We’re rewarding the question askers & reputations are being recalculated! Read more.

Questions tagged [training]

The tag has no usage guidance.

Filter by
Sorted by
Tagged with
-1
votes
0answers
4 views

BERT pretraining hardware information [on hold]

I want to pre train BERT with a dataset of legal documents. Can I do it on google colab with TPU runtime ?
0
votes
2answers
32 views

Oversampling/Undersampling only train set only or both train and validation set

I am working on a dataset with class imbalance problem. Now, I know one needs to oversample or undersample only the train set and not the test set. But my issue is: whether to oversample the train set ...
2
votes
1answer
352 views

Why is performance worse when my time-series data is not shuffled prior to a train/test split vs. when it is shuffled prior to the split?

We are running RandomForest model on a time-series data. The model is run in real time and is refit every time a new row is added. Since it is a timeseries data, we set shuffle to false while ...
0
votes
0answers
8 views

saving a model during training of an RL agent

I am training an RL agent using PPO2 algorithm. Iam using stable-baselines library. During the training process, my rewards are slowly increasing and stabilizing, but are falling down suddenly. I ...
66
votes
2answers
46k views

Training an RNN with examples of different lengths in Keras

I am trying to get started learning about RNNs and I'm using Keras. I understand the basic premise of vanilla RNN and LSTM layers, but I'm having trouble understanding a certain technical point for ...
0
votes
1answer
55 views

How to prepare training data for deep learning models

I am working on a project which involves the application of deep learning models. I have collected training data. In collected images, I have more than one object in interest. I am not very clear how ...
0
votes
1answer
98 views

Unsupervised Learning and Training Data

As far as I know, we need to use training data to find out the relation between the features, also known as input values, and labels, that are output values, in supervised learning. After that, by ...
0
votes
1answer
18 views

Target encoding before split on skewed data

Hi My data is distributed like follow: And I only have categorical variables on many many levels. As I need to make a regression task I thought about doing leave one out encoding on my categories. ...
0
votes
0answers
14 views

CNN models comparison

I coded a 38 layer CNN and 8 layer CNN but there's something wrong in my 38 layer CNN, which doesn't learn anything. Not able to fugure out what's wrong. They were trained on CIFAR.
0
votes
1answer
52 views

Transformer for neural machine translation: is it possible to predict each word in the target sentence in a single forward pass?

I want to replicate the Transformer from the paper Attention Is All You Need in PyTorch. My question is about the decoder branch of the Transformer. If I understand correctly, given a sentence in the ...
4
votes
2answers
180 views

Training a model sample by sample

I'm training a Scikit model but it seems that in all examples, they call the fit method on the entire training set. What I want to do however is call it per sample (...
2
votes
1answer
659 views

Incrementally Train BERT with minimum QnA records - to get improved results

We are using Google BERT for Question and Answering. We have fine tuned BERT with SQUAD QnA release train data set (https://github.com/google-research/bert , https://rajpurkar.github.io/SQuAD-explorer/...
0
votes
1answer
39 views

What could make a set of the train data more predictive than the whole train data

I took a sample of my training data and balanced it and then trained my model. The results obtained are more accurate than using the whole set of train data (balanced or imbalanced). My question is: ...
0
votes
1answer
33 views

Potential problems with expanding training set

The problem is a binary classification one. My dataset contains users with activity over multiple days, where they all start with class 0 and can become class 1 after a certain activity (which is not ...
3
votes
1answer
224 views

Why do most GAN (Generative Adversarial Network) implementations have symmetric discriminator and generator architectures?

For example, if the discriminator is a vanilla network of n layers, each with n(i) units, then, typically, the generator will also be a vanilla network of n layers, each with n(n-i) units (except the ...
0
votes
0answers
4 views

Training Algorithm for pointcloud data (3d point data)

I have a "PointNet" neural network which theoretically can work with any number of points as input. I have trained the model using an equal number of points from each object. That is fairly simple and ...
0
votes
0answers
11 views

Accuracy to big when train yolov3

I am trainning yolov3. My result is not as my expected. I think accuracy must be less then 1.0 But I got accuracy and avg too large, in this case is 1577.76 average. Do I still working well?
1
vote
1answer
78 views

Training data : forecasted or actual?

I am working on a time series prediction problem. I am using keras models for machine learning. For this prediction, weather variables are used as input. They can be of two types: forecasted and ...
0
votes
2answers
18 views

Suggestions on how to explain 'models' & 'algorithms'

I guess other members of this Stack have ran in to this before, but I may be wrong: Have you ever been approached and asked to explain the difference between models and algorithms? This happened to be ...
1
vote
0answers
11 views

Grape disease detection

i'm trying to realize a detector for diseased grape leaves, for this par of the project i'm just interested in detecting lets say, the percentage of diseased to healty leaves and/or place a flag where ...
0
votes
0answers
9 views

What should be ideal ratio for number of unique target label vs number of training set samples

For a multiclass classification problem in Machine Learning, is there any rule for ratio of number of target class values vs number of training samples? For example, I have 2000 records to train on ...
8
votes
5answers
8k views

tool to label images for classification

Can anyone recommend a tool to quickly label several hundred images as an input for classification? I have ~500 microscopy images of cells. I want to assign categories such as 'healthy', 'dead', 'sick'...
28
votes
4answers
24k views

Should a model be re-trained if new observations are available?

So, I have not been able to find any literature on this subject but it seems like something worth giving a thought: What are the best practices in model training and optimization if new observations ...
2
votes
3answers
3k views

Training set, validation set, and test set with Orange

Is it possible with Orange (only using its widgets, without writing Python code) to implement the following typical machine learning processes? Train a training set, Validating a validation set (e.g....
1
vote
1answer
61 views

How to split a dataset into train and test sets for time series (multiple step-multiple output forecasting)?

I am trying to use an LSTM neural net to do multiple step / multiple output forecasting (I predict multiple values in one time knowing some values in the past). But, I have realized that I must be ...
1
vote
1answer
13 views

How to deal with annotation errors?

I know my annotators are not perfect, sometimes making mistakes. What would be the best way to deal with the annotation errors for my training data? Thanks!
0
votes
0answers
3 views

Is there any relation between binary/ternary quantization using in deep learning and fuzzy?

I am new with binary/ternary quantization but its structure seems to have some relation with fuzzy. Am I in right way? Is there any relation between binary/ternary quantization using in deep learning ...
0
votes
2answers
30 views

How to deal with a feature that has lot of categorical values?

I know this question has been asked before and I have tried a few things but those things are not working as expected for my usecase. I have a 500 length feature vector. One of these features is a ...
0
votes
1answer
96 views

LSTM loss function and backpropagation

I'm trying to understand the connection between loss function and backpropagation. From what I understood until now, backpropagation is used to get and update matrices and bias used in forward ...
0
votes
1answer
16 views

Transfer learning VGG16 does not work as expected. (Detect tacos as hamburgers)

I am new in this field of machine learning, to test I wanted to do a simple project. Create a cnn capable of recognizing hamburger images. As I do not have the ability to collect more than 10,000 ...
0
votes
0answers
21 views

How do I compare more than 20 deep learning models?

I have to compare several deep learning models (CNNs) based on the same dataset. For estimating the model skill's I use the train_test_split instead of ...
2
votes
3answers
6k views

Determine useful features for machine learning model

I am working with a dataset with hundreds of features. I wish to create a simple machine learning model using 7-10 features from the original dataset. My question is this: What quantitative metrics ...
0
votes
1answer
57 views

clarification on train, test and val and how to use/implement it

So far I think I understood the differences between the training, test and validation set. Basically it is like in this image: Training set: The data where the model is trained on Validation set: ...
0
votes
2answers
105 views

What can be the cause of a sudden explosion in the loss when training a CNN (Deeplab)

I am training the following deeplab CNN: https://github.com/tensorflow/models/tree/master/research/deeplab During training I see the following loss: The first 50k steps of the training the loss is ...
1
vote
1answer
159 views

Training an ensemble of small neural networks efficiently in TensorFlow 2

I have a bunch of small neural networks (say, 5 to 50 feed-forward neural networks with only two hidden layers with 10-100 neurons each), which differ only in the weight initialization. I want to ...
2
votes
1answer
49 views

Given a machine learning algorithm, what is the minimum size of the training set for it?

I understand that the more data we have, the more reliable is our model trained on that data. I also understand that the more parameters a machine learning model has, the more training data it ...
1
vote
1answer
17 views

How to construct validation set for time series for NN?

I would like to train my model with a validation set. As the data is a time series I have to use timeseriessplit: ...
1
vote
0answers
18 views

How to teach algorithm to mimic paths in a certain enviroment

I have a set of scenarios which represent the movement of a car in a certain environment containing some obstacles. So for each scenario I have the position of the car (x,y,t) and a description of the ...
0
votes
0answers
5 views

Training data for Tesseract

I have to train my tesseract to detect different variations of a letter for example, u,û,ü,ù should all mean u. Is this possible and if yes how should I train it and how should the dataset be made ...
0
votes
0answers
22 views

Issues with Implementation of CNN based on Paper

I am attempting to duplicate a CNN in a paper, and am having issues with bad accuracy and loss not decreasing past 40. As described in this paper, specifically on pages 6/7, the network architecture ...
0
votes
1answer
39 views

How to deal with a constant value as an output from neural network?

I am using feedforward neural network for regression and what I get as a result of prediction is a constant value visible on the graph below: Data I use are typical standardised tabular numbers. The ...
2
votes
1answer
52 views

Training a model where each response in the observation data has a different known varience

I have a dataset where each response variable is the number of successes of N Bernoulli trials with N and p (the probability of success) being different for each observation. The goal is to train a ...
1
vote
3answers
191 views

Meaning of stratify parameter

I'm training a Neural Network and I'm trying to divide my data into training and testing sets. I have a lot of output classes and for some of them I have as little as 2 examples, so I would like to ...
2
votes
1answer
76 views

how is correct usage of the validation split in neural networks?

I have a dataset separated in train, test and validation splits. After each epoch, I evaluate the loss and accuracy in the validation split. When the loss in validation split is not better, I stop ...
0
votes
0answers
18 views

How to avoid different accuracies when training with subsets?

when trying to train a CNN with randomly selected small subsets (each same size) of the training data set, I get different results in accuracy (the accuracy varies from 0.75 to 0.85). I determine the ...
3
votes
2answers
66 views

Why can we not split train test data with 0.01 as parameter or 99% training data

Most of the blogs mention about a good thumb rule to be 80-20 split for the train and test respectively. My special case is a time series dataset and it is for the stock prices, which IMO is very ...
1
vote
1answer
100 views

Optimization methods used in machine learning

I don't have too much knowledge in the field of ML, but from my naive point of view it always seems that some variant of gradient descent is used when training neutral networks. As such, I was ...
0
votes
0answers
20 views

Beating Roulette with Neural Networks, YoloV3, and PyTorch

Background: I am in my last semester of electrical engineering, and I am working on my senior design project. The senior design project is a two-semester design project in which students outline, or ...
2
votes
1answer
24 views

Running multiple times of a model is for model randomness or data randomness?

When a paper report the average and std of a model on a dataset, it means that they have changed the split of training and test sets and run the model multiple times or they just run the model on ...
3
votes
2answers
645 views

In the context of Deep Learning, what is training warmup steps

I found this term "training warmup steps" in some of the papers, what exactly does this term mean? Has it got anything to do with "learning rate"? If so, how does it affect?