise HTTP/1.1" 200 23
[19/Dec/2021 21:51:52] "GET /kernel_url HTTP/1.1" 200 0
[19/Dec/2021 21:51:52] "GET /is_enterprise HTTP/1.1" 200 23
[19/Dec/2021 21:51:52] "GET /rendering_kernel_url HTTP/1.1" 200 0
[19/Dec/2021 21:51:52] "GET /keycloak_url HTTP/1.1" 200 0
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
[19/Dec/2021 21:51:53] "GET /static/img/add-button.svg HTTP/1.1" 304 0
[19/Dec/2021 21:51:53] "GET /static/fonts/nunito-sans-v5-latin-700.d0b6d0f.woff2 HTTP/1.1" 304 0
[19/Dec/2021 21:51:53] "GET /rygg_url HTTP/1.1" 200 0
[19/Dec/2021 21:51:53] "GET /static/fonts/roboto-regular-webfont.736b705.woff2 HTTP/1.1" 304 0
[19/Dec/2021 21:51:53] "GET /static/fonts/icomoon.1b9c004.woff HTTP/1.1" 304 0
[19/Dec/2021 21:51:53] "GET /static/fonts/roboto-bold-webfont.ffa77d5.woff2 HTTP/1.1" 304 0
[19/Dec/2021 21:51:53] "GET /static/fonts/nunito-sans-v5-latin-regular.9c6a1ff.woff2 HTTP/1.1" 304 0
[19/Dec/2021 21:51:53] "OPTIONS /app/updates_available/?token=83rAJF4MKzmqhCRnlbAK33i8qp7HcIkefh532Ww_DMw HTTP/1.1" 200 0
[19/Dec/2021 21:51:53] "OPTIONS /projects/default/?token=83rAJF4MKzmqhCRnlbAK33i8qp7HcIkefh532Ww_DMw HTTP/1.1" 200 0
[19/Dec/2021 21:51:53] "OPTIONS /app/version/?token=83rAJF4MKzmqhCRnlbAK33i8qp7HcIkefh532Ww_DMw HTTP/1.1" 200 0
[19/Dec/2021 21:51:53] "OPTIONS /app/is_enterprise/?token=83rAJF4MKzmqhCRnlbAK33i8qp7HcIkefh532Ww_DMw HTTP/1.1" 200 0
[19/Dec/2021 21:51:53] "GET /app/version/?token=83rAJF4MKzmqhCRnlbAK33i8qp7HcIkefh532Ww_DMw HTTP/1.1" 200 20
[19/Dec/2021 21:51:53] "GET /app/is_enterprise/?token=83rAJF4MKzmqhCRnlbAK33i8qp7HcIkefh532Ww_DMw HTTP/1.1" 200 23
[19/Dec/2021 21:51:53] "GET /projects/default/?token=83rAJF4MKzmqhCRnlbAK33i8qp7HcIkefh532Ww_DMw HTTP/1.1" 200 223
[19/Dec/2021 21:51:53] "OPTIONS /datasets/?filename=undefined&project_id=1&token=83rAJF4MKzmqhCRnlbAK33i8qp7HcIkefh532Ww_DMw HTTP/1.1" 200 0
[19/Dec/2021 21:51:53] "GET /datasets/?filename=undefined&project_id=1&token=83rAJF4MKzmqhCRnlbAK33i8qp7HcIkefh532Ww_DMw HTTP/1.1" 200 1760
[19/Dec/2021 21:51:53] "GET /datasets/?filename=undefined&project_id=1&token=83rAJF4MKzmqhCRnlbAK33i8qp7HcIkefh532Ww_DMw HTTP/1.1" 200 1760
[19/Dec/2021 21:51:53] "POST /mixpanel/track/?ip=1&_=1639950713296 HTTP/1.1" 200 0
[19/Dec/2021 21:51:53] "GET /app/updates_available/?token=83rAJF4MKzmqhCRnlbAK33i8qp7HcIkefh532Ww_DMw HTTP/1.1" 200 21
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
[19/Dec/2021 21:51:58] "POST /mixpanel/engage/?ip=1&_=1639950718269 HTTP/1.1" 200 0
[19/Dec/2021 21:51:58] "POST /mixpanel/engage/?ip=1&_=1639950718271 HTTP/1.1" 200 0
[19/Dec/2021 21:51:58] "GET /mixpanel/decide/?verbose=1&version=3&lib=web&token=1480b2244fdd4d821227a29e2637f922&distinct_id=modeler%40perceptilabs.test&ip=1&_=1639950718266 HTTP/1.1" 200 0
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001

I can’t tell if the repeating sequence at the end ever stops.

I’m primarily interested in using ML for trade systems, but I’m also wanting to get an education in data science and seek out ways that data-driven methods can produce greater societal impact. The idea of building models of complex systems and applying scientific investigation to those models, is something that’s very interesting to me, and I kind of see endless applications for it.

Interesting, seems like the backside Kernel is allowed to access it?
Check your firewall maybe?
Otherwise check if something else is using port 5001
Both are possible

1 Like

You’re welcome & you are welcome :wink:

Thanks for providing the detail; @bamgm14 has some good suggestions to check - I’ve got nothing to add really except to ask whether you have an browser security extensions that maybe, might, have an effect… if so, disable them safely and see if it makes any difference…

Failing that, one of the PL guys will surely get you sorted shortly!

Hmmm… yet another person interested in ML for trading (that makes four of us that I know of).

FWIW my personal view is that step 1 is to produce naive forecasts, then to suck in assorted other Stuff to taste and then build a proper Reinforcement Learning system. I have started re-implementing in Python (with numpy, numba, etc.) a chaos detector I wrote a couple of years back, in Mathematica originally. The idea is to provide an extra category for any ML prediction system and a tweak (somehow, TBD) to say effectively “don’t try to learn when it’s chaotic”… took me a looong time to get the 1st version running so don’t hold you breath for news but, count me as an interested party generally! :smiley:

Hi @rmrfetc and welcome to the forum! :slight_smile:

The log looks fine for me as well so I’m just going to highlight @bamgm14’s great suggestion to check if something else already is using port 5001, which you can do by running:
netstat -a -n -o | find "5001"

Looking forward to hear more about your modeling as soon as you get up and running as well! :smiley:

PS
I am one of those people
XD

I don’t see any hits for port 5001. Since I’m on macOS, I used these commands:

@rmrfetc strange, sometimes it just takes a while for the rendering-kernel to start up though. How long have you waited for it without results?

I think this is similar issue(also waited for several hours)

My system specs:
windows 10 (20H2) wsl2 Ubuntu 18.04.2 LTS(also tried Ubuntu 20.04 LTS)
Python 3.8.12 (also tried 3.7)
PerceptiLabs 0.13.7

windows 10 (20H2) wsl2 Ubuntu 18.04.2 LTS

PS C:\Windows\system32> wsl --shutdown
PS C:\Windows\system32> netstat -ano|findstr 8080
PS C:\Windows\system32> netstat -ano|findstr 8000
PS C:\Windows\system32> netstat -ano|findstr 8001
PS C:\Windows\system32> netstat -ano|findstr 5000
PS C:\Windows\system32> netstat -ano|findstr 5001
PS C:\Windows\system32> wsl
To run a command as administrator (user "root"), use "sudo <command>".
See "man sudo_root" for details.

(base) dasomoh@DESKTOP-DQK40K5:/mnt/c/Windows/system32$ netstat
Active Internet connections (w/o servers)
Proto Recv-Q Send-Q Local Address           Foreign Address         State
Active UNIX domain sockets (w/o servers)
Proto RefCnt Flags       Type       State         I-Node   Path
(base) dasomoh@DESKTOP-DQK40K5:/mnt/c/Windows/system32$ netstat -ano|find 8000
find: ‘8000’: No such file or directory
(base) dasomoh@DESKTOP-DQK40K5:/mnt/c/Windows/system32$ netstat -ano|grep 8000
(base) dasomoh@DESKTOP-DQK40K5:/mnt/c/Windows/system32$ netstat -ano|grep 8001
(base) dasomoh@DESKTOP-DQK40K5:/mnt/c/Windows/system32$ netstat -ano|grep 8080
(base) dasomoh@DESKTOP-DQK40K5:/mnt/c/Windows/system32$ netstat -ano|grep 5000
(base) dasomoh@DESKTOP-DQK40K5:/mnt/c/Windows/system32$ netstat -ano|grep 5001
(base) dasomoh@DESKTOP-DQK40K5:/mnt/c/Windows/system32$ lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description:    Ubuntu 18.04.2 LTS
Release:        18.04
Codename:       bionic
(base) dasomoh@DESKTOP-DQK40K5:/mnt/c/Windows/system32$ conda activate perceptilabs
(perceptilabs) dasomoh@DESKTOP-DQK40K5:/mnt/c/Windows/system32$ python --version
Python 3.8.12
(perceptilabs) dasomoh@DESKTOP-DQK40K5:/mnt/c/Windows/system32$ pip list|grep perceptilabs
perceptilabs            0.13.7
(perceptilabs) dasomoh@DESKTOP-DQK40K5:/mnt/c/Windows/system32$ perceptilabs -v=3
Operations to perform:
  Apply all migrations: admin, api, auth, contenttypes, sessions
Running migrations:
  No migrations to apply.
  Your models in app(s): 'api' have changes that are not yet reflected in a migration, and so won't be applied.
  Run 'manage.py makemigrations' to make new migrations, and then re-run 'manage.py migrate' to apply them.
PerceptiLabs: Starting
Performing system checks...

System check identified no issues (0 silenced).
December 28, 2021 - 16:14:48
Django version 3.2, using settings 'static_file_server.settings'
Starting development server at http://127.0.0.1:8080/
Quit the server with CONTROL-C.
2021-12-29 01:14:49,774 - INFO - perceptilabs.utils:4 - Initialized sentry for environment 'prod' and release '0.13.7'
sh: 1: xdg-open: not found
Performing system checks...

PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs:    rygg on port 8000
System check identified no issues (0 silenced).
December 28, 2021 - 16:14:51
Django version 3.2, using settings 'rygg.settings'
Starting development server at http://127.0.0.1:8000/
Quit the server with CONTROL-C.
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001

windows 10 (20H2) wsl2 Ubuntu 20.04 LTS

PS C:\Windows\system32> wsl --shutdown
PS C:\Windows\system32> netstat -ano | findstr 5000
PS C:\Windows\system32> netstat -ano | findstr 5001
PS C:\Windows\system32> netstat -ano | findstr 8000
PS C:\Windows\system32> netstat -ano | findstr 8001
PS C:\Windows\system32> netstat -ano | findstr 8080
PS C:\Windows\system32> wsl
(base) dasomoh@DESKTOP-DQK40K5:/mnt/c/Windows/system32$ conda activate perceptilabs
(perceptilabs) dasomoh@DESKTOP-DQK40K5:/mnt/c/Windows/system32$ python --version
Python 3.8.12
(perceptilabs) dasomoh@DESKTOP-DQK40K5:/mnt/c/Windows/system32$ perceptilabs -v=3
Operations to perform:
  Apply all migrations: admin, api, auth, contenttypes, sessions
Running migrations:
  No migrations to apply.
  Your models in app(s): 'api' have changes that are not yet reflected in a migration, and so won't be applied.
  Run 'manage.py makemigrations' to make new migrations, and then re-run 'manage.py migrate' to apply them.
PerceptiLabs: Starting
Performing system checks...

System check identified no issues (0 silenced).
December 28, 2021 - 14:23:09
Django version 3.2, using settings 'static_file_server.settings'
Starting development server at http://127.0.0.1:8080/
Quit the server with CONTROL-C.
2021-12-28 23:23:10,109 - INFO - perceptilabs.utils:4 - Initialized sentry for environment 'prod' and release '0.13.7'
sh: 1: xdg-open: not found
Performing system checks...

PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs:    rygg on port 8000
System check identified no issues (0 silenced).
December 28, 2021 - 14:23:11
Django version 3.2, using settings 'rygg.settings'
Starting development server at http://127.0.0.1:8000/
Quit the server with CONTROL-C.
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001
PerceptiLabs: Waiting for services to listen on these ports:
PerceptiLabs:    rendering-kernel on port 5001

Hi @dasom,
We have had some issues with running on WSL, here’s a quick FAQ for it which may help: https://docs.perceptilabs.com/perceptilabs/support/perceptilab-faqs#why-does-perceptilabs-fail-to-run-on-wsl
Let me know how it goes :slight_smile:

@robertl I tried waiting 10 minutes before I gave up. In order to see that PL is still printing the waiting message, I have to clear the terminal every time it reaches page-length, and I was only willing to do that for about 10 minutes before giving up.

Hi @rmrfetc,
I just now realized that you were using an M1 machine. Unfortunately a bunch of packages (TensorFlow included) are still trying to catch up to the new hardware and we are seeing a variety of issues related to it. This is likely the case why things are not running properly for you.
If you have another machine with different hardware then it would be worth trying on that instead.

Worth an M1 sticky message somewhere?

It would be, we’ll add it to our FAQ :slight_smile: