Location
Badges
Activity
Ratings Progression
Challenge Categories
Challenges Entered
Airborne Object Tracking Challenge
Latest submissions
Machine Learning for detection of early onset of Alzheimers
Latest submissions
See Allgraded | 134413 | ||
graded | 134407 | ||
graded | 133845 |
5 Puzzles 21 Days. Can you solve it all?
Latest submissions
See Allfailed | 169961 | ||
failed | 169910 | ||
failed | 169891 |
Latest submissions
See Allgraded | 127115 | ||
graded | 127112 | ||
graded | 127051 |
5 Problems 21 Days. Can you solve it all?
Latest submissions
See Allgraded | 124129 | ||
graded | 123956 | ||
graded | 123942 |
5 Puzzles 21 Days. Can you solve it all?
Latest submissions
See Allgraded | 126941 | ||
failed | 126940 | ||
graded | 126932 |
Latest submissions
See Allgraded | 121542 | ||
graded | 119205 | ||
graded | 119101 |
Participant | Rating |
---|---|
nima | 64 |
Participant | Rating |
---|
Multi-Agent Behavior: Representation, Model-17508f
AI Blitz #6
#1 Solution to Blitz6 ChessWin prediction
Over 3 years agoOh, that makes sense. I wasnβt aware humans still play this game.
#1 Solution to Blitz6 ChessWin prediction
Over 3 years agoThanks to organizers for this fun competition!
The key to solve all this puzzles was βChess Configurationβ puzzle, which asked to recognize FEN from images. I tried 2 approaches, but both had trouble with black pawns on dark background, so they gave .001 score, just below most competitors. Luckily, I got a teammate.
Chess Win Prediction
For this one I installed stockfish and called it from python API, as below. It disagreed with provided labels 7% of time, so probably labels were incorrect.
engine = chess.engine.SimpleEngine.popen_uci("stockfish")
limit = chess.engine.Limit(time=.2)
def predict_win(fen, turn):
"""
fen: fen string
turn: 'b' or 'w'
"""
ext_fen = f'{fen} {turn} - - 0 1'
board = chess.Board(ext_fen)
result = engine.analyse(board, limit)
score = result['score'].white() # from white perspective
if score.is_mate():
value = score.mate()
else:
value = score.score()
winner = 'white' if value > 0 else 'black'
return winner
Hockey: Player localization
HAC Software did not pay #3 prize; leaked data; solution
Over 3 years agoNo prize My solution was disqualified from the 3rd prize, because it made predictions on the subset of videos, not all of them. I disagree with this decision from the org.
Leak Sample submission had a leak: it was constructed as:
SampleSubmission = GroundTruth + Delta
Just submitting sample_submission.csv file gave a high score 130089.426 and some participants have figured it out. It is possible that Delta was not random, may be even constant. So fitting this Delta would provide a pretty high LB score. I didnβt try it, but it is possible that some participants did.
Solution The core of my approach was to find the position, angle and focus( or zoom factor) of the camera in the physical space. I used the following projections:
(3d physical space) <-> (camera lens) <-> (2d rink plane)
(ImageCoords) <-> (camera lens)
The approach was to figure out angles between individual frames in the video and collate frames to build a panoramic view of the rink. Then I could project each image to the panoramic view on camera lens and then onto 2d rink space. However, I noticed that some videos had a fixed camera, that didnβt move, rotate or zoom, so I decided to drop all this complexity and final submission was pretty simple: just find that fixed camera coordinates and angle and use them to project. This only worked for 3 videos, but was enough for #3 place.
Learning to Smell
#3 solution to Learning to Smell
Over 3 years agothanks. yeah, that step gave a small boost.
I had 2 ways to get validation scores:
- out of fold score for individual models: around .39
- average prediction of fold models on the held out set: around .41
LB: around .31
Itβs possible that the distribution of the test smells was different from what we had in the train set or molecules were structurally different, hence the discrepancy between LB and validation. In previous rounds results were much closer.
#3 solution to Learning to Smell
Over 3 years agoThis was one of my favorite challenges so far, because the problem formulation is very simple and it attempts to get insight into one of our primal but neglected basic senses. My solution was far behind top 2 competitors, so I feel like I was missing some crucial ingredient, so I am looking forward to learn about their approach.
The core of my approach is neural net on fingerprints.
-
Data: union of various fingerprints extracted with
rdkit
from theSMILES
in train setfrom rdkit import Chem from rdkit.Chem import AllChem from rdkit.Chem import MACCSkeys mol = Chem.MolFromSmiles(smiles) fp0 = MACCSkeys.GenMACCSKeys(mol) # MACCS keys fp1 = AllChem.GetMorganFingerprintAsBitVect(mol, 2, 256) # Morgan fingerprints fp2 = Chem.RDKFingerprint(mol) fp3 = [len(mol.GetSubstructMatch(Chem.MolFromSmarts(smarts)) > 0 for smarts in smarts_inteligands] # smarts_inteligands has about 305 smarts patterns
-
Preprocessing: drop constant and duplicate fingerprints
-
Model:
from torch import nn hidden_size = 512 dropout = .3 output_size = 75 nn.Sequential( nn.Linear(input_size, hidden_size), nn.ReLU(inplace=True), nn.Dropout(dropout), nn.BatchNorm1d(hidden_size), nn.Linear(hidden_size, hidden_size), nn.ReLU(inplace=True), nn.Dropout(dropout), nn.BatchNorm1d(hidden_size), nn.Linear(hidden_size, output_size), )
-
Training was done over 5 folds, each one for 25 epochs with
nn.BCEWithLogitsLoss
. The model tried to predict probabilities of 75 smells. -
The last step was to come up with 5 prediction sequences starting from individual smell probabilities. For this I sampled smells using their predicted probabilities and found the sequence with the best
jaccard
score. Then found the next sequence with the best incrementaljaccard
score and so on. -
Bells and whistles. Some of the things that made small improvements:
- label smoothing
- weighting labels for training
- weighting fingerprints based on their estimate importance
- Things that didnβt work:
- PCA on features and on labels
- UMAP on features and on labels
- pretraining on 109 labels
- continous version of IOU loss instead of BCE for training
- various learning rate schedulers
- dropping fingerprints with high correlation to others
- trying another dropout/learning rate
Launching the 3rd and Final Round of Learning to Smell Challenge π
Over 3 years agoThey claim 97% accuracy. Does this mean that the problem is solved?
Requesting early feedback for Round 2
Almost 4 years agoany progress on this? it seems your evaluator doesnβt find vocabulary
Requesting early feedback for Round 2
Almost 4 years agoI have a similar error:
βSubmission Vocabulary contains Unknown smell words : blackcurrant,dairy,seafoodβ
https://gitlab.aicrowd.com/plemian/learning_to_smell/issues/1
how to fix this?
Time Series Prediction
#1 Solution to TIMSER Blitz5
Over 3 years agoWell, when I submitted MSE=3333333.333 solution, I already knew the answer, so just added some noise to make it look cute.
But to your point, I looked at the distribution of fractional values and ran a couple of linear regressions. Fractional values in this data had some distinct pattern. Single stock prices usually get adjusted because of splits and dividends, so the distribution of their fractional values didnβt match the pattern. Indices on the other hand donβt get adjusted, so I searched for the combination of indices.
Overall, I think it was a great puzzle and a lot of fun to solve.
#1 Solution to TIMSER Blitz5
Over 3 years agoIf you download prices of this 2 indices and add their βOpenβ columns you will get the solution. The prices came from indices, not individual stocks
#1 Solution to TIMSER Blitz5
Over 3 years agoNow that the results are in, time to share the solution.
The data was generated with the following formula:
DowJonesIndex.Open + NasdaqIndex.Open
Data could be downloaded from the following links:
https://finance.yahoo.com/quote/^DJI/history?p=^DJI
https://finance.yahoo.com/quote/^IXIC/history?p=^IXIC
IMGCOL
#1 Solution IMGCOL Blitz5
Over 3 years agomy guess is because the colorizers library opens image using PIL, so it has different convention on channel ordering than cv2.
#1 Solution IMGCOL Blitz5
Over 3 years agoColorizers library worked well for this competition: https://github.com/richzhang/colorization.git
Below is the code:
import cv2
import glob
from colorizers import *
use_gpu = True
def color(colorizer, img_path):
# default size to process images is 256x256
# grab L channel in both original ("orig") and resized ("rs") resolutions
img = load_img(img_path)
(tens_l_orig, tens_l_rs) = preprocess_img(img, HW=(256,256))
if (use_gpu):
tens_l_rs = tens_l_rs.cuda()
# colorizer outputs 256x256 ab map
# resize and concatenate to original L channel
img_bw = postprocess_tens(tens_l_orig, torch.cat((0*tens_l_orig,0*tens_l_orig),dim=1))
out_img = postprocess_tens(tens_l_orig, colorizer(tens_l_rs).cpu())
return out_img
def main():
# load colorizers
#colorizer_eccv16 = eccv16(pretrained=True).eval()
colorizer_siggraph17 = siggraph17(pretrained=True).eval()
if use_gpu:
#colorizer_eccv16.cuda()
colorizer_siggraph17.cuda()
colorizer = colorizer_siggraph17
stage = 'train'
data_dir = '{INSERT_YOUR_DATA_DIR}/' + stage + '_black_white_images/' + stage + '_back_white_images/'
out_dir = '../' + stage + '_color_images/'
fnames = glob.glob(data_dir + '/*')
print(fnames[:10])
for cnt, fname in enumerate(fnames):
imagename = fname.split('/')[-1]
outname = out_dir + imagename
res = color(colorizer, fname)
res = res[:,:,[2,1,0]] # reorder channels
cv2.imwrite(outname, np.clip(res * 256, 0, 255).astype(int))
if cnt % 100 == 0:
print(cnt)
main()
AI Blitz 5 β‘
TXTOCR
#2 Solution TXTOCR Blitz5
Over 3 years agoTesseract.
The data was 1 or 2 english words written on the image. Tesseract generally did ok in reading the text from the image, except when it encountered fancy font or the word was at the edge of the image partly invisible.
-
Preprocessing
Binarization did help tesseract better handle images, it made the text written in black on white background:def binarize(fname): img = cv2.imread(fname) for d in range(3): img2 = img[:, :, d] med = np.median(img2) img[:, :, d] = abs(img2 - med) bw = np.sum(img, axis=2) bw = bw / np.max(bw) * 255 # scale bw = 255 - bw fname = fname.replace('.png', '_bin.png') cv2.imwrite(fname, bw)
Also tried to resize image from 256x256 to 512x512.
-
Run tesseract:
tesseract img_bin.png out --psm 7 -l eng
-
Vocabulary check:
Lastly, check if prediction is made of actual words, and if not - try different type of preprocessing.
import enchant
usdict = enchant.Dict(βen_USβ)
usdict.check(βwordβ)
Sound Sentiment Prediction
#1 Solution to SOUSEN Blitz5
Over 3 years agoThe trick to this competition was to identify negative reviews with high confidence. My approach had 2 easy steps:
- Convert speech to text.
- Classify text based on sentiment.
In both steps there was no training involved, only inference using pretrained models from internet.
-
Convert speech to text.
For this I used a free model found on torchhub. It wasnβt 100% accurate, but did ok.import pandas as pd import torchaudio import torch from glob import glob device = torch.device('cpu') # gpu also works model, decoder, utils = torch.hub.load(repo_or_dir='snakers4/silero-models', model='silero_stt', language='en', # also available 'de', 'es' device=device) (read_batch, split_into_batches, read_audio, prepare_model_input) = utils # see function signature for details stage = 'test' files = sorted(glob(f'{INSERT_YOUR_PATH_TO_WAV_FOLDER}/*.wav')) bsize = 10 batches = split_into_batches(files, batch_size=bsize) ids = [f.split('/')[-1].split('.')[0] for f in files] res = [] for i, batch in enumerate(batches): minput = prepare_model_input(read_batch(batch), device=device) output = model(minput) for j, example in enumerate(output): res.append([ids[i * bsize + j], decoder(example.cpu())]) df = pd.DataFrame(res) df.columns=['wav_id', 'text'] df.to_csv(f'text_{stage}.csv', index=False)
-
Compute sentiment
For this task I usedtransformers
library, which gives βPOSITIVE/NEGATIVEβ sentiment of a phrase and a confidence in its prediction.from transformers import pipeline
nlp = pipeline(βsentiment-analysisβ)
bulk = 50res = []
for i in range(math.ceil(len(z) / bulk)):
r = nlp(list(df.iloc[bulk * i: bulk * (i + 1)][βtextβ].values))
res.extend( r)rdf = pd.DataFrame(res).rename(columns={βlabelβ: βsentimentβ})
d = pd.concat([df, rdf], axis=1)
d = d.sort_values(βwav_idβ).reset_index(drop=True)
Finally, submit 0(=negative) only when the model is very confident:
d[βlabelβ] = 2
d.loc[(d.sentiment == βNEGATIVEβ) & (d.score > .995)] = 0
d.to_csv(βsubmission.csvβ, index=False)
OBJDE
#1 Solution OBJDE Blitz5
Over 3 years agoFor this competition the baseline https://www.aicrowd.com/showcase/baseline-for-objde-challenge actually worked pretty well. My guess is that the labels were generated from a similar model as in the baseline, so sticking to this architecture would give better results.
There were 2 minor tweaks to make it work:
-
Creating Data cell:
'category_id': i[1]['category_id'].values[0],
replace with
'category_id': i[1]['category_id'].values[n],
-
Submitting predictions cell:
new_boxes.append([b[0]*w, b[2]*h, b[1]*w, b[3]*h])
replace with
new_boxes.append([b[0]/w, b[2]/w, b[1]/h, b[3]/h])
I ran the training for 5000 steps:
Creating Model cell:
cfg.SOLVER.MAX_ITER = 5000
and then continued for another 10000 with smaller lambda:
cfg.SOLVER.BASE_LR = 0.00025 / 10
cfg.SOLVER.MAX_ITER = 10000
Insurance pricing game
OSError: libgomp.so.1: cannot open shared object file: No such file or directory
Almost 4 years agohow to install this library βlibgomp1β in the zip submission?
Flatland Challenge
Computation budget
Almost 5 years agoIs 8 hour limit enforced? My submission 7 took 42 hours: https://gitlab.aicrowd.com/plemian/flatland-challenge/issues/7
Do I understand correctly, that we roughly have 1minute( = 8h/200 ) per test?
Claim your AWS Credits | Report your improvement in baseline score
Over 3 years agoclaiming credits for Classification task submission_id=127051 F1score=.814