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.
11,275
questions
0
votes
0
answers
7
views
Decent specification for a starter machine learning system
I will be working on applying machine learning models on biological data. I am a bioinformatician and new to data science / machine learning. The data set (tentatively) contain data from 6k samples, ...
- 101
0
votes
0
answers
16
views
How do you appropriately measure the real mean squared error of a box cox transformed linear regression model?
My understanding is that it can make sense to transform the outcomes of a linear regression model to make them more normally distributed. That's because it could 1) help me find more linear ...
- 101
0
votes
1
answer
12
views
feature engineering mechanism
why do we need to rescale some feature having large range I know we do it for faster rate of gradient descent ,but still how does rescaling works? and it doesn't break the model
and does rescaling ...
1
vote
1
answer
22
views
Doubt in gradient , vanishing gradient problem in Back propagation
As per my knowledge, in back propagation- loss function or gradient is used to update the weights.
in back propagation, weights became small w.r.t gradients, this leads to vanishing gradient problem.
...
- 37
1
vote
0
answers
20
views
How to find own way in data field? [closed]
I need help finding my path in the data field. I just completed my MSc in Big Data at university. During my studies, I gained approximately Junior level Python and SQL programming skills, worked with ...
- 11
0
votes
0
answers
15
views
Seeking Solutions for Generating Text Descriptions from Diagrams, Infographics, Charts, etc
I am currently on a quest to find an efficient way to generate meaningful text descriptions (or alt text) from visual representations such as diagrams, infographics, charts, plots and the like. ...
- 101
0
votes
1
answer
26
views
Spam tweets dataset needed [closed]
Where can I find labelled spam tweets dataset that has at least 50000 rows. I've searched everywhere; but there are very few datasets available, and those datasets are way too small.
I am in urgent ...
0
votes
0
answers
23
views
Please give me suggestions to identify scammers from communication data
I have some communication behavior data of carrier subscribers, containing information on calls made, duration and base stations. What do I need to do with this kind of data or what kind of machine ...
- 9
1
vote
1
answer
20
views
0
votes
2
answers
125
views
How to interpret k=1 when summing over k? Is there something wrong with this equation?
I am having a hard time understanding the following equation to do with document clustering.
Is there something wrong with it?
$$\ln \left(\prod_{n=1}^N \sum_{k=1}^K p\left(x_n \mid z_n, k=1\right) p\...
- 57
0
votes
0
answers
36
views
Error in tsfresh feature extraction: TypeError: ufunc 'isnan' not supported for the input types
I'm trying to use the tsfresh library for feature extraction from time series data. Specifically, I'm using the extract_features function with the EfficientFCParameters as the default feature ...
0
votes
0
answers
10
views
Approaches to improve the performance in defect prediction of source code?
I have the task to do defect detection on C source code (on function level) starting from this repo (using RoBERTa): https://github.com/microsoft/CodeXGLUE/tree/main/Code-Code/Defect-detection
It does ...
0
votes
0
answers
13
views
Binary latent representation
I've been working on a problem where I got stuck at encoding my data into a binary latent representation, most of the methods out there aren't really working for my case.
I have input for the encoder ...
0
votes
0
answers
10
views
What kind of learning do I need ? (use-case specific)
Consider a scenario where I have a model trained on gesture videos (say a 3D ResNet). I am looking for a technique (or a combination) that allows me to further train the model every time I have a new ...
- 139
0
votes
1
answer
29
views
Clustering: is it common that data just cannot be grouped?
I'm currently in the middle of a clustering project and struggling to get acceptable results, is it commonplace that datasets just can't be clustered?
Context:
I'm trying to cluster a relatively small ...
- 101
0
votes
1
answer
27
views
Neural Nets: Difference between activation and activation function, error on Wikipedia?
I'm reading the Wikipedia page on backpropagation and have some questions about the following equations:
$$
\frac{d C}{d a^L}\cdot \frac{d a^L}{d z^L} \cdot \frac{d z^L}{d a^{L-1}} \cdot \frac{d a^{L-...
- 101
0
votes
1
answer
20
views
Query about Sigmoid activation function calculation
While applying sigmoid activation function (in finding y label), I have calculated it as below:
y = 0.35 + (0.8 * 0.1) + (0.3 * 0.6) + (-0.2 * 0.4) = 0.53
sigmoid_y = 0.625
how do we take threshold ...
- 37
0
votes
0
answers
38
views
what is the difference between word2vec and doc2vec
As we know Word2Vec is a non-contextual embedding, here it maps the words in global vocabulary and returns their corresponding vectors (at word level).
In case of Doc2Vec, hope this is also non-...
- 37
0
votes
1
answer
42
views
RMSE too high when trying to create a machine learning model in Python
I am new using Python/ML. Right now, I am trying to create a model to forecast the expected call volume for a company. However, the RMSE that I am getting is higher than expected
Here is my code. I ...
0
votes
1
answer
22
views
more insights about Word2Vec implementation
As we know Word2Vec is non-contextual embedding (at word level). As per my knowledge, BOW is statistical embedding technique (word level).
we can perform Word2Vec embedding in two approaches: 1. CBOW. ...
- 37
0
votes
0
answers
33
views
Proven approaches for labeling recordings for training a speech-to-text model
To train a deep learning model, for converting speech to text, we need labeled data. How should this data be arranged?
I'm interested in building the model from scratch and not using ready-made ...
- 101
0
votes
0
answers
29
views
An OCR model that can be easily improved on the user side
I am building OCR software, for this purpose I trained a model on many types of fonts, the model is SVM or NN but it is not ...
- 101
0
votes
1
answer
42
views
Is it ok to normalize data using minmaxscalar on dependent variable?
I'm trying to make a sales prediction using the column X = item_amount and y = item_price_total, I'm confused whether it's okay to normalize data on the dependent variable using minmaxscalar?
With the ...
- 1
0
votes
1
answer
27
views
what is the difference between NSP and text prediction
In BERT, NSP (Next Sentence Prediction) is for predicting next sentence based on context and Text prediction task is also for predicting next word or phrases.
So, both are for predicting next sentence ...
- 37
0
votes
1
answer
27
views
what are the best sources to learn about statistics?
am interested in quant analytics. hope statistics is heart for quant analysis.
so, what are the best sources to learn core statistics (in depth) and what is the best approach to learn statistics?
- 37
0
votes
0
answers
9
views
What are the Bidirectional methodoly based pre-trained models in NLP
BERT model is useful to solve many NLP tasks (11+) like Sentiment analysis, Q&A, summarization and NER.
what are the other predefined model similar to BERT?
what is best way to find them in NLP?
- 37
0
votes
0
answers
26
views
Language model gradient sensitivity to insignificant tokens
I'm trying out a method to identify important training samples for a given test-time prediction. What it essentially boils down to is calculating the gradient of a test-time prediction and ordering ...
- 11
0
votes
0
answers
20
views
Generator loss keeps increasing while discriminator keeps decreasing
I am trying to build a GAN to generate LEGO images however my generator is not working at all. I have tried changing the learning rates but it caused the loss to go even more higher, sometimes into ...
0
votes
1
answer
40
views
What's Best way in selecting right model for document comparison
We have different pre-trained models like BERT, USE, ELMo, Word2Vec, FastText, etc..,
we have documents in different sizes (large, medium, small). now, we want to do document similarity. how can we ...
- 37
1
vote
1
answer
48
views
What are the preprocessing steps for text classification after removing stopwords?
I am working on an NLP project where I have text that I need to categorize based on topics (The data is 2 columns, text and topic).
Something that I am stuck on now is the preprocessing part. What are ...
- 13
1
vote
1
answer
97
views
How to handle OOV in non-contextual embedding (word2vec, Glove, FastText)?
how non-contextual embedding (Word2Vec, Glove, FastText) handle OOV (incase if given word is not available in vocabulary)
- 37
1
vote
2
answers
67
views
What is the ELMO approach to learn contextual embedding?
BERT, GPT, and ELMo used the contextual embedding. but, their approach of learning contextual embedding is different.
so, what is the ELMo approach to learn contextual embedding?
- 37
0
votes
0
answers
9
views
Recommendation for ML framework that allows running untrusted code to train models while maintaining data privacy?
We are a deep learning company looking to find a way to allow contractors to create and run training code that we can then run on private datasets, without giving the contractors access to all the ...
- 101
0
votes
0
answers
266
views
CUDA error: "no kernel image is available for execution on the device" - which PyTorch version to use?
I'm running some ML python code on Amazon's EC2 - the machine has a GPU: NVIDIA A10G. The ML code is written using PyTorch.
When I run the code I get error:
...
- 173
0
votes
1
answer
138
views
Difference between Word2Vec and contextual embedding
am trying to understand the difference between word embedding and contextual embedding.
below is my understanding, please add if you find any corrections.
word embedding algorithm has global ...
- 37
0
votes
0
answers
19
views
Machine Learning Geographical Data with NaNs
I have some features (physical properties related to geographical etc. of the Earth), I have a target that I'd like to predict. Sometimes this target is covered with clouds so I cannot see its actual ...
- 91
0
votes
0
answers
34
views
I have a strange loss behavior in Tensorflow/Keras. It goes smooth towards better values, and then suddenly it increases rapidly before falling again
I have a model that trains on a dataset of 100000 data points. The model is a single dense layer with a single neuron.
During training, I use a batch size of 1000 and train for 250 epochs. After ...
- 1
1
vote
2
answers
39
views
which hyperparameters are returned as best in cross validation?
The description on the RandomizedSearchCV says this about best hyperparameters :
"Estimator that was chosen by the search, i.e. estimator which gave highest ...
- 31
0
votes
0
answers
37
views
how to predict the arpu for a monthly cohort dynamically?
The main idea of this project, is to predict the ARPU (Average Revenue Per User) 11 month after subscription of a cohort with a monthly subscription, using minimum number of delays (a delay is a month ...
1
vote
1
answer
35
views
improving Neural network regression model
I have the following toy data (which closely mimicks my original larger data used for the project):
...
- 139
0
votes
1
answer
77
views
Why COST FUNCTION AND MSE IS CALLED THE SAME?
Why are the cost function and mean squared errors called the same thing? WHEN THE COST FUNCTION IS 1/2M AND THE MSE IS 1/N. AND M=N
2
votes
1
answer
46
views
Difference among ID3, C4.5, C5.0
The C4.5 algorithm uses information gain ratio instead of information gain like ID3, and it also adds pruning.
What does C5.0 add more? Is there any example of code? I looked on the web but there is ...
- 152
0
votes
0
answers
25
views
Unsupervised clustering approach validated using internal data
I have used mclust package in R software for unsupervised clustering approach and choosing the clustering result according to the minimum BIC value. Can I use the cross validation method and calculate ...
- 1
0
votes
0
answers
24
views
cost function including constrain on vector norm
I'm playing with collaborative implementation using numpy.
As a reminder, we are given a matrix $R$ of user ratings for movies. Let's assume there are 3 users and 4 movies. The data matrix we are ...
- 1
0
votes
1
answer
29
views
How to get the relationship between a result and a data
I have big data (at least 10 variables each row and millions of lines) and result (for example if y=0.2 and x=0.4 and z=0.9 so there is a system failure).
I need to find the relationship between the ...
- 1
0
votes
1
answer
22
views
Inverse reinforcement learning with trajectories only
Inverse reinforcement learning (IRL) is a task that can learn a reward from other agent behaviour. Most IRL paradigms assume that dynamics of environment is known, that is the transition probability ...
- 515
1
vote
2
answers
85
views
Why are normal distributions so important in deep learning?
I am currently reading on normalization/standardization techniques as well as batch normalization in deep learning and I don't really understand why normal distributions are so important inside deep ...
1
vote
0
answers
28
views
Feature Importance in Stacked Model
I have built a stacked model using mlxtend StakingCVClassifier. I want to know the feature importance scores now. Is there any way I can calculate feature importance scores for the stacked model? If ...
- 11
-1
votes
2
answers
88
views
How to generate synthetic dataset with dependent and independent features?
I want to generate a synthetic dataset, in python, composed of 10 features. 5 are dependent among each others and 5 are independent.
Generate dependent features.
...
- 57
3
votes
0
answers
57
views
Where can I find implementation of the various improvements of K-nearest neighbors (KNN)?
I have been facing some challenges where traditional KNN algorithm perform well. I'd like to explore more advanced knn solutions. While researching possible solutions, I came across a paper titled
<...
- 2,513