All Questions
12,296
questions with no upvoted or accepted answers
22
votes
1
answer
3k
views
Understanding Timestamps and Batchsize of Keras LSTM considering Hiddenstates and TBPTT
What I'm trying to do
What I am trying to do is predicting the next data-point $x_t$ for each point in the timeseries $[x_0, x_1, x_2,...,x_T]$ in the context of a date-stream in real-time, in theory ...
KenMarsu
10
votes
1
answer
621
views
A clear visualization of a two-way ANOVA
To provide a full yet simple picture of a 3-level, one-way ANOVA, I use the following visualization where variation within each group (the filled circles) and variation between the groups (black ...
- 101
9
votes
2
answers
367
views
Are there any graph embedding algorithms like this already?
I wrote an algorithm for generating node embeddings based on the graph's topology. Most of the explanation is done in the readme file and the examples.
The question is:
Am I reinventing the wheel?
...
- 201
9
votes
4
answers
3k
views
Loss Function for Probability Regression
I am trying to predict a probability with a neural network, but having trouble figuring out which loss function is best. Cross entropy was my first thought, but other resources always talk about it in ...
- 191
9
votes
0
answers
2k
views
Why is my Keras model not learning image segmentation?
Edit: as is turns out, not even the model's initial creator could successfully fine-tune it. This is most likely a problem of implementation, or possibly related to the non-intuitive way in which the ...
- 199
9
votes
0
answers
2k
views
AdaBoost implementation and tuning for high dimensional feature space in R
I am trying to implement the AdaBoost.M1 algorithm (trees as base-learners) to a data set with a large feature space (~ 20.000 features) and ~ 100 samples in R. ...
- 91
8
votes
1
answer
1k
views
Gensim LDA model: return keywords based on relevance (λ - lambda) value
I am using the gensim library for topic modeling, more specifically LDA. I created my corpus, my dictionary, and my LDA model. With the help of the pyLDAvis library I visualized the results. When I ...
- 81
8
votes
0
answers
132
views
Training value neural network AlphaGo style
I have been trying to replicate the results obtained by AlphaGo following their supervise learning protocol. The papers specify that they use a network that has two heads: a value head that predicts ...
- 101
8
votes
3
answers
692
views
Chi-square as evaluation metrics for nonlinear machine learning regression models
I am using machine learning models to predict an ordinal variable (values: 1,2,3,4, and 5) using 7 different features. I posed this as a regression problem, so the final outputs of a model are ...
- 181
8
votes
1
answer
253
views
How to predict advantage value in deep reinforcement learning
I'm currently working on a collection of reinforcement algorithms: https://github.com/lhk/rl_gym
For deep q-learning, you need to calculate the q-values that should be predicted by your network. There ...
- 181
8
votes
0
answers
3k
views
Python : Feature Matching + Homography to find Multiple Objects
I'm trying to use OpenCV via Python to find multiple objects in a train image and match it with the key points detected from a query image. For my case, I'm trying to detect the tennis courts in the ...
- 81
8
votes
1
answer
3k
views
Decision Trees - C4.5 vs CART - rule sets
When I read the scikit-learn user manual about Decision Trees, they mentioned that
CART (Classification and Regression Trees) is very similar to C4.5,
but it differs in that it supports numerical ...
- 181
7
votes
3
answers
4k
views
Why is 10000 used as the denominator in Positional Encodings in the Transformer Model?
I was working through the All you need is Attention paper, and while the motivation of positional encodings makes sense and the other stackexchange answers filled me in on the motivations of the ...
7
votes
4
answers
4k
views
How to perform feature selection on dataset with categorical and numerical features?
I am working on a dataset with 30 columns (29 numerical, 1 non-ordinal categorical). I hot-encoded the categorical feature and reached at 35 columns. To improve training efficiency, I want to perform ...
- 71
7
votes
1
answer
9k
views
How is WordPiece tokenization helpful to effectively deal with rare words problem in NLP?
I have seen that NLP models such as BERT utilize WordPiece for tokenization. In WordPiece, we split the tokens like playing to play and ##ing. It is mentioned that it covers a wider spectrum of Out-Of-...
- 696
7
votes
0
answers
3k
views
Tensorflow v1 Dataset API AttributeError with ndim
I'd like to make pipeline for optimizing Gpu and Cpu.
Dataset
It's about 10000 datapoint and 4 description variables for the regression problem.
...
- 71
7
votes
0
answers
2k
views
Using the Python Keras multi_gpu_model with LSTM / GRU to predict Timeseries data
I'm having an issue with python keras LSTM / GRU layers with multi_gpu_model for machine learning.
When I use a single GPU, the predictions work correctly ...
- 171
7
votes
0
answers
1k
views
Multivariate, multistep forecasting with LSTM
I want to use an RNN with LSTM to forecast multiple steps into the future, based on multiple inputs. I have some ideas for different ways to approach this, but I'm afraid I'm missing the "right way" ...
- 171
7
votes
0
answers
2k
views
Fine tuning accuracy lower than Raw Transfer Learning Accuracy
I've used transfer learning on Inception V3 with ImageNet weights on Keras with Tensorflow backend on python 2.7 to create an image classifier. I first extracted and saved the bottleneck features from ...
- 71
7
votes
0
answers
925
views
ALS in Spark: what loss function is it minimizing?
I’ve playing with the MovieLens ratings dataset under Spark’s ALS and a manual implementation of ALS and comparing results with the same hyperparameters. I’d like to know this exactly in order to make ...
- 854
7
votes
1
answer
2k
views
Comparing transition matrices for Markov chains
I have a population, each unit of which exists in one of several states that change over time. I am using first-order Markov chains to model these state transitions.
My population can be segmented ...
- 1,105
7
votes
0
answers
512
views
differences between LSQR and FTRL when working with very sparse data
I have a 2M instances dataset with millions of very very sparse dummy variables created using the hashing trick = ...
- 1,357
7
votes
1
answer
336
views
When to use which multiple testing correction?
There are a large number multiple testing p-value correction methods. e.g.:
...
- 294
6
votes
2
answers
131
views
How sklearn SVM find the initial hyperplane before Optimisation?
The optimization goal of the SVM is to maximize the distance between the positive and negative hyperplanes. But before optimizing, how does sklearn first find the positive and negative support vectors ...
- 261
6
votes
3
answers
171
views
What Clustering Method Should I Use?
My data is a group of 10 thousand points (each having an node location (x,y)) that are spread across a plane. They are also chromatically-colored based on their weight.
I need to finalize a bayesian ...
6
votes
0
answers
138
views
Unable to transform (greatly performing) Autoencoder into Variational Autoencoder
Following the procedure described in this SO question, I am trying to transform my (greatly performing) convolutional Autoencoder into a Variational version of the same Autoencoder. As explained in ...
- 61
6
votes
2
answers
124
views
Data transformations in hierarchical classification
I am building a hierarchical text classifier using the Local Classifier Per Parent Node (LCPN) approach with the 'siblings' policy
as described in the A survey of hierarchical classification across ...
- 161
6
votes
1
answer
8k
views
Keras - Implementation of custom loss function with multiple outputs
I am trying to replicate (a way smaller version) the AlphaGo Zero system. However, in the network model, I am having a problem. The loss function I am supposed to implement is the following:
$$l = (z -...
- 205
6
votes
2
answers
240
views
What does big O mean in KNN optimal weights?
Wiki gives this definition of KNN
In pattern recognition, the k-nearest neighbors algorithm (k-NN) is a
non-parametric method used for classification and regression. In both
cases, the input consists ...
- 623
6
votes
0
answers
327
views
Optimal implementation of vanilla DQN loss in Keras
I've implemented vanilla DQN for continuous/non-images (no CNN) states in keras. But, I'm not sure if my implementation of the loss computation is optimal.
For reminder the loss is defined as : $loss=...
- 61
6
votes
0
answers
86
views
Fitting model to differenced time series
I have a time series on daily stock price of company(2013 data points).I took a first order difference and the following acf and pacf plots of the differenced series were obtained. However, I am ...
- 231
6
votes
2
answers
3k
views
How to deal with missing data for Bernoulli Naive Bayes?
I am dealing with a dataset of categorical data that looks like this:
...
- 161
6
votes
2
answers
325
views
Training stateful LSTM with different number of sequences
I'm using a stateful LSTM for stock market analysis, and I have varying amounts of data for each stock, ranging from 20 years to just a few weeks (i.e. for newly listed stocks).
I use 3 years of data ...
- 740
6
votes
2
answers
2k
views
Metrics to evaluate features' importance in classification problem (with random forest)
I want to evaluate the importance of each of the features of a 2000x60 dataset in a classification problem with random forest.
The most widely used ones apparrently are:
Cross Entropy-Information ...
- 1,057
6
votes
3
answers
1k
views
how does XGBoost's exact greedy split finding algorithm determine candidate split values for different feature types?
Based on the paper by Chen & Guestrin (2016)
"XGBoost: A Scalable Tree Boosting System", XGBoost's "exact split finding algorithm enumerates over all the possible splits on all the features to ...
- 71
6
votes
2
answers
197
views
Gridsearch XGBoost for ensemble. Do I include first-level prediction matrix of base learners in train set?
I'm not quite sure how I should go about tuning xgboost before I use it as a meta-learner in ensemble learning.
Should I include the prediction matrix (ie. df containing columns of prediction results ...
- 161
6
votes
1
answer
178
views
Predicting change of shapes/coordinates
I'm trying to find a way to predict/calculate how a shape (e.g. outline of a glacier) will change in the future—based on its history (previous shape) and additional factors (e.g. Δtemperature).
In my ...
- 161
6
votes
1
answer
2k
views
What is the minimum number of times a word needs to appear in word2vec training corpus for quality results?
When training a word2vec model with, eg, gensim, you can specify the minimum times a word needs to be seen (with the parameter min_count). The default value for this seems to be 5.
Are there any ...
- 203
6
votes
0
answers
299
views
Confidence value in AdaBoost?
I read this introduction about AdaBoost (http://www.cs.man.ac.uk/~nikolaon/~nikolaon_files/Introduction_to_AdaBoost.pdf), and am curious why confidence for each model is defined as
$$\alpha_j=\frac{...
- 143
6
votes
0
answers
372
views
Maths of Xavier initialization
The paper I read is Glorot et al (2010). And the math part is in Section 4.2.1.
Formula (5) and (10) make sense to me but I cannot derive formula (6) and (7) myself from (2) and (3).
I found many ...
- 61
6
votes
0
answers
95
views
Classify driver based on time-series sensor data
I want to build a model that can detect which driver is driving now the car based on a dataset that contains 20 driver records for 3600s each driver ( the dataset contains all the car sensors values ...
- 61
6
votes
0
answers
130
views
Learning a logical function with a 2 layer BDN network - manual weight setting rule question?
So I am trying to construct a 2-layer network of binary decision neurons as proposed by McCullough and Pitts (1943) to learn a logical function (a composition of AND's and OR's) such as:
$((\neg x_1\...
- 61
6
votes
0
answers
606
views
Adversarial Learning for Semantic Segmentation
I am incorporating Adversarial Training for Semantic Segmentation from Adversarial Learning for Semi-Supervised Semantic Segmentation.
The idea is like this:
The discriminator takes as input a ...
- 61
6
votes
0
answers
230
views
Connect output node to next hidden node in RNN
I'm trying to build a neural network with an unconventional architecture and a having trouble figuring out how. Usually we have connections like so, where $X=$ input, $H=$ hidden layer, $Y=$ output ...
- 2,238
6
votes
2
answers
419
views
Visualizing 28 different variables with 28 different colors?
ColorBrewer seems to be very useful in selecting a color pallet to represent factors that have up to 12 possible values.
I have 28.
Is it a horrible idea to represent 28 variables with color? If so,...
- 161
6
votes
0
answers
3k
views
6
votes
0
answers
1k
views
Keras objective function shared between outputs
Is there any way to implement a loss function that is shared between outputs? I have a 2D image output and scalar classification that are both used by a single loss function.
I have attempted writing ...
- 161
6
votes
0
answers
110
views
Fixed-radius range search in non-Euclidean space
I'm trying to find an indexing data structure most suitable for my metric space:
set of IP network related data (IP addresses, ports, TCP flags, ...),
distance function is continuous, non-Euclidean ...
- 61
6
votes
0
answers
12k
views
Tuning Gradient Boosted Classifier's hyperparametrs and balancing it
I am not sure if it is a correct stack. Maybe I should have put my question into crossvalidated.
Nevertheless, I perform following steps to tune the hyperparameters for a gradient boosting model:
...
- 286
6
votes
0
answers
289
views
how to propagate error from convolutional layer to previous layer？
I've been trying to implement a simple convolutional neural network. But I've been stuck at this problem for over a week.
To be specific, assume there are 3 layers in a convolutional pass, marked as ...
- 161