Questions tagged [machine-learning]
Machine Learning is a subfield of computer science that draws on elements from algorithmic analysis, computational statistics, mathematics, optimization, etc. It is mainly concerned with the use of data to construct models that have high predictive/forecasting ability. Topics include modeling building, applications, theory, etc.
10,473
questions
0
votes
0
answers
10
views
Relation between batch size, number of steps, and learning rate
Taking alphazero training setup as a reference:
700k total steps
batch-size of 4096
initial LR of 0.2
What would be an equivalent setup for a batch-size of 1024? ...
- 101
0
votes
0
answers
8
views
How to reduce the size of Bert model(checkpoint/model_state.bin) using pytorch
I used torch.quantization.quantize_dynamic to reduce the model size but it is reducing my prediction Accuracy score.
I'm using that model file inside the Flask and ...
0
votes
0
answers
11
views
How do machine learning applications communicate with one another?
My understanding is that if a company creates a machine learning application, they use an API to ensure that applications talk to one another, which is what I found in my initial research (https://www....
- 101
0
votes
0
answers
5
views
How to handle tags/lists with CatBoost?
I have database like this:
Id, type, category1, category2, tags
1, ‘cosmetics’, 123, 456, [446, 354]
2, ‘electronics’, 234, 213, [55, 978, 12]
…
And I want to predict some value with ...
- 1
0
votes
0
answers
10
views
Selecting Right data and amount of data
Multiple Data Sources are needed to make Credit or Fraud Decision on (loan/credit card) applications. But each data pull costs money to buy. So how do we optimize
a) Number of data sources to pull ...
2
votes
2
answers
32
views
How to avoid numerous Hyperparameter tuning in ML?
Suppose I have developed a dynamic system for forecasting the future of some specific stocks. As time passes, the train set will change dynamically. For a better understanding, consider this example:
...
- 121
1
vote
1
answer
18
views
Grid Searching seed in randomized machine learning
I was wondering if tuning a seed with cross-validation in order to maximize the performance of an algorithm heavily based on a randomness factor is a good idea or not. I have created an Extra Tree ...
- 13
1
vote
0
answers
9
views
How can I implement lambda-mart with lightgbm?
I have a learning to rank task at hand and I want to use the lightgbm implementation of LambdaMART. I'm also following this notebook.
...
- 664
1
vote
0
answers
7
views
0
votes
1
answer
9
views
How to Scale target feature
How should I scale target feature? Should I use scaler as fit_transform on y_train, and just fit on y_test to avoid leaking data?
0
votes
1
answer
41
views
Newly discovered learning rule
Does anyone know how this algorithm performs the learning process for neural networks?
I've stumbled over this solution. It works, but I don't know how and why.
It's neuron-local and works without ...
- 101
1
vote
0
answers
17
views
On-Device Football Detection Model not performing up to the par ; misdetections
I have trained a football detection model. I have so far trained the models using RCNN, SSD (backbone MobileNet), CenterNet and others. SSD and Centernet, so far have been the best in terms of speed ...
- 11
0
votes
1
answer
22
views
+100
How to exercise Quality Assurance Engineering principles to Artificial Intelligence systems?
In deterministic (software) systems we have a set of business requirements and ideally, given enough resources, such a system can be fully defined of which are the expected outputs for each inputs or ...
1
vote
2
answers
24
views
Loss decreases, but Validation Loss just fluctuates
I've been trying to implement object detection using a CNN architecture like this:
...
0
votes
1
answer
20
views
Odd error when training neural network with Keras - Error occurred when finalizing GeneratorDataset iterator
I'm attempting to train a neural network to evaluate chess positions. I have around 100 CSV files each with about 10,000 positions, leading to roughly 1,000,000 positions in total. Because of the ...
- 59
0
votes
1
answer
24
views
What what will happen if all the layers of a MLP or any DL architecture are set as same in the beginning?
Setting the initial weights as all zeros will have the output dependent on the bias and setting the weights of all the neurons of a layer as same, will update the gradients in same way thus removing ...
- 313
0
votes
0
answers
9
views
Using the dnn model on data with a small amount attribute
I have made some Medical information classification model using tensorflow and keras.
I make some classification model that have two input.
That are the time series data[like signal data] and two of ...
2
votes
1
answer
32
views
Why we do random sampling when we select the training set?
The usual workflow when building a machine learning model starts with random splitting the data set into training and test set.
What I can't understand is why we do this. For example lets say we have ...
- 131
1
vote
0
answers
19
views
How to do Feature clustering?
I have different datasets and I want to find out the features that are similar among the datasets. The datasets are of varying sizes.
example:
dataset1 has columns a,b,c,d,e
dataset2 has columns m,n,o,...
0
votes
0
answers
10
views
What configuration of output neurons to use for detecting bias
I am trying to make a deep learning model that detects political bias in media articles for my local community. There are two political parties here and I have a dataset of biased articles from both. ...
- 125
1
vote
1
answer
28
views
MAE divided by median metric
I have a regression task for which my best models has a Mean Absolute Error (MAE) of approximately 15,000. The median value of the target variable is approximately 150,000.
I want to report that the ...
- 65
2
votes
1
answer
17
views
Choosing Right Optimiser and Data Scaling
The choice of optimiser and how data is scaled are both very important things in machine learning, yet they are not hyperparameters (as far as I am aware). It is also not necessarily obvious which ...
- 61
2
votes
0
answers
10
views
What is the correct way to compute hit rate in recommender systems?
I'm working with the famous Movielens 1M dataset and implemented some simple recommender algorithms. While computing the hit rate, I found that it's very low $(\approx 0.008)$ but the papers seem to ...
- 131
0
votes
0
answers
18
views
Error in model.fit in keras [migrated]
I was building a model for a classification problem in keras. However, it is throwing the following error:-
...
1
vote
1
answer
21
views
Different results between hyperparameter optimisation and actual training/val values
If I want to do a hyperparameter optimisation on a dataset using e.g. hyperband or random search, I note that some of the models being randomly chosen seem to have rather good R2 scores, MSE etc.
I ...
- 61
0
votes
0
answers
10
views
Keyword suggestion rule learner question
I have a mock exam question as the following:
Q:
You have a set of documents D = {di}, where each document is assigned to an arbitrary number of keywords from a fixed set of keywords. There is no ...
- 11
0
votes
0
answers
5
views
Why would an affine forward layer take unflattened inputs?
I'm doing Stanford's CS 231n on my own.
I'm tasked with implementing the forward function for an affine forward layer.
Here's the doc comment:
...
- 115
0
votes
0
answers
16
views
Mix of time-dependent and constant features for a transformer
I'm using the transformer architecture to predict future time-points from previous time-points.
Each item of the input sequence is a vector of ...
0
votes
0
answers
17
views
How to calculate dataset and feature sparsity/density
I have a dataset with 8 features and 30,000 samples but which is probably a sparse sampling. I would like to quantify how sparse or dense the dataset and individual features are, as described in the ...
- 101
0
votes
0
answers
20
views
Are there examples of quantization aware neural networks
I'm looking for examples of Machine Learning / Neural Networks examples that work with quantized weights, activation functions,.... The simple approach of training with floating point parameters and ...
- 101
0
votes
0
answers
13
views
Which dataset is used to measure system performance? [closed]
For a machine learning or deep learning model, we have a testing, train, and validation dataset. Which set is used to measure system performance?
- 175
0
votes
0
answers
10
views
Images get rotated during training
I am trying to train a ssd_mobilenet_v2_keras for object detection on a dataset of more or less 6000 images. The problem is that images are rotated randomly during training (or at least, this is what ...
- 101
1
vote
1
answer
43
views
How to select relevant columns from a dataset with many features
I have a dataset with a large number of potential features (>100) and I am interested in finding a relatively small subset of these (maybe on the order of 5, or 20) features which is best suited to ...
- 35
0
votes
0
answers
10
views
How can we determine if the model does require an image or not while predicting?
I've this sentiment analysis task, where the dataset comprises an image and a comment for prediction. I want to determine if the image is really necessary for the task or not, is there any way I can ...
- 1
0
votes
0
answers
14
views
Is Proximal Gradient Descent exactly same with coordinate descent for LASSO
We have the lose function of LASSO:
$$L(\theta)=F(\theta) + \lambda||\theta||,\quad F(\theta) = \sum\limits_i(y^{(i)}-x^{(i)}\cdot\theta)^2.$$
And the minimal ...
- 127
0
votes
0
answers
11
views
Why is the PPO agent in RL giving negative rewards after each iteration during the training process and what are the possible hyperparameter values?
I am using the mujoco simulator as my training environment. I loaded Ant-v3 for the agent to train on. It is persistently producing negative rewards after each iteration performed.
- 1
0
votes
1
answer
16
views
What is meant by averaging inhibits it in the paper 'Attention is All You Need'?
Could anyone explain to me about the sentence below? What is meant by averaging inhibits it?
Multi-head attention allows the model to jointly attend to information
from different representation ...
0
votes
1
answer
19
views
Why would we add regularization loss to the gradient itself in an SVM?
I'm doing CS 231n on my own. I'm looking at this solution to a question that implements a SVM.
Relevant code:
...
- 115
0
votes
1
answer
19
views
How to measure similarities between two datasets with same features?
I have multiple datasets with the same features, a few numerical and a few categorical. The only difference is that they are market behavior for different countries. I wanted to know if there is a way ...
- 103
-1
votes
0
answers
7
views
Object Detection for Image where border is BBOX annotation
I'm wanting to train a object detection model where it contains images of different vehicles. let's say this is a sample image for which I want to use in my training set. as you can see this image ...
0
votes
1
answer
31
views
Automated feature selection - Best practice to avoid data leakage?
This question relates generally to all automated feature selection approaches. In my particular scenario, we have a python package called tsfresh and multiclass classification.
What has been done so ...
- 29
1
vote
0
answers
21
views
Hashtag-based Tweet similarity
I have a big dataset consisting of tweets including hashtags and I want to build a hashtag-based similarity engine to get the most similar tweets given a set of hashtags.
In the end I would like to ...
- 11
0
votes
0
answers
36
views
Why isn't my backprop matching autograd?
Am attempting to implement backpropagation for a deep learning course but my backprop gradients don't seem to be matching the gradients you get from autograd. Here's the code:
Is my math incorrect ...
1
vote
1
answer
13
views
Advice on vision ML classifier pipeline
I'm a neuroscientist and I've become quite good at using ML to predict a variety of variables of interest, however have no experience with vision ML. My aim in 2022-2023 is to learn vision ML so that ...
0
votes
0
answers
12
views
How to tune LSH implementation for better precision and recall [closed]
I am using locality sensitive hashing (LSH) algorithm/technique with murmur hash in the near-real-time system to find near-duplicate documents. But its precision and recall values are very low (around ...
- 111
1
vote
0
answers
20
views
Near duplicate detection algorithms for a near real time system
I'm looking for near-duplicate detection algorithms or techniques for a near-real-time system with large document volumes. I know LSH is the most popular industry-standard algorithm for syntactical ...
- 111
5
votes
4
answers
377
views
Calculating Potential Usefulness of Acquiring Additional Data [closed]
Imagine Anne has a labeled training dataset for a machine learning prediction problem. There is an opportunity to acquire more data from an agent, at a cost. However, before she decides to acquire ...
- 59
1
vote
1
answer
11
views
Model to implement Question Answering System over structured data
I need to write a program(like a chatbot) that retrieves an answer from a CSV datafile based on a question user asks. So for example if the CSV stores list of products and its specifications in 5-10 ...
0
votes
0
answers
14
views
Are neural networks(or any other machine learning tools) capable of predicting the next element in "1 1 0 1 0 0 1 0 0 0 ..." sequence?
Is neural network can solve tasks that involving some type of counting, for example this sequence
...
1
vote
1
answer
20
views
Calculate RMSE based on R squared and vice versa
If for example I have the value of RMSE can I calculate the $R^2$? And vice versa if I have the value of $R^2$ can I calculate the value of RMSE?
I have all predictions, dataset, training set, and ...
