GithubHelp home page GithubHelp logo

Comments (7)

ArlindKadra avatar ArlindKadra commented on June 16, 2024

Hey :),

Thanks for finding our work interesting. The first thing that comes into my mind, is that maybe the arguments might not be set ideally. For example, with a large dataset, func_eval_time needs to cover the time for a full function evaluation.

parser.add_argument(
'--func_eval_time',
type=int,
default=1000,
)

I would suggest running for one epoch initially and not for 105, to see if the problem persists and then to adjust accordingly.

parser.add_argument(
'--epochs',
type=int,
default=105,
)

from welltunedsimplenets.

ArlindKadra avatar ArlindKadra commented on June 16, 2024

I will consider the issue resolved, if it still does not work, you can reopen it and we can discuss it further :).

from welltunedsimplenets.

DonIvanCorleone avatar DonIvanCorleone commented on June 16, 2024

Sorry for the delayed response.

I tested you proposal with a highly reduced dataset = 7500 samples and a much longer func_eval_time = 86400 / epoch = 1. This setup achieved a result within ~ 45 minutes.
The moment I increased the dataset to 15000 samples with same func_eval_time and epoch it fails again.
By the way: I increased the number of workers to 8.

Any further suggestions? Any help highly appreciated :)

from welltunedsimplenets.

DonIvanCorleone avatar DonIvanCorleone commented on June 16, 2024

HI @ArlindKadra, by any chance some additional ideas?

from welltunedsimplenets.

ArlindKadra avatar ArlindKadra commented on June 16, 2024

Hey @DonIvanCorleone,

We do have an experimental branch that you could try out, it is reg_cocktails-pytorch_embedding. Theoretically, it should perform better in your case if there are multiple features and possibly some of them are categorical with a high dimensionality.

from welltunedsimplenets.

DonIvanCorleone avatar DonIvanCorleone commented on June 16, 2024

Hey @ArlindKadra,

thanks for your support. I tried yesterday and today to get your new approach running but at the moment something broke "entirely".

I am trying to dockerize the entire project and integrated all the required steps from this repository and the image seems to be properly setup but unfortunately in the very end I always receive the following new error after starting the process:

[ERROR] [2022-10-28 10:30:17,276:Client-AutoPyTorch:734d2973-56ab-11ed-815f-0242ac110002:1] get_smac_object() got an unexpected keyword argument 'initial_configurations'

Its unclear to me, whats happening and more specifically Why? :)
It seems to me, that the setup of autoPyTorch and this repository is more complicated than expected and I don`t want to steal more support time from you. If you ever find the time to create a wokring dockerized version of your solution it would be awesome, but at the moment I don't see whats not working properly.

Many thanks again
Cheers

from welltunedsimplenets.

DonIvanCorleone avatar DonIvanCorleone commented on June 16, 2024

For the sake of completion these are the installed modules

Package Version


absl-py 1.3.0
autoPyTorch 0.2
cachetools 5.2.0
catboost 1.1
certifi 2022.9.24
charset-normalizer 2.1.1
click 8.1.3
cloudpickle 2.2.0
ConfigSpace 0.6.0
contourpy 1.0.5
cycler 0.11.0
Cython 0.29.32
dask 2022.10.0
distributed 2022.10.0
emcee 3.1.3
flaky 3.7.0
fonttools 4.38.0
fsspec 2022.10.0
google-auth 2.13.0
google-auth-oauthlib 0.4.6
graphviz 0.20.1
grpcio 1.50.0
HeapDict 1.0.1
idna 3.4
imageio 2.22.2
imgaug 0.4.0
importlib-metadata 5.0.0
Jinja2 3.1.2
joblib 1.2.0
kiwisolver 1.4.4
liac-arff 2.5.0
lightgbm 3.3.3
locket 1.0.0
lockfile 0.12.2
Markdown 3.4.1
MarkupSafe 2.1.1
matplotlib 3.6.1
minio 7.1.12
msgpack 1.0.4
networkx 2.8.7
numpy 1.23.4
oauthlib 3.2.2
opencv-python 4.6.0.66
openml 0.12.2
packaging 21.3
pandas 1.5.1
partd 1.3.0
Pillow 9.2.0
pip 22.2.2
plotly 5.11.0
protobuf 3.19.6
psutil 5.9.3
pyarrow 10.0.0
pyasn1 0.4.8
pyasn1-modules 0.2.8
pynisher 0.6.4
pyparsing 3.0.9
pyrfr 0.8.3
python-dateutil 2.8.2
pytz 2022.5
PyWavelets 1.4.1
PyYAML 6.0
regex 2022.9.13
requests 2.28.1
requests-oauthlib 1.3.1
rsa 4.9
scikit-image 0.19.3
scikit-learn 0.24.2
scipy 1.9.3
setuptools 63.4.1
Shapely 1.8.5.post1
six 1.16.0
smac 1.4.0
sortedcontainers 2.4.0
tabulate 0.9.0
tblib 1.7.0
tenacity 8.1.0
tensorboard 2.10.1
tensorboard-data-server 0.6.1
tensorboard-plugin-wit 1.8.1
threadpoolctl 3.1.0
tifffile 2022.10.10
toolz 0.12.0
torch 1.12.1
torchvision 0.13.1
tornado 6.1
typing_extensions 4.4.0
urllib3 1.26.12
Werkzeug 2.2.2
wheel 0.37.1
xmltodict 0.13.0
zict 2.2.0
zipp 3.10.0

from welltunedsimplenets.

Related Issues (7)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.