Winners Annoucement

Round 2 of the Food Recognition Challenge is up!

A BIG THANKS to all the participants who made this round successful.

It is really impressive to see the scores improve by such a great margin from Round-1 and also the increased participation in Round-2.

Without further ado, it’s time to announce the winners of this round.

Congratulations to team rssfete comprising @rohitmidha23 and @shraddhaamohan who achieved the winning score of 63.4 (Average Precision). Together they bag the Singularity prize and the travel grant of 2500 CHF to visit AMLD(Applied Machine Learning Days). We also want to thank them for the resources notebooks that they made available publically for other participants as a useful reference.

We also want to congratulate @simon_mezgec for his continued participation and an impressive score of 59.2 (Average Precision) and extend to him a full travel grant of 2500 CHF to visit AMLD 2021 as well. Seeing such perseverance in participants is really amazing and motivates us as we work hard towards organizing more such challenges.

To everyone who participated, we thank you once again and encourage you to discuss your approach (successful or not) in the discourse forums.,Such open discussions help encourage everyone to participate and learn together as a community .

Stay tuned for some exciting announcements coming up shortly

View post 4

Train using mmdetection and submit via Colab (Round 2)!

View post 0

Submissions not finishing

@simon_mezgec
While debugging your submissions, we realised that you have changed nothing but only an addition of new model (epoch_1.pth -> model.pth), and its started taking 20 hours. Before the change it was just taking 3.5 hours.

We would like to know what is the difference between the baseline model (epoch_20.pth, epoch_1.pth) vs your model (model.pth) to find out why it’s taking more than 5x time.

@hannan4252 We are debugging your submissions and will let you know accordingly once we figure out the issue.

View post 0

Editing Docker file

@hannan4252
I assume this line is installing tensorflow 2.0:
https://gitlab.aicrowd.com/hannan4252/food_project/blob/Submission-v3.0.6/Dockerfile#L40
RUN pip3 install tensorflow-gpu as the version number hasn’t been mentioned for this package.

View post 0

Evaluation Criteria

We use the official cocoeval for evaluation, the code for it is available at: https://github.com/cocodataset/cocoapi/blob/master/PythonAPI/pycocotools/cocoeval.py
There you can search for ‘score’ key and check how it is being used for evaluation purposes,
If you observe though scores are not used in the calculation formulae but it’s used for the purpose of sorting.

View post 2

Evaluation Criteria

@gloria_macia_munoz the output format (annotations) and details regarding that are available at https://github.com/AIcrowd/food-recognition-challenge-starter-kit/blob/master/Dataset%20Utils.ipynb

View post 0

Evaluation Criteria

@gloria_macia_munoz as far as the implementation the background class is also considered for calculation of IoU.

https://www.jeremyjordan.me/evaluating-image-segmentation-models/, https://towardsdatascience.com/metrics-to-evaluate-your-semantic-segmentation-model-6bcb99639aa2

View post 0

Which libraries installs are **actually** needed on AIcrowd?

The requirements from this repository would be sufficient:

Nikhil Rayaprolu

View post 1

Instructions, EDA and baseline for Food Recognition Challenge

https://github.com/YBIGTA/pytorch-hair-segmentation/issues/37 this could be an issue, if you hadn’t used git lfs to pull the models.

View post 0

Instructions, EDA and baseline for Food Recognition Challenge

We are releasing a notebook with data analysis on the Food Recognition Dataset and then a short tutorial on training with keras and pytorch. This lets you immediately jump onto the challenge and solve the challenge.
https://colab.research.google.com/drive/1A5p9GX5X3n6OMtLjfhnH6Oeq13tWNtFO#scrollTo=ok54AWT_VoWV

Along with the notebook, we are also releasing the starter codes in both keras (using matterport maskrcnn) and pytorch (using mmdetection). Also, these starter codes have the submission format required to make a successful submission to AICrowd.

mmdetection (pytorch): https://gitlab.aicrowd.com/nikhil_rayaprolu/food-pytorch-baseline
matterport-maskrcnn (keras - tensorflow) - https://gitlab.aicrowd.com/nikhil_rayaprolu/food-recognition

View post 1

Extension of Deadline till December 31st

You could access the notebook and the baseline repositories at:
https://colab.research.google.com/drive/1A5p9GX5X3n6OMtLjfhnH6Oeq13tWNtFO

View post 0

Extension of Deadline till December 31st

We are extending the deadline to December 31st. We are also preparing a blog post and a baseline model to help participants with the challenge.

View post 2

Issue with aicrowd_helpers.py

@shivam @mohanty can you look through this?

View post 0

Issue with aicrowd_helpers.py

In the meantime, we will modify the helpers to handle the local evaluation too.

View post 0

Issue with aicrowd_helpers.py

For now, The local debug.sh needs you to remove lines relevant to aicrowd_helpers from run.py, as it requires redis.

View post 0

Issues with submitting

here is the initial version of the baseline submission, please go through README on changes required for submission, and raise your further queries here.

View post 0

I don't know how to submit

I am part of the organising team,
here is the initial version of the baseline submission, please go through README on changes required for submission, and raise your further queries here.

View post 0

I don't know how to submit

Also, we will be providing an example in the next 24 hours that smoothens your submission process. This example contains a baseline model and the structure of the repository for successful submission.

View post 0

Issues with submitting

hello, shraddhaamohan, we will be providing an example in the next 24 hours that smoothens your submission process. This example contains a baseline model and the structure of the repository for successful submission.

View post 0