Questions tagged [tensorflow]

A Python library for deep learning developed by Google. Use this tag for any on-topic question that (a) involves tensorflow either as a critical part of the question or expected answer, & (b) is not just about how to use tensorflow.

225 questions with no upvoted or accepted answers
Filter by
Sorted by
Tagged with
1
vote
0answers
115 views

How do I set up my hyper-parameter space for optimizing a convolutional neural network (using packages Skopt and Tensorflow)

I just finished building a 1D CNN using TensorFlow, and I want to optimize a variety of hyper-parameters using Scikit-Optimize (skopt) (although, I would be willing to use whatever optimization ...
1
vote
0answers
21 views

Reference for Inception-v2

Cross-posted from Data Science StackExchange. The "Rethinking" paper doesn't describe the actual implementation of the Inception-v3 model in Tensorflow: an accurate description is written in model....
1
vote
0answers
23 views
1
vote
0answers
482 views

Many false positives in a custom SSD model with Tensorflow object detection API

My model has 2 classes (no background class) and is trained using transfer learning with ssd_mobilenet_v2_coco. It detects and classifies well the objects it was trained on. However, on new images it ...
1
vote
0answers
40 views

CTC classification for License Plate with two lines

I have gone through this tutorial and have understanding how CTC works for end to end text recognition. But for recognition of images with texts in two lines Would it be able to use CTC for ...
1
vote
0answers
140 views

Forming Conditional Probabilities in tensorflow

This is a repost from stack, but I think it may be more applicable here. Granted there is even a bounty for it there, if someone could answer it. https://stackoverflow.com/questions/53421179/forming-...
1
vote
1answer
222 views

Question about bounding boxes to handle false positives

I have trained a model to detect vehicle number plates. The issue is that it returns matches of partial plates, with high confidence. To eliminate partials I want to add two new boxes to the images ...
1
vote
0answers
40 views

Feature ranking for ANN

I am doing a regression analysis on 1 hidden layer feed forward NN and have 18 input features with 1 output. I Intend to do some feature ranking so it can give me an idea of the most important to the ...
1
vote
0answers
105 views

How does the internal state of Tensorflow's “dtf.nn.dynamic_rnn()” change?

I am trying to do a uni-variate forecasting and I have come across this question quite a lot. This question is regarding TensorFlow's "dtf.nn.dynamic_rnn()". NOTE: When I say internal state I am ...
1
vote
1answer
478 views

Neural networks to predict a nonlinear curve

I want to model a complex nonlinear function using neural networks (keras). Training data: input - 8500 x 176 matrix of features, output - 8500 x 8 matrix, each row corresponds to 8 points which ...
1
vote
1answer
38 views

Which algorithm for classification problem?

I want to create a ML (DL) model, that predicts the success of Facebook page-posts, based on historical data. My dataset represents a couple thousands posts, labeled 1 (successful) and 0 (...
1
vote
0answers
422 views

How to implement custom loss function on keras for VAE

I have implemented a custom loss function. While training the model, I want this loss function to be calculated per batch. ...
1
vote
0answers
496 views

Need help writing a neural network for a Pokemon battle

I'm trying to write a neural network that's able to select the optimal course of action in a Pokemon battle. In a battle, there are two different types of actions: use one of the four moves known by ...
1
vote
0answers
50 views

Decision tree / boosting tree in Tensorflow?

When I was looking for some way to break the bottleneck of memory limitation in xgboost, I found there are some boosting tree algorithm that is implemented in Tensorflow such as here. I have some ...
1
vote
0answers
35 views

How should I choose the Encoder hyperparameters to make its memory state suitable for the Decoder in a Bidirectional Neural Network?

I'm trying to implement Neural Machine Translation following the tutorial on the Tensorflow website here https://www.tensorflow.org/versions/r1.8/tutorials/seq2seq and I was able to build a ...
1
vote
0answers
38 views

Do encoders/decoders preserve composition?

Let's assume that we have an encoder E and a decoder D implemented for instance as a Convolutional Encoder-Decoder, namely ED. Let's take two inputs X and Y, and some composition operator o operating ...
1
vote
0answers
268 views

custom loss function to optimize payoff via binary decision

I have written a custom loss function that is supposed to optimize a payoff via a binary decision. However, the neural networks is struggling to convert, and I'm suspecting that there's something ...
1
vote
1answer
150 views

why the parameter 'scale' of tf.layers.batch_normalization is disabled when next layer is relu?

In the tensorflow documentation of tf.layers.batch_normalization,it is said" When the next layer is linear (also e.g. nn.relu), this(the parameter of 'scale' ) can be disabled since the scaling can be ...
1
vote
0answers
27 views

Character level RNN for converting word forms

I want to build a char RNN to convert word form from one to another, for example, singular nouns such as lion to lions. However ...
1
vote
0answers
556 views

Training of multiple time-series with different lengths

I have a lot of time series with different lengths. I would like to know what are the best practices to fit them to a Bidirectional LSTM model. The problem is a Binary Classification of Sequence to ...
1
vote
0answers
21 views

Wildly different results in not that different conditions

First I trained the network on a cloud computer, then bought a GPU and trained it locally. My code is the same, yet the results are extremely different. I tried training various networks and the trend ...
1
vote
0answers
343 views

How to select the regularization parameter between two losses?

In deep learning, the total loss commonly consists of a task-specific loss and a weight regularized loss: loss = loss_specific + lambda * reg_loss In my case (...
1
vote
0answers
25 views

Regression Model Predictions with Pseudo-Random Results

Situation I'm performing an experiment in which I will use machine learning to build a model around how fast people generally (voluntarily) react to a set of stimuli. To performs this, I will be ...
1
vote
0answers
64 views

How to use Tensorflow with an already existing Keras LSTM model

I want to perform a reinforcement learning experiment on top of an LSTM model. The LSTM model performs an entity recognition on four entities (Products, Person, Location and others). Now I want to ...
1
vote
0answers
98 views

DCGAN: How to nearest neighbor on the training set in sense of compressed sensing?

I want to find the nearest neighbor of the output compared to the training set based on this framework: https://github.com/carpedm20/DCGAN-tensorflow (I asked already generally in this post: ...
1
vote
0answers
197 views

how do you initialize the confidence score for r-cnn (Region-Convolutional Neural Network)

I follow the arguments for training an r-cnn to derive the bounding box and confidence score. https://www.cv-foundation.org/openaccess/content_cvpr_2014/papers/...
1
vote
0answers
345 views

TensorFlow: Predicting a value based on historic data

I'm trying to predict a dependent variable, which is a result of some independent variables. There may be a number of independent variables at play, but if I can approach the problem with 2 ...
1
vote
0answers
556 views

A3C implementation using Tensorflow

I'm trying to implement my own version of the Asynchronous Advantage Actor-Critic method, but it fails to learn the Breakout game. My code was mostly inspired by Arthur Juliani's and OpenAI Gym's A3C ...
1
vote
1answer
78 views

Is there a way to handle associative arrays in machine learning without casting to a fixed index?

I have a (very large) data set that I'd like to set up a categorization algorithm for. The data has the following form: ...
1
vote
0answers
300 views

How to implement “Bilinear loss” function in R/tensorflow

I was reading Resheff et al 2017[1] and trying to implement their idea for a loss function in R. I am not sure if I understand correctly. It appears simple enough: Specifically, let $a_{i,j}$ ...
1
vote
0answers
323 views

Xavier initialisation of weights for 3D convolutions

I am using 3D convolutions in tensorflow. I want to initialise the weights using Xavier initialisation. Tensorflow has an implementation of Xavier initialisation for 2D convolutions, from this ...
1
vote
1answer
105 views

Feedback on approach for supervised code completion with tflearn

We got a machine learning task in a lecture regarding deep neural networks this semester. I already have an approach which works but I am 100% sure that there is a better one. Thus I request you to ...
1
vote
0answers
93 views

Strange behavior in Tensorflow with increasing data feed speed?

I'm not sure if this is more of a Stack Overflow or a Cross Validated sort of thing, but maybe somebody has some insight as to this. I'm currently trying to train a convolutional, stacked, denoising ...
1
vote
0answers
139 views

What is the benefit of word embedding wrapper over simply adding an extra layer in RNNs

I am trying to build a sequence to sequence model using tensorflow. Tensorflow provides some functions for learning embeddings for words - tf.nn.embedding_lookup and tf.contrib.rnn.EmbeddingWrapper ...
1
vote
0answers
214 views

Unknown image classification category

I followed Tensor flow for poets and classified images into 4 categories c1,c2,c3,c4. Now I have lot of images whose category is not known but I know for sure that those images falls into only one of ...
1
vote
0answers
586 views

Online training RNN (LSTM) with Tensorflow

I have developed a basic LSTM model which predict energy consumption for t+1, now I have been looking how to deploy and retrain my model : How can I retrain my model continuously with recent data, I ...
1
vote
0answers
2k views

CRF (TensorFlow) log-likelihood becomes positive

I am currently doing a multiclass classification task on sequence data and am using tf.contrib.crf.crf_log_likelihood to compute sentence level log-likelihood values. In particular it implements a ...
1
vote
1answer
299 views

CNN with fixed batch size - repeat to fill or reduce batch size?

I'm setting up a CNN that can handle differently-sized examples so long as all examples within the same batch are of the same size. As a trade-off for this flexibility, I need to fix the batch size. (...
1
vote
1answer
379 views

What is the loss/cost function for a mixture of experts model?

For a mixture of experts model, is each expert trained individually, or are they trained with respect to the overall loss? It seems each would be trained individually, since its not a requirement ...
1
vote
0answers
85 views

RNN optimization for bigger datasets

I am training bi-directional RNN model with attention for answer selection referring this. I am attaching screenshot of tensorflow graphs. My dataset consist of Question and 5 multiple choice ...
1
vote
0answers
464 views

Deep Autoencoder in TensorFlow to learn pseudo-sentence embeddings

I am stuck with a problem since days which is to find a right set-up for an Autoencoder that learns dense representations of sparse textual representations, word order is not relevant which is why I ...
1
vote
0answers
160 views

Embedding Phrase using Recurrent Neural Network

I am trying to get a word2vec embedding of a given phrase using an LSTM RNN encoder. For example, "How are you?" => [some word2vec representation] I have pretrained model of word2vec(i already wrote ...
1
vote
0answers
399 views

Which cost function to use for float labeling?

I'm using TensorFlow to label time series using an LSTM, such that the labels to attribute to each timestep in the data are floats comprehended between -1 and +1. I'm struggling to find an adequate ...
1
vote
0answers
202 views

Deep CNN-BLSTM network's loss and cost suddenly start fluctuating - is this normal?

I have a CNN feeding into a BLSTM feeding into a fully connected layer trying to do classification. I've noticed some odd behaviour approx. 70% into my epoch - the accuracy and cost start fluctuating ...
1
vote
0answers
74 views

How to teach neural network to get meaning similarity measure for 2 texts

I have an NLP task: there are database with 2 bunch of texts (50000 strings) 1) customer complaints (i.e. "i have a problem with authorization system ...") 2) answers to this complaints (i.e. "Go to ...
1
vote
0answers
767 views

Preprocessing required for grayscale X-ray images when performing transfer learning with inception-resnet-v2?

I am attempting to fine-tune the inception-resnet-v2 model with grayscale x-ray images of breast cancers (mammograms) using TensorFlow. As the images consists of both the left and right breasts, I ...
1
vote
0answers
374 views

Multi GPU CNN performance way better than single GPU

I'm training a CNN for text classification using tensorflow. When I train the model using multiple GPUs per the tensorflow docs, the performance on the dev set goes up 5%! In the multi GPU case, the ...
1
vote
0answers
243 views

Fitting a Known Function with a Feedforward Neural Network

Given the following known function $$ f(x, t) = e^{-t}\|x\|_2 - 1, $$ where $x \in R^2$ and $t \in R$, I would like to fit it with a neural network by sampling pairs of the form $\{(x,t),f(x,t)\}$ ...
1
vote
0answers
1k views

Training accuracy suddenly drops to zero after a while in Tensorflow?

I was training a ConvNet on CIFAR10 based on the code presented in "Deep MNIST for Experts" tutorial by TensorFlow. After around a few thousand epochs with training accuracy gradually increasing as ...
1
vote
1answer
2k views

Parameters Grid Search for Keras LSTM on Time Series

How do you do grid search for Keras LSTM on time series? I have seen various possible solutions, some recommend to do it manually with for loops, some say to use scikit-learn GridSearchCV. Feedback ...