Questions tagged [deep-learning]
Deep Learning is an area of machine learning whose goal is to learn complex functions using special neural network architectures that are "deep" (consist of many layers). This tag should be used for questions about implementation of deep learning architectures. General machine learning questions should be tagged "machine learning". Including a tag for the relevant software library (e.g., "keras", "tensorflow","pytorch","fast.ai" etc) is helpful.
deep-learning
27,889
questions
-2
votes
0
answers
8
views
Why we use concatenation in U net [closed]
why we use concatenation in unet and how did we get this grey lines. pls can someone elaborate this I'm new to this
I tried U net architecture but I'm unable to understand how we get the grey lines i....
0
votes
0
answers
23
views
Sign Language alphabet identification using Machine Learning giving wrong predictions
I'm just a beginner on machine learning and I've watched tutorial videos on YouTube in order to come up with my code. Our professor has only given us less than a week to learn and complete this ...
- 1
0
votes
0
answers
11
views
How to Estimate GPU Memory for training and inference, Data Requirements, and Training Time for Large Language Models?
Today, I faced this question during an interview for an ML Engineer position. I didn't answer it perfectly at the time. How should I answer it ideally?
Assume we have models like Transformer, BERT, ...
- 1,437
-2
votes
0
answers
13
views
Binary classification: Accuracy always equal to 1 [closed]
training a CNN for a binary image classification task. However accuracy is always shown as 1 on validation and test scores. the loss curve does not show overfitting though.
tried to add class weights ...
- 1
-1
votes
0
answers
26
views
Accuracy improving gets so slow after some epoches in neural network training [closed]
I have something like 70 million sample to train a neural network model, the accuracy improving goes very well and fast until somewhere like 25-30 epoch, after 25-30 epoch it gets so slow.
e.g.
epoch ...
- 1
-4
votes
0
answers
29
views
Which machine learning technique should I use to solve my sequence problem? [closed]
I have a set of data like the below:
Column A
Column B
Column C
Column D
Tom
Monday
Blue
0,3,20,36,80,98,100
Dan
Friday
Red
0,15,45,100
Sarah
Sunday
Red
0,6,31,91,100
Dan
Monday
Yellow
0,21,86,...
- 1
0
votes
0
answers
16
views
Darts (Python) TiDE model gives irrational results when switched to probabilistic mode
Let's consider this simple code:
import yfinance as yf
import pandas as pd
from darts import TimeSeries
import darts.models as dm
import darts.utils.likelihood_models as dl
import lightning.pytorch as ...
- 741
-2
votes
0
answers
10
views
Compare hyperparameters influence on model performance
I want to assess which Hyperparams seem to influence the most my results, if there is any kind of "threshold" value for a given or several hyperparams after which the model does not improve ...
- 7
-1
votes
0
answers
19
views
Seeking Techniques to Automate Transformation of Irregular Excel Layouts to Structured Datasets
I am facing a challenge with automating the transformation of irregular Excel layouts into structured datasets. These Excel files often contain:
Merged cells
Hierarchical columns
Annotations and notes ...
-2
votes
0
answers
23
views
Choosing the Right Normalization Layer for Audio Classification Neural Network [closed]
I am building an audio classification neural network. I have learned that it is essential to normalize the data during network design.
However, I discovered that there are various types of ...
- 1
-1
votes
0
answers
14
views
Extracting text from an TMT (thread mill test) report [closed]
I have a task that involves extracting specific values from a TMT PNG image of a table. Depending on the required output, I need to extract either a specific value from a table cell or some text from ...
-2
votes
0
answers
11
views
How to annotate VQA dataset? [closed]
Can't find a useful annotation platform or software? How did everyone solve this problem. And how to organize the JSON of the dataset.
I used label me and label studio, but the results were not good, ...
-2
votes
0
answers
13
views
Seeking advice on creating datasets and tools for real-time Indian Sign Language conversion project [closed]
I am currently working on a project aimed at real-time conversion between audio/text and Indian Sign Language (ISL), utilizing 3D character animation. However, I am encountering challenges in two key ...
-3
votes
0
answers
28
views
Why the Intel Iris performs better than RTX 3050 laptop GPU in training a Yolov8 model [closed]
I have Dell G-15 Gaming laptop with Core i5 12500h and Rtx 3050. I am new to deep learning and AI. I am training a model with Yolo v8 for Automated Number Plate Detection with a custom dataset of 28k ...
-2
votes
0
answers
37
views
More CPU cores lead to less performance in Deep Reinforcement Learning
We are working on a deep RL project and are encountering an issue where switching the training from our laptops to a workstation with more CPU cores causes the performance to drastically drop.
An ...
- 7