Questions tagged [training]

The tag has no usage guidance.

Filter by
Sorted by
Tagged with
10
votes
3answers
5k 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?
0
votes
2answers
18 views

Hello, when i'm training my model with 80% data and testing with 20% data the accuracy is 49% and without split it's 99%

Hello, when i'm training my model with 80% data and testing with 20% data the accuracy is 49%. And when i'm training my data without splitting it's giving around 99%. I'm confused. Please help me with ...
1
vote
1answer
23 views

Drastic increase in accuracy while using pickle file with sklearn

I trained a xgboost classifier and it gave an accuracy of 49.99 % and i saved that model into a pickle file. When i ran the same data with pickle file (.pkl) it's giving an accuracy of 88.99 percent. ...
0
votes
1answer
30 views

Question about balancing training data for sentiment analysis (machine learning)

My question is about when to balance training data for sentiment analysis. Upon evaluating my training dataset, which has 3 labels (good, bad, neutral), I noticed there were twice as many neutral ...
1
vote
2answers
30 views

What is the purpose of the 'train model' step in data mining?

My understanding is that training a model is something done in machine learning using training data so that the model can predict values when new data is given to it. Data mining is the process to ...
2
votes
1answer
21 views

Transformer masking during training or inference?

I'm working through Attention is All you Need, and I have a question about masking in the decoder. It's stated that masking is used to ensure the model doesn't attend to any tokens in the future (not ...
0
votes
1answer
15 views

Distribution of training dataset in binary classification problem

Is it important to have the same amount of positives and negatives in our training dataset for a binary classification problem? Or for example, it doesn't matter if we have 70% positives and 30% ...
0
votes
0answers
13 views

DNN predicting the same value for train+test Data

I have trained a Deep Neural architecture for regression problem and after the hundred's of epochs, model predicting the same output for both training and testing data. When I reduced the batch size, ...
0
votes
2answers
21 views

how do the number of classes in an object detection model affect accuracy?

If I have, say, a Yolo or RetinaNet Object Detection Model... if I train it with 10 vs 50 classes, (assuming 3000 training data images per class), will the model with 10 classes perform similarly to ...
0
votes
0answers
7 views

TF object detection - The total number of detected objects is not increasing

I'm building a model to recognize fishes in the aquarium (150 different fishes). I'm using a faster_rcnn_inception_v2_coco_2018_01_28 model for transfer learning from TF object detection API. I have ...
0
votes
0answers
25 views

Ram required to train YOLO v4

What is the GPU ram required to train YOLO v4? More generally what is the ram requirements to work with the more interesting architectures and bigger datasets? I see Nvidia has released the RTX 3090 ...
0
votes
1answer
74 views

Does save_best_only in Keras prevents overfitting?

I'm training a CNN and using: model_checkpoint = ModelCheckpoint(os.path.join(output_artifacts,'weights.h5'), monitor='val_acc', save_best_only=True) I trained ...
0
votes
1answer
17 views

[under/over]-sampling teaches model the wrong distribution?

TLDR: Will under/oversampling during the training phase teach the model the wrong distribution and adversely affect accuracy? Let us assume you want to train a classifier to differentiate between ...
2
votes
2answers
41 views

How to gather training data for simple voice commands?

I'm trying to build a machine learning model for recognizing simple voice commands like up, down, left, etc. On similar problems based on images, I'd just take the picture and assign a label to it. ...
1
vote
0answers
23 views

libraries for multiple machine NN training?

As detailed here, the way to go to break NN training over multiple machines/threads, is decompose training data set on multiple chunks and send to each node, then sum results back in main node. There ...
1
vote
1answer
23 views

How to use External Data Sets in test set

I have a doubt regarding usage of external datasets like gdp rate, unemployment rate... etc., in test set for time series prediction. These datasets are historical and can be used along with train set,...
1
vote
1answer
593 views

Error: An operation has `None` for gradient with categorical_crossentropy

I am trying to train my discriminator network using Keras with TensorFlow backend. The network is meant to classify the input into one of the 9 output labels. I am passing a 2D input (height, width, ...
1
vote
2answers
136 views

How interpret keras training loss without compare with validation loss?

I have several implementation of the same neural network, but each one with different starting parameter. This is one of my plot comparing the training loss of the base experiment with the training ...
0
votes
1answer
18 views

How to use Cross Validation data set?

I am new to data science, and the dataset I am working on is divided into train set, test set, and validation set. However, till now I was splitting the data with ...
24
votes
2answers
11k views

What is the advantage of keeping batch size a power of 2?

While training models in machine learning, why is it sometimes advantageous to keep the batch size to a power of 2? I thought it would be best to use a size that is the largest fit in your GPU memory /...
1
vote
1answer
41 views

Making sense of loss and accuracy curves

This is an issue that I have come across over and over again. Loss (cross-entropy in this case) and accuracy plots that do not make sense. Here is an example: Here, I’m training a ReNet18 on CIFAR10. ...
8
votes
4answers
810 views

Understanding how convolutional layers work

After working with a CNN using Keras and the Mnist dataset for the well-know hand written digit recognition problem, I came up with some questions about how the convolutional layer work. I can ...
0
votes
0answers
9 views

How to use an encoder to do feature extraction

I'm newbie with all of data science. I have a pre-trained U-Net network from which I get its encoder. Now I have to use a picture to get its features. With the whole U-Net I do this with fit method: <...
0
votes
1answer
18 views

GRU and LSTM does not “take risk” predicting

I tested LSTM and GRU models to predict the exchange rate between currencies. I do not take the raw price but a the delta with the previous day, so the data is stationnary around zero. My problem is ...
2
votes
4answers
337 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
45 views

Is a test set necessary after cross validation on training set?

I'd like to cite a paragraph from the book Hands On Machine Learning with Scikit Learn and TensorFlow by Aurelien Geron regarding evaluating on a final test set after hyperparameter tuning on the ...
3
votes
1answer
46 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 ...
2
votes
2answers
46 views

When a dataset is huge, what do you do to train with all the images on i t?

I'm using Python 3.7.7. I'm trying to load a lot of NIFTI images using SimplyITK and Numpy from the [BraTS 2019 dataset][1]. This is the code I use to load the images into a numpy array. ...
0
votes
1answer
37 views

Given is the result of the model performance. Help me with this MCQ

You also evaluate your model on the test set, and find the following: Human-level performance 0.1% Training set error 2.0% Dev set error 2.1% Test set error 7.0% What does this mean? (Check the ...
0
votes
1answer
44 views

High probabilities of success for wrong predictions

I'm studying the behavior of machine failures in a production scenario. For this, I generated random data to form my imbalanced training set, consisting of categorical data, which indicate whether or ...
1
vote
1answer
46 views

Tool for test/train automation

I need to test different datasets as well as different algorithm implementations. The current workflow looks like: Perform feature extraction from train set Train classifier on this features Feed ...
2
votes
1answer
308 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
17 views

Escaping from overfitting hell: introducing regularization vs increasing training data

I am trying to identify noisy intervals in geomagnetic data using logistic regression, working with scikit-learn. Here is a typical spectrum of the data that I am working with: In this example, the ...
0
votes
1answer
48 views

What's Joint Training in Neural Networks?

I'm having a hard time trying to find a good explanation of the process of Joint Training in Neural Networks. I already understand the concepts of Fine Tuning and Feature Extraction, and i know it has ...
0
votes
0answers
18 views

Weird train and validation curves

I would like to train a GCN for protien-ligand binding affinity regression. I use GCNConv from pytorch geometric, ReLU for all activations and Dropout (0.2) after 2 dense layers each. Using ReLU for ...
0
votes
0answers
21 views

Training loss is increasing while testing loss is decreasing and Accuracy stands still

I am following the Udacity Intro to Pytorch on Exercise 7 which is to make a model which can recognize dogs or cats. Here is my code. ...
1
vote
1answer
164 views

ValueError trying to use a pickled scikit-learn model

I am new to data science and trying to learn something. I was able to complete the prediction with 98% accuracy and i saved it as pickle model. Now while trying to predict using this model I am ...
-1
votes
3answers
43 views

Why training of a neural network will require multiple iterations? [closed]

I can't understand why training of a neural network will require multiple iterations (theoretically)? Can anyone explain why, please?
1
vote
1answer
330 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
1answer
32 views

Bad results for training set and good results for test set

When training an algorithm, I have ROC = 0.5896, Sensitivity = 0.3333, and Specificity = 0.8375. When considering the test set, Sensitivity = 1 and Specificity = 1. This could happen or is a problem? ...
0
votes
3answers
24 views

Problem of having imbalanced classes in the test set while using oversampling

I have an imbalanced dataset. My classes are 0 and 1. The number of 0 class instances is about 20 times more than the number 1 class instances. I know that I should apply oversampling after train test ...
1
vote
0answers
30 views

Fine tuning a automatic speech recognition model with my own dataset

I'm using wav2letter to develop a speech-to-text system. wav2letter has pre-built acoustic and language models which is great, however the audio that I am transcribing from is unique in comparison to ...
3
votes
0answers
44 views

Help interpreting GAN output, and how to fix it?

After a few tries, I had trained a GAN to produce semi-sensible output. In this model, it almost instantly found a solution and got stuck there. The loss for both the discriminator and generator were ...
-1
votes
1answer
23 views

How can I predict the true label for data with incomplete features based on the model learned by data with complete features? [closed]

for example, the model was learned by training data with complete features (f1,f2,f3,f4,f5,f6) but, I wonder the model can test data with incomplete features (f1,f2,f3) to attach true label into these ...
1
vote
0answers
8 views

Checkerboard artefacts vs distinct objects in GANs

I found a very good solution for getting rid of checkerboard artefacts in GANs: https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix/issues/190 Instead of using Transposed Convolution, use bilinear ...
4
votes
1answer
217 views

How many epochs to run during hyperparameter search?

If I'm doing a hyperparameter search and comparing two different hyperparameters (but not number of epochs), is there some established rule of thumb for how many epochs to run? If I just compare ...
1
vote
1answer
105 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 ...
0
votes
2answers
37 views

Validation Curve Interpretation

I have been reading about the validation_curve function from scikit learn. When I run this it takes too long. Therefore, I am plotting the results from grid search ...
3
votes
0answers
25 views

Repeated k-fold Cross Validation for time series data?

I have a relative small sample size (330 with 45 features) + it's time series data. I want to train my LightGBM regression model for best generalized RMSE score and want to use repeated CV. I use ...
1
vote
1answer
205 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
2 3 4 5
9