🚀 Code Submission Round Launched 🚀

Hi @dami,

We have re-run these submissions along with the fix for your issue.
It should pass automatically or will show relevant logs in case of failure now on.

View post 1

No debug logs was given in my fail submission

Hi @dami, I have shared the error and potential solution in your GitLab issue page given it includes information about your approach.

Let us know in case it helps.
You can reply back on the comment here or on the GitLab issue page.

View post 0

🚀 Code Submission Round Launched 🚀

Hi @wac81, the leaderboard contains the best submission from each participants (not all the submissions). I assume that may have caused the confusion.

You can see all your submission here (they are merged for all the tasks in this challenge, but you can identify them using relevant repository name):
https://www.aicrowd.com/challenges/esci-challenge-for-improving-product-search/submissions?my_submissions=true

Best,
Shivam

View post 0

🚀 Code Submission Round Launched 🚀

Hi @vitor_amancio_jerony, thanks for the logs.

We have identified the bug during the evaluation phase which caused the models to load twice, and is now fixed.
We have also restarted your latest submission and monitoring if any similar error happens to it.

View post 1

🚀 Code Submission Round Launched 🚀

Hi all, in case you feel your submission is running quite slow online v/s your local setup.
It might be a good idea to verify torch or relevant packages are installed properly.

Here is an example for torch:

In case you are confused how to verify for your package, please let us know and we can release relevant FAQs.

View post 0

🚀 Code Submission Round Launched 🚀

Hi @wac81, it has failed due to timeout.

View post 0

🚀 Code Submission Round Launched 🚀

Hi @LYZD-fintech,

All the successful submissions would be considered for the final ranking in this challenge.

Best,
Shivam

View post 0

🚀 Code Submission Round Launched 🚀

Hi @LYZD-fintech,

Each submission has dedicated compute resources.

The time elapsed, that is reported on the issue page is currently wrong and will be fixed soon, it shows the total time from the submission to completion (instead of start of execution of your code to the end). The timeout however is properly implemented and only considers the running time.

We provision a new machine dynamically for each submission, due to which the time elapsed might have been higher when there are a high number of submissions in the evaluation queue (multiple machines got provisioned)

I hope that clears any confusion.

Best,
Shivam

View post 0

🚀 Code Submission Round Launched 🚀

Hi @wac81, your submissions have been evaluated.

View post 0

🚀 Code Submission Round Launched 🚀

Hi @vitor_amancio_jerony @qinpersevere & all,

The image building code has been updated on our side, and the repository size is no longer a restriction.

Please keep in mind that the resources available for inference are same for everyone i.e. Tesla V100-SXM2-16GB.

View post 2

🚀 Code Submission Round Launched 🚀

Hi @zhichao_feng, your submissions have failed due to 90 minutes timeout for each public & private phase runs. (it got 92% in 90m)

View post 0

🚀 Code Submission Round Launched 🚀

Hi @wac81,

The issue is resolved and I noticed you were able to make the submission just now.
Please let us know in case you face any other issue.

View post 0

🚀 Code Submission Round Launched 🚀

Hi @animath3099, the issue has been resolved. You can try to make a submission now.

Note: Please remove debug: true from your aicrowd.json, due to which no slots remaining error is coming.

View post 0

🚀 Code Submission Round Launched 🚀

Hi @animath3099, can you share the link to the issue page? We are looking into it.

View post 0

🚀 Datasets Released & Submissions Open 🚀

Hi @olivertautz, that’s not the case, and any team member can submit in the challenge (not only the team member).

Can you please help by sharing relevant URL (Issue page, submission ID, etc) and/or a screenshot?

View post 0

Fatal: pack exceeds maximum allowed size

Hi @pn1,

Please use git-lfs to upload larger files in your repository.

The relevant commands are available in starter kit, or you can refer this for example.

View post 0

🚀 Code Submission Round Launched 🚀

Hi @vitor_amancio_jerony,

Can you check if weights can be pruned in your approach?

We are checking internally to increase the allowed submission/model size meanwhile, and will update you soon.

View post 2

🚀 Datasets Released & Submissions Open 🚀

Hi @olivertautz,

You can find the participate button on the challenge page.
https://www.aicrowd.com/challenges/esci-challenge-for-improving-product-search

image1324×813 212 KB

View post 0

🚀 Code Submission Round Launched 🚀

Hi @qinpersevere,

We don’t want you to be restricted because of the submission limit.

Due to this, the submissions can have a maximum size of 10GB [to prevent abuse].
Along with it, the limit is only applied on the files present in your current tag (HEAD i.e. part of the submission). You only need to remove any unwanted [or old] models from your HEAD, without worrying about deleting them from git history, etc.

In case your usecase require more than 10GB, please let us know.

Best,
Shivam

cc: @xuange_cui, @TransiEnt, @ystkin, @masatoh

View post 2

How to install pytorch in your submissions? Can my pytorch use GPU?

Potential Issue:

In case the challenge you are participating has GPUs available during evaluation, and your submission is running slower than when you run it locally. One of the potential reason can be wrong torch version installed.

Why is this a common occurrence?

It is undoubtedly one of the most popular frameworks for the deep learning users, BUT there is a special caveat.
The PyPI releases for the PyTorch do not have GPU releases, and due to that it is easy to just add torch in your requirements.txt and end up using the CPU version of pytorch.

How to identify if you are using correct torch version?

In the build logs provided to you, please check the package downloaded.
For example below is a CPU version as it is missing cuXXX prefix:

Collecting torch
    Downloading torch-1.12.0-cp37-cp37m-manylinux1_x86_64.whl

The GPU version examples:

torch-1.10.2-cu111-cp39-cp39-win_amd64.whl
torch-1.8.0-cu111-cp36-cp36m-linux_x86_64.whl

Package

torch

Usage:

>>> import torch
# you can add this in your code & print in the logs of your code as well to verify that everything works
>>> torch.cuda.is_available() 
True

Solution

I am using requirements.txt. What should I do?

You can add extra-index-url before mentioning torch or at the top of your requirements.txt.

In case you want to point to a specific CUDA version and PyTorch version, the example is as follows (you can open the link to know if your pytorch version support your cuda version):

--extra-index-url https://download.pytorch.org/whl/cu111
torch==1.10.1+cu111

I am using environment.yml. What should I do?

The installation is a bit easier with Conda, and the sample file looks as follows. (note: you can mention python version of your choice, etc as well)

name: myenv
channels:
  - conda-forge
  - defaults
  - pytorch
dependencies:
  - python=3.8
  - cudatoolkit=10.2   # cuda version
  - pytorch=1.11       # torch version
  - torchvision
  - torchaudio
  - pip:
      - pandas
      - <....other pip dependencies goes here....>

View post 1