Loading
Feedback

liberifatali 222

Location

VN

Badges

0
1
2

Activity

Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Mon
Wed
Fri

Ratings Progression

Loading...

Challenge Categories

Loading...

Challenges Entered

Classify images of snake species from around the world

Latest submissions

See All
graded 10024
graded 6995
failed 6992

Predict Labor Class

Latest submissions

See All
graded 73700
graded 73698
graded 73692

Real Time Mask Detection

Latest submissions

See All
graded 73807
graded 73198
graded 72790

Predict Win Depth

Latest submissions

See All
graded 73622
graded 73523
graded 73475

Classify Scrambled Text

Latest submissions

See All
graded 71529

Recognize Orientation of a 3D Skeleton

Latest submissions

See All
graded 74031
graded 73938
graded 73831
Gold 0
Silver 1
Boltzmann's Favourite
May 16, 2020
Bronze 2
Trustable
May 16, 2020
Newtonian
May 16, 2020

Badges


  • May 16, 2020

  • May 16, 2020

  • May 16, 2020

  • May 16, 2020
  • Has filled their profile page
    May 16, 2020

  • May 16, 2020

  • May 16, 2020

  • May 16, 2020

  • May 16, 2020
  • Kudos! You've won a bronze badge in this challenge. Keep up the great work!
    Challenge: Snake Species Identification Challenge
    May 16, 2020
  • Kudos! You've been awarded a silver badge for this challenge. Keep up the great work!
    Challenge: Snake Species Identification Challenge
    May 16, 2020
Participant Rating
alchemi01 240
Participant Rating

Hello guys

18 days ago

Ok, I uploaded Maskd and Scrbl.

About the prize, how can we claim?

Hello guys

18 days ago

This is my github username: https://github.com/LiberiFatali.
Could you add me so I can push?

Hello guys

18 days ago

Congratulation on 1st place :smiley:
I will arrange source codes of my parts for github.

Hello guys

19 days ago

great job ! :partying_face:

Hello guys

19 days ago

Yeah. I have sent to you via email.

Hello guys

19 days ago

The images I sent is the result of the pre-processing script.
Good job on 4.167 :slight_smile:

Hello guys

19 days ago

Besides horizontal flipping, I have no idea of other augmentation that can also give xRot label.

Hello guys

19 days ago

Flipped images (_flip.png) and corresponding xRot (in label.csv) are included.
There’s no preprocess script. I did preprocessing and saved to _flip.png, also append label in label.csv

Hello guys

19 days ago

Images are too big for github. So I shared google drive and send code by email.
Flipped images (_flip.png) and corresponding xRot (in label.csv) are include

  • Let me know if you want RGB cropped image. I think black and white image would be better, since they do not have color distraction when compute similarity.

  • There’s the idea to train CNN classifier with 360 class (degrees) on cropped black and white skeleton (something similar to MNIST classifier). I haven’t had time to try it out.

Hello guys

20 days ago

Well it’s night here. However, I sense your passion to win, so I will upload the cropped image and code for Skelly. Wait a moment !

What’s your github account?

Hello guys

20 days ago

My email is clessvna@gmail.com. I haven’t had Telegram yet.

Hello guys

20 days ago

We’re in safe position now, with 1st place at 4/5 challenges.

I brief ideas about my approaches here. Can you share ideas for Labor and Chess?

About newest improvement on Skelly, my approach is multi-steps as below:

  1. For both train and test images:
  • Binarize the full image by thresholding
  • Do dilation so skeleton parts are connected
  • Segment whole skeleton out of the full image
  • Save black and white image of skeleton area to new file
  1. For train images only:
  • Augument data by flipping horizontally
  1. For test images only:
  • Apply nearest neighbor classifer where each test sample is compared with all train database to get the nearest neightbor. Then the rotation of that test sample is the rotation of found neighbor.

About MaskD, I utilize a pretrained model with some tweaks in prediction script.

About SCRBL, I apply transformer model and also reach perfect score.

Codes and scripts are put in many files. I haven’t gather them so it’s not convenient to send now. After this, I will try to arrange time to clean source codes; then we can publish.

Hello guys

20 days ago

I think we can open source after the challenge ends. Do you guys using any chat app?
We also have better position on Skelly.

Hello guys

20 days ago

I just join BlitzCA. Nice to meet you !

Btw, good job 1st place on Labor and Chess :clap:

About team

20 days ago

(topic withdrawn by author, will be automatically deleted in 24 hours unless flagged)

Interview

About 1 year ago

Yeah. It sounds good.

Cheers,

Interview

About 1 year ago

Dear gloria_macia_munoz,

Thank you for the invitation. Does the interview include my video stream or just voice?
I will be more comfortable by answering questions via text.
Also, I’m on GMT+7 timezone.

My email address is hieunp131@gmail.com

Thanks,

MASKD

MMdetection unable to form final test file

20 days ago

Looking at the starter/baseline code. There’s the example for preparing the submission.

CHESS

About the dataset

23 days ago

Thanks for the reply.

I make another submission without external data. However, the leaderboard doesn’t seem to be updated:
https://www.aicrowd.com/challenges/aicrowd-blitz-2/problems/chess/submissions/73348

Is it ok to make submission now? @ashivani

About the dataset

23 days ago

While trying to find approaches for this problem, I encountered a dataset online.

After training the model on this extra dataset, I make the submission; the MAE is 0.0 and F! is 1.0 as in here:
https://www.aicrowd.com/challenges/aicrowd-blitz-2/problems/chess/submissions/73209
https://www.aicrowd.com/challenges/aicrowd-blitz-2/problems/chess/submissions/73222

Does the rule allow this? @mohanty
If this is the case that the rule does not allow, could you help to remove my above 2 submissions?

Furthermore, I think we should make clear about this public dataset. So all participants can start with equal data.

Snake Species Identification Challenge

Evaluation stuck [Edit : Evaluation took a long time]

About 1 year ago

Mine is still “waiting_in_queue_for_evaluation” since 6 hours ago.

Evaluation stuck [Edit : Evaluation took a long time]

About 1 year ago

Is the evaluation still running? My submission hasn’t been evaluated.

Cannot upload trained model to gitlab.aicrowd

About 1 year ago

I used git-lfs.
Now how to debug if anything goes wrong when submitting?

Cannot upload trained model to gitlab.aicrowd

About 1 year ago

It says that the size is too big. What should we do? @mohanty

liberifatali has not provided any information yet.