All Questions

28,346 questions
Filter by
Sorted by
Tagged with
3 views

Can a siamese neural network distinguish expected from unexpected changes?

Please redirect me to another stack exchange if this isn't the appropriate forum. I am interested in finding a neural network architecture that can detect distance between two inputs. As an example, ...
9 views

How do I sum the last three rows based on condition?

I have a table like this: Date Revenue Outlier Month 01/2018 333 No M0 02/2018 125 No M1 03/2018 136 No M2 04/2018 123 No M0 05/2018 496 No M1 06/2018 134 No M2 07/2018 223 Yes M0 08/2018 145 ...
8 views

Can data science be used to detect fileless malware cyber attacks?

Is it possible to detect fileless malware attacks with a machine learning model or with any other data science application? Fileless malware exists only in a computer’s random-access memory (RAM), ...
6 views

Training for Named entity recognition with sparse labels

I am training an NER pipeline - It was pretrained. My label appears sparsely - Once every 100 or 200 sentences. Do I have to train my pipelines with ALL sentences or can I speed the training up by ...
10 views

Correct way of computing dice score for image segmentation?

In binary image segmentation, for given a set of images, it's true mask and predicted mask. How to compute dice score?, should I compute dice score for each image separately and then find mean across ...
10 views

Looking for modern algorithms using NN Language Model implementations addressing polysemy in NLP tasks, including text classification, question answering and topic modeling. Transfer/Zero-short ...
7 views

How to remove spikes from data with Python using signal.find_peaks

I need to make a regression model to estimate data values in future. Train set contains occasional spikes that make my model less accurate, thus I'm trying to locate and remove them. I've used ...
10 views

plot time series in text by python

I have a txt file (here) and i have tried to plot it with pandas, xarray of python I tried with ...
7 views

Margin of error for imbalanced discrete set

I'm evaluating the performance of a classifier regarding its false negatives. The classifier performed over 9090 samples, from which 9000 were labeled as negative. I randomly chose 800 samples (out of ...
16 views

Interpretation of Loss and validation Loss in Keras

I am building a model to predict one label by taking one feature as an input. The two variables seems to be strongly correlated. I wanted to build a (sequential) Neural Network model with Keras in ...
15 views

Feature egineering

I have hourly data regarding the number of visits to the website. A single record consists of date, time(0-23) and number of visits. In a separate table I have data about the display of advertising ...
7 views

Difference between MAP@K for recommendations, MAP from Precision Recall Curves and Macro-Precision

I have been using the 3 metrics independently for a while now, but trying to figure out if they are actually 3 separate things (with similar-looking definitions/names) or there is some underlying ...
12 views

Cross Validation with Artificial Neural Network in Keras

I needed to use cross-validation in the Neural network. I used kf = KFold(n_splits=5, shuffle=True) and ...
19 views

Selecting the split point for continuous attribute

For the above table, midpoints for possible split points are 22.5 and 35. I have calculated the entropy and gain for each value and 35 had the minimum Entropy and highest gain. Is it correct ? Entropy ...
17 views

Implementing a confusion matrix for a CNN model

I have created a convolutional neural network model which classifies malware based on the Malimg dataset. Below is the model and the results from the first run: ...
19 views

Dividing dataset into Training, validation and test set

I have a dataset of 535 classes (258 194 images) and I want to divide it into train, validation and test set. I am a new learner in image classification with CNN, should I do that manually or is there ...
7 views

How to do encode this target vector containing strings

Consider a target vector like ...
2 views

Use tf.image.ssim_multiscale for a batch of images in sequential model

I am trying to use SSIM as loss value for my Keras Sequential model. The output value is a set of 6 images - (6, 32, 28, 3). I want to implement a custom loss function for the model.fit() function but ...
11 views

Precision or Recall when dealing with critical cases

I am new to AI and I don't get really well Precision and Recall for my case. I have to create an AI that classify mutliples objects in order to accept them or not inside a machine. The problem is that ...
8 views

How to train videos without split into frames?

I am a beginner to machine learning. I'm building a tennis stroke classification system using action recognition. The dataset consists of different stroke videos. In all the tutorials I referred, the ...
16 views

Data visualization multiple choices multiple condition conditions

Say I have a element that base on a certain condition (1, 2, 3 or 4), it chooses a decision (A,B or C). What diagrams can be used to present this data if I have more than one thousand elements so that:...
13 views

How to improve machine learning model using 2+ datasets

I am building a supervised machine learning model which (for example) predicts heart failure (yes/no). I have two datasets from 2 different labs A and B, which both have decent distribution, aka it's ...
8 views

Classification based on color clustering

I need to classify some domain specific images by analysing their color distribution. I have annotated data; this last classification step is supervised. After some preprocessing and masking and other ...
9 views

Data wrangling dates

I have a feature with data creation dates. I have normalized them all to the same format and split them to 'day', 'month' and 'year' columns. But now I have a question. Should I apply normalization or ...
7 views

Choosing correct Period for SARIMA on a 4 month (120 day observation) dataset

I'm testing SARIMA on the first 4 months (observations are in days, total 120 days) of this dataset: https://api.covidtracking.com/v1/states/ny/daily.cs I'm not sure what value I should assign to the ...
16 views

Non linear regression using scikit learn [closed]

What are the different non linear regression techniques and how many of them are available on scikit learn?
6 views

Coupling Layer in RealNVP and NICE

Why is a coupling layer called so, in NICEand RealNVP density estimation? Does it have anything to do with the coupling of probability distributions? What is the intuitive goal of using these layers?
8 views

Cluster images labels in some given categories using word embeddings

Given: set of images Labels in string format each one. Also I've given a set of Categories, also in string. ($Images \neq Categories$) Goal: I need to map given labels to given categories to "...
20 views

Original k-Means Research Paper

I'm having difficulty searching for the original published paper proposing k-Means as an algorithm. I have been inspired to find it as reference for similar work, inspired by this TowardsDataScience ...
9 views

How to impute when you have multiple variables not important to the column with the missing values?

I have a dataset with 18000 rows and 192 columns. I have a specific column with more than 2000 rows missing. I've tried some types of imputations but they just take to long or just seem not good ...
9 views

Is it possible to load a model object to splunk to score?

I have been working on some graph embedding deep learning algorithms for anomaly detection using some log files. My issue is that I need to use splunk found on a system different than my model make ...
12 views

...
9 views

Tf/Keras LocallyConnected layers

I recently stumbled upon these layers, and it's an interesting idea, but I'm not sure what one would do with it. What is the application of a Convolutional layer that doesn't share its weights? Where ...
15 views

What does updated alpha mean in LDA model?

I'm trying to understand LDA model by reading through implementations of the algorithm. Many implementations update alpha during training iterations with codes like: ...
14 views

Analysis of changing correlation

Given a correlation matrix M and after an event, M changes. What is the best way to analyse the change? Verify significance (presumably T-test, ANOVA..) Currently looking at the %, count & average ...
18 views

Can identifiers be used to train a model?

I recently participated in some Machine Learning competition where we were asked to decide whether a rider should accept or not a course (~2k riders and ~140k courses). It came up that some of the ...
105 views

Follow Up Question - Why use fit when already have fit_transform? [duplicate]

This is a follow up question to: What's the difference between fit and fit_transform in scikit-learn models? I want to know why should we use fit at all when we ...
28 views

How to have Multiple classes in a single video?

I am building a swimming stroke classification system using a CNN. l I assume each stroke contains 3 steps/classes ('Ready', 'Impact', 'Finish'). I want to train a model which will predict whether the ...
15 views

Is the usage of the “momentum” significiantly superior to the conventional weight update

The "momentum" adds a little of the history of the last weight updates to the actual update, with diminishing weight history (older momentum shares get smaller). Is it significiantly ...
15 views

how to do regression and classification in a same time?

I was thinking of doing a job including regression firstly, then doing the classification. I read lots of sources that are saying it is Semi-Supervised learning ...
17 views

3D CNN model always predict the same values

I am currently trying to build a 3D-CNN model to predict video quality. When I trained the model with a small dataset (300 videos) it predicts different values, but when I use a huge dataset for ...
25 views

Does Homoscedasticity applies only for linear regression models?

In statistics, a sequence of random variables is homoscedastic if all its random variables have the same finite variance. This is also known as homogeneity of variance. (Wikipedia) Is this assumption ...
11 views

What is the architecture of PoseNet?

PoseNet is state of art approach and I am using it for a pose estimation. I read on the internet that pose net uses models like MobileNet or VGG etc. I would like to know that is PoseNet itself a ...
19 views

Pandas groupby show value counts

I have a dataframe with 3 columns: job_type (5 categories), age_group (5 categories) and answer (0, 1, 2 or 3) job_type | age_group | answer By ...
13 views

Theoretical question around multiclass classification

Assume there's a customer dataset with monthly installments repaying goods purchased in advance. The objective is to build a model predicting who is going to be bad debtors and who can afford to buy ...
11 views

Representing multi-channel input signals with a single signal

I am working on an EEG signal classification problem. My dataset consists of EEG signals stored as 19X30000 NumPy arrays. Each row represents a single channel. For now, I am converting each of the ...
24 views

Testing accuracy very low, while training and validation accuracy ~ 85%

I have a training dataset of 10000 pictures and a test dataset of 15000 pictures. There are 23 types of birds. First of all, I imported the necessary ...
9 views

Keras EarlyStopping callback: Why would I ever set restore_best_weights=False?

The point of EarlyStopping is to stop training at a point where validation loss (or some other metric) does not improve. If I have set ...
16 views

Understanding math notation in infoGAN paper

I'm reading this paper about mutual information in infoGAN infoGAN_paper_link and already have the code to run it. I pretty much found code for it which is fine and dandy except for the fact that I ...
25 views

Does multi-head attention remove the need for self-attention?

The title may be confusing but suppose I were to build Transformer Neural Network with a masking network that utilizes multi-head attention (like that in SepFormer), would adding self-attention in the ...

15 30 50 per page