Stack Exchange Network

Stack Exchange network consists of 175 Q&A communities including Stack Overflow, the largest, most trusted online community for developers to learn, share their knowledge, and build their careers.

Visit Stack Exchange

Questions tagged [training]

The tag has no usage guidance.

3
votes
1answer
23 views

Validation data shall be in broken down into batches or not?

I am using fit_generator to train the model. The training dataset is being read from a generator function which gives data in a constant batch size. Now I want to ...
0
votes
0answers
12 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
19 views

review: gradient descent, epochs, validation in neural network training

These days, training data aren't put in gradient descent all at once. Rather, they are put in batch after batch. Gradient descent is run once for each batch of training data. When all batches are ...
1
vote
1answer
33 views

Using SMOTE for Synthetic Data generation to improve performance on unbalanced data

I presently have a dataset with 21392 samples, of which, 16948 belong to the majority class (class A) and the remaining 4444 belong to the minority class (class B). I am presently using SMOTE (...
0
votes
1answer
20 views

Accuracy and Loss in MLP

I am trying to explore models for predicting whether the a team will win or lose based on features about the team and their opponent. My training data is 15k samples with 760 numerical features. Each ...
0
votes
0answers
38 views

What to do when Kfold is not enough?

I have a dataset made of roughly 100 time-series and my final goal is to obtain a classification of each point (detection problem). To do so I have labels so I decided to use an XGB model to perform ...
0
votes
0answers
16 views

How to train non image data in batches from disk?

I am working on a project where I have 50 .npy files with each of shape (77156, 30, 50, 1) representing ...
1
vote
1answer
13 views

Change rate of cross validation data, after training

Say we have N of labeled data, and we need to take some part for the cross validation (we will skip test part for this case). We ...
0
votes
0answers
12 views

FaceNet training, tripletloss not decrease but accuracy increase then stuck,what are possible causes?

as you can see,triplet loss(pink curve in the left) do not change,but accuracy increase then stuck,what are possible reasons?
0
votes
0answers
14 views

How can I train a RNN on different sets of data?

I have a practical question about RNN training, the answer to which I can't seem to find no matter how hard I try (googled a lot). Can I train an RNN that has a time element in it, where said time ...
1
vote
1answer
27 views

Issues with training SSD on own dataset

I'm new to ML and trying to train a SSD300, with some Keras-Code github.com/pierluigiferrari/ssd_keras I found on github. For training I'm using an own (very small) dataset of objects that are not in ...
0
votes
1answer
28 views

Is “adding the predictions to the real data for new training and prediction” a good idea for LSTM?

Considering we have trained our model with a lot of data for "many-to-one" prediction. Then we like to forecast the future data of next 10 days. So we use last 60 of existent data and predict the ...
1
vote
0answers
11 views

Performance diagnostics in mxnet gluon (e.g. plotting training vs validation loss over time)?

Tensorflow has tensorboard, is there any recommended way to plot classification error/loss over time in mxnet?
2
votes
1answer
32 views

Is splitting the data set into train and validation applicable in unsupervised learning?

I am having a tough time implementing all the steps of setting up support vector machine (SVM) for unsupervised learning. My data set is labelled but for educational purposes I am learning ...
1
vote
1answer
15 views

Suggestions for labeling data for named entity recognition [closed]

Is it good to label the data based on sub category than parent category? For example: for drugs data ... label the drugs dose as drug_dose or label the drug dose as different type of dose like ...
1
vote
1answer
25 views

Is it ok to train the model only on the interested part of the data?

Let's say I have a dataset where one feature is Car type : say 'A', 'B' and 'C'. The test set consists of samples where car type is 'A' always. Therefore, should I train my model only on the subset ...
0
votes
0answers
23 views

Validation curve

I'm learning about data science and I've been checking several tutorials. Now I'm trying some validation curves on the problem sample I'm resolving and I'm having some troubles with it. This is the ...
3
votes
0answers
31 views

Training deep CNN with noisy dataset

I am training a Mask RCNN model with a train dataset that has been generated from some simple computer vision operations (color thresholding) and some morphological filtering. The train set captures ...
3
votes
4answers
116 views

Weights not converging while cost function has converged in neural networks

My cost/loss function drops drastically and approaches 0, which looks a sign of convergence. But the weights are still changing in a visible way, a lot faster than the cost function. Should I ensure ...
2
votes
3answers
37 views

weight training speed too slow in CNNs

I'm writing my own CNN code from scratch. Though I got fast, converged and satisfactory results, the trained weights change very little in value (while cost/loss function drops in time rapidly in a ...
2
votes
1answer
116 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 ...
1
vote
1answer
23 views

Computing number of batches in one epoch

I have been reading through Stanford's code examples for their Deep Learning course, and I see that they have computed ...
0
votes
0answers
15 views

How to train LSTM with previous cell's prediction as an input in Keras?

At the moment I'm using a simple Keras model to learn a sequence of items and after it using the trained model to generate new sequences . I want to change the training to be in the same manner as ...
0
votes
1answer
31 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
0answers
19 views

Train and Test Error dependence on size of data

I was reading about Adaboost Alogorithm and learned that if initially we split the dataset equally for training and testing, apply the algorithm. And then gradually start increasing the size of train ...
2
votes
0answers
27 views

Correct approach to usage of class labels in cell imaging data

As part of a group project at university, we are given a series of videos of cell cultures over a 24 hour period. A number of these cells (the "knockout" cells) have had a particular gene removed, ...
1
vote
0answers
19 views

Unexpected shape of “training curves” in NN

I'm trying to find the best configuration for my NN (in terms of batch size, learning rate etc) and noticed the following unexpected behavior. The AUC scores, computed on validation data, as ...
2
votes
1answer
49 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 ...
1
vote
1answer
35 views

Face Recognition (Scalability Issue)

Background I would like to build a face recognition model for registration and login for some kind of service. For example, using this approach (CNN + SVM). When a new user wants to register a ...
1
vote
0answers
48 views

Training a convolutional neural network for image denoising in Matlab

first time posting in this stack exchange. I am currently trying to train CNNs to remove Poisson noise from images. The software I am using is Matlab 2018b, however the results I am getting are poor. ...
0
votes
0answers
14 views

I have created a model that classification sentences. How do I create a good dataset when I train this model further?

I used bidirectional lstm I have a model that classification as spam and general trained with about 130,000 data. The model has 90% accuracy for sentences over a certain length, but 75% accuracy for ...
0
votes
0answers
24 views

How to cross validate a DNN model?

You have a binary classification model giving a decent score on selected metrics. The model has been trained using early stopping. The epoch with the lowest loss is kept. Now you want to cross ...
0
votes
1answer
20 views

Confusion on Delta Rule and Error

I'm currently reading Mitchell's book for Machine Learning, and he just started gradient descent. There's one part that's really confusing me. At one point, he gives this equation for the error of a ...
0
votes
2answers
30 views

Normalization before or after resizing

I'm training deep learning network using images (to be exact - I'm solving semantic segmentation problem). What's the proper order of resizing (I need to resize images to fixed width X height) and ...
0
votes
0answers
4 views

effects of number of running time periods on examining the DQN quality?

What are the effects of the number of running time periods on examining the DQN quality? I mean "T": time periods of training and testing. If there is not an obligation to set it to a value in the ...
0
votes
0answers
19 views

How apply Reinforcement Learning in the following case?

Suppose that I have to move from point A to B and I have to choose among 3 different paths. But we don't know the traffic in each path, so what is the training rule to use to learn the best behaviour? ...
1
vote
0answers
197 views

“10-year-challenge” data for age algorithms? [closed]

Both on FB and IG, I see people posting themselves before 10y and now. I have no idea how this challenge started. Could it be a way to collect a colossal amount of data, that could be used to train ...
0
votes
1answer
26 views

Difference in labelling and normalizing train/test data

I am working on a dataset comprised of almost 17000 data points. Since it's a financial dataset and the components are many different companies, I need necessarily to split it by date. Therefore, ...
1
vote
3answers
47 views

Is it possible to make a CS:GO Machine Learning AI? [closed]

I am not an expert on Machine Learning, Neural Networks or NEAT. In fact, I probably have no clue what I'm talking about. My question is if you can make a learning AI that learns to play complex ...
0
votes
0answers
16 views

Semantic segmentation training on images of different sizes - good practices

What are the good practices of handling images when training the neural networks for semantic segmentation, but the images have different sizes and aspect ratios? Also, how to properly handle small ...
1
vote
0answers
16 views

Strangeness in validation loss between CPU vs GPU when training CNN

I've been training an implementation of Mask R-CNN and it was training very successfully on my CPU but I've just set up my GPU and it is giving some strange results when looking at my validation loss. ...
1
vote
0answers
122 views

How to train the generator in a recurrent GAN (Keras)

I am trying to train a Recurrent GAN that is meant to generate geospatial movement data (sequences of 3-tuples of latitude, longitude and time). You may simply consider it a sequences of vectors with ...
1
vote
1answer
24 views

Conjugated gradient method. What is an A-matrix in case of neural networks

I am reading about conjugated gradient methods to understand how they exactly work. I understand that a pair of vector $u$ and $v$ are conjugated with respect to $A$ if $u^TAv=0$. I also read that $A$ ...
2
votes
4answers
69 views

Neural Network unseen data performance

I started dabbling in neural networks quite recently and encountered a situation which is quite strange (at least with my limited knowledge). The problem I'm using a NN is a regression problem which ...
1
vote
2answers
151 views

What's the advantage of multi-gpu training in real?

The decreasing speed of training loss is almost the same between one gpu and multi-gpu. After averaging the gradients, the only benefit from multi-gpu is that the model seems to see more data in the ...
-1
votes
0answers
53 views

Triplet loss training problem

My results are very poor and I cannot make out the reason on why is it so? I am using euclidean distance measure for hard mining of triplets. It is prior to training with the initial random set of ...
2
votes
1answer
43 views

Sequential Modelling: Multiple Sequence to One or Sequence to Sequence

Suppose I have a single sequence of $x_1, x_2, ..., x_n$ and corresponding labels $y_1, y_2, ..., y_n$. An example would be a person makes website visits $x_i$ and the label $y_i$ tells us if there ...
0
votes
1answer
20 views

Can accuracy become worse on the training set with more epochs?

I know that overfitting occurs when the accuracy on the training set improves but the accuracy on the validation set decrease. So, we must stop the training. I would like to know if this is a rule ...
0
votes
2answers
64 views

Validation accuracy is always close to training accuracy

I am trying to tune the hyperparameters of a LSTM I have to do time series forecasting. I have noticed that my validation accuracy is always very close to my training accuracy. I am not sure whether ...