Questions tagged [kaggle]
Kaggle is an online community for data scientists and machine learning practitioners owned by Google.
119
questions
0
votes
2
answers
37
views
What methods do top Kagglers employ for score gain?
I’m currently taking a course on ML and part of my final grade is my position on a Kaggle competition (private one) regarding a classification task. The majority of groups tend to have a similar ...
0
votes
0
answers
13
views
Finding/analyzing data competitions that share the individual model's predicted outcomes?
I would like to analyze past machine-learning data competitions, focusing on bias and variance of predictions. I thought I could look at past kaggle competitions, but unfortunately kaggle typically ...
- 149
0
votes
0
answers
21
views
Pip download does not see package
In kaggle comp I try to store all the packages to build them later without internet connection.
At first I use freeze to get packages, then download. Problem is, that
despite 'freeze' show us /home/...
- 3
0
votes
0
answers
33
views
Segmentation-models-pytorch returns black image
I tried to use Segmentation-models-pytorch first time by following guideline and using it on my task on kaggle(HuBMAP), here is link for official guide:
https://github.com/qubvel/segmentation_models....
- 3
0
votes
0
answers
22
views
Missing weights file in kaggle notebook
I was trying to train the DistilBert model on AG news dataset in kaggle by running this code.
I created a version of the notebook and let it run in the background. After the training was done i was ...
- 1
1
vote
1
answer
85
views
Where do Kaggle Masters Work?
If you become a Kaggle Grand Master, it should increase your job prospects, but does it guarantee a job at a FANG company for instance is a question of interest.
1
vote
1
answer
272
views
Filling na values with condition from other column [closed]
I am working on famous titanic dataset and I want to replace na values in Age column but on such a condition that these people whose Pclass=3 receive 25, Pclass=2 29 and Pclass=1 38.
I was trying to ...
- 11
0
votes
0
answers
44
views
Train AI to recognize a Snapchat's bitmoji's gender
I have found this https://www.kaggle.com/competitions/bitmoji-faces-gender-recognition/overview/description competition. I need the file module that returns either male or female but cant find it. How ...
1
vote
1
answer
51
views
why do some people don't use scikit-learn library instead of writing long python functions?
The other day I was exploring Kaggle and I found most of them don't use scikit-learn or any other library but instead they write these long python scripts.
For ...
- 11
0
votes
1
answer
102
views
Where can I practice multivariate outlier detection?
Can anyone provide me with a dataset, hopefully on Kaggle, where I can practice my skills in outlier analysis?
I have been studying this topic for quite a while, but I can't find a case study to apply ...
- 121
0
votes
1
answer
53
views
Need help understanding how this Neural Network is working
This is a model I came across, and I need some help understanding how it works
It uses South German Credit Prediction data set from Kaggle
...
1
vote
1
answer
22
views
Changing the predicted variable from price to price/km due to better visual correlation
I'm working on a dataset of Uber Rides from Kaggle. Of the important variables there are pickup and drop-off coordinates, passenger count, datetime of pickup, distance and the final price. I'm ...
2
votes
2
answers
977
views
Massive difference in accuracy of KNN depending on random_state
pardon the noob question but I am baffled by the following behavior. My model has MASSIVELY different results based on the random seed. I want to train a KNN classifier on the famous Kaggle Titanic ...
- 123
11
votes
4
answers
2k
views
Why SMOTE is not used in prize-winning Kaggle solutions?
Synthetic Minority Over-sampling Technique SMOTE, is a well known method to tackle imbalanced datasets. There are many papers with a lot of citations out-there claiming that it is used to boost ...
- 6,082
1
vote
1
answer
116
views
Difference between model score on test part and Kaggle public score
I tested my CatBoostModel model on part of data and get 0.92 score, but Kaggle public score was 0.9. I found new hyperparameters via randomsearch, new model score was 0.925, but on Kaggle score fell ...
- 125
0
votes
1
answer
50
views
Why performance varies among validate set, public testset and private testset?
When practicing with classical kaggle competitions, such and Titanic, House pricing, and so on, I followed the traditional process that I learned from textbook:
split training data into trainig set ...
- 3
1
vote
1
answer
1k
views
Keras ImageDataGenerator unable to find images
I'm trying to add image data to a Kaggle notebook so I can run a convolutional neural network but I'm having trouble doing this via ImageDataGenerator. This is the ...
- 135
2
votes
1
answer
455
views
Loading medical imaging data from multiple folders
I have a fairly basic mathematical and implementational understanding of ML algorithms and CNNs, and I am trying to think of an approach for this task: https://www.kaggle.com/c/rsna-miccai-brain-tumor-...
- 103
1
vote
1
answer
99
views
Creating and training a Multilayer perceptron model with few data
Are there any ways to create a deep multilayer perceptron model that is capable of making accurate regression predictions based on the training done using around 1000 unique data?
I'm currently ...
- 13
3
votes
2
answers
349
views
How to pose a non-trivial ML task for students?
As I am working in teaching, I would like to provide my students a (basic) machine learning classification challenge in the next semester. I really like the idea of giving them a challenge on an ...
- 131
2
votes
0
answers
90
views
When to stop the final model training?
Let's say I'm participating in a Kaggle image recognition competition.
Firstly, I create a train/validation split and find the good hyperparameters for my model. Here the stopping criterion is when ...
- 21
3
votes
1
answer
137
views
Can data leakage be sometimes acceptable?
I have recently started using kaggle and I have stumbled on a few examples of practices I would consider do be data leakage. Many of them were done by people well established on the platform and I ...
- 115
0
votes
0
answers
497
views
Why am I getting good validation scores, but poor test scores in Kaggle competition
I am participating in a Kaggle multiclass classification competition. The submissions will be scored based on the 'logloss' score. I am using Keras and Scikit libraries and a deep learning network ...
- 101
0
votes
1
answer
2k
views
Good model but bad confusion matrix?
I am trying to understand the code here.
The output [12] shows that the model accuracy is above 90% even for the validation set, but the confusion matrix in [16] ca not even achieve 50% accuracy, and ...
- 41
1
vote
0
answers
35
views
About the Licence of Shared Codes on the Kaggle Site [closed]
I have one job offer and there are many codes on the Kaggle site based of that job offer which could be used for the company usage like the result of this google search, and Salomon Tetelepta codes:
...
- 111
1
vote
0
answers
128
views
Recommendation System with ALS Implicit
I created a model for Recommending top 10 items to users similar to the approach used here https://github.com/benfred/implicit/blob/master/examples/lastfm.py
I wanted to evaluate the model using NDCG ...
- 21
1
vote
0
answers
90
views
Predicted probabilities of Multi Label Classification
I'm currently working on a Kaggle Competition wich objective is to predict probabilities of an ID belonging to each class. There are 4 posible classes.
The data is tabular and because it's a Kaggle '...
- 11
-1
votes
1
answer
106
views
Lack of standardization in Kaggle jupyter notebooks when using lasso/ridge?
I've recently started using Kaggle, and I've noticed that for a lot of these jupyter notebooks written by others, when they use Ridge/Lasso, they don't standardize the non-categorical numerical ...
2
votes
1
answer
31
views
Image classification vs medical grading problems
For image classification problems like cat vs dogs, the output layer is 2.
Image classification problems like diabetic retinopathy seem to be more of a grading classifier. Although the targets range ...
- 23
0
votes
1
answer
38
views
Parse documents to obtain subjective sentiment
I'm working on a project which deals with MRC (Machine Reading Comprehension).
I would like a machine to read an article and give me the sentiments based on a provided token.
For Instance given the ...
- 101
0
votes
1
answer
92
views
Trying to run a kaggle notebook
I'm found an interesting problem on kaggle and more or less solved it with my limited knowledge of machine learning. I was curious how other people solved and checked the solution with the highest ...
0
votes
0
answers
25
views
Size drastically increased when images converted to hdf5 format
I converted Kaggle Dog vs Cat dataset from image to hdf5 format, in order to fasten the learning time. To my surprise, 540MB training data converted to more than 30GB of data in hdf5 file format. Why ...
- 679
0
votes
1
answer
20
views
Searchable list of Kaggle challenges
Is there a way to search Kaggle for a list of current and prior challenges, or an external site that does that?
When I search on Kaggle it will only bring up solution notebooks and datasets, it doesn'...
- 165
1
vote
1
answer
112
views
Dropping missing rows in two dataframes
I have two files :
Test_data - contains the features of a dataset to find predictions for
Submission_data - contains two columns : The index column for test data and another column for its ...
- 75
0
votes
0
answers
24
views
Is there a certain threshold over which to accept or reject predictors based on correlation values with the target variable?
I have been working on the Titanic dataset. After some feature manipulation, I printed out the correlation values between my target variable Survived and all the ...
- 1
2
votes
1
answer
373
views
How do I read the cord_19_embeddings_2020-07-16.csv from the COVID-19 Open Research Dataset Challenge (CORD-19) on Kaggle?
What I'm trying to do
I wanted to use the CORD19 word embeddings csv to map them to certain findings from the rest of the dataset, but as we can see there are no stings in the first column.
The way I ...
- 151
7
votes
3
answers
11k
views
Kaggle notebook Vs Google Colab
What are the major differences between Kaggle notebook and Google Colab notebook?
To work on a dataset my first step is to start a Kaggle notebook but then I cant help thinking what could be the ...
- 71
1
vote
2
answers
277
views
How to find correlation between categorical data and continuous data
I'm working on imputing null values in the Titanic dataset.
The 'Embarked' column has some. I do NOT want to just set them all to the most common value, ...
- 13
0
votes
1
answer
61
views
Dropping attributes leads to better classifier accuracy? (Titanic Set)
I am currently tackling the Titanic Dataset on Kaggle. The goal is to find a classifier that can predict whether a passenger will survive or die.
The dataset has features that I believe are strongly ...
- 1
2
votes
2
answers
326
views
How to divide a dataset for training and testing when the features and targets are in two different files?
I am trying to divide a dataset into training dataset and testing dataset for multi-label classification. The datset I am working on is this one. It is divided into a file which contains the features ...
0
votes
0
answers
108
views
Using word embeddings for kaggle?
Not sure, if this is the right forum so redirect me if it wrong.
I have started on an NLP problem in kaggle. There i have word embeddings from google news, wiki, glove
in a zipped folder. I want to ...
- 1
1
vote
2
answers
47
views
Is this over-fitting or something else?
I recently put together an entry for the House Prices Kaggle competition for beginners. I decided to try my hand at understanding and using XGBoost.
I split Kaggle's 'training' data into 'training' ...
- 309
1
vote
0
answers
95
views
How to train a simple Machine learning model in batches?
I have a dataset for multiclass classification of text data. The number of samples in training data are 1,20,000. If I extract features using TF-IDF vectoriser of the sklearn library it gives about 80,...
- 153
1
vote
0
answers
28
views
What to do when feature engineering and parameter tuning don't add to the base model performance
I've been working on using LogisticRegression from scikit to try the Titanic Kaggle comp.
I've found something interesting, and that is that no amount of feature engineering and paramater tuning is ...
- 309
1
vote
0
answers
76
views
Multiple choice gap-fill question (with distractors) dataset for evaluating NLP algorithms
I am looking for a standard gap-filling multiple-choice exercise (with distractors) dataset that can be used to evaluate the NLP gap-filling ML algorithms. I expect the dataset to contain questions ...
- 53
2
votes
1
answer
236
views
Sklearn Pipeline for mixed features: numerical and (skewed) categorical
I am working on a dataset from Kaggle (housing price prediction). I have done some pre-processing on the data (missing values, category aggregation, selecting ordinal vs one-hot). I am trying to ...
- 65
2
votes
1
answer
1k
views
Extract features from Decision tree leaf nodes
Recently came across a coursera course on "How to win Kaggle competitions" where they explain how we can engineer a categorical feature from each leaf node of the decision tree.
[Video Link][1]
I ...
- 245
4
votes
1
answer
410
views
AUC ROC metric on a Kaggle competition
I am trying to learn data modeling by working on a dataset from Kaggle competition. As the competition was closed 2 years back, I am asking my question here. The competition uses AUC-ROC as the ...
- 65
0
votes
1
answer
90
views
Python and Titanic competition how to get the median of specific range of values where class is 3
I am trying to solve Kaggle's titanic competition. In the test set, there is only 1 row having Fare value as null. It's so easy to replace it with median or average of all Fare values.
But I am ...
- 163
-1
votes
1
answer
219
views
Meaning of 'hue" in seaborn [closed]
I know what "hue" does a little
I'm studying kaggle and the image is about bike sharing demand analysis
What I want to know is
shouldn't the second images's sum of each point's y axis be the first ...
- 3