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 ...

-1
votes
0answers
13 views

Identifying blank images using Tensorflow/Keras

I have a training set wherein I have a set of blank and non-blank imageseg. /test/1/blank/files & /test/1/non_blank/files. Similarly, I have the same structure for test images.I would like to ...
-2
votes
0answers
12 views

Caffe: second Slice layer does not working [on hold]

In my proto there are 2 Slice layers. the fist one is working well and appears in the log. The second one seems does not work and it does not even appear in the log!
0
votes
0answers
16 views

DDPG - Actor Critic Network does not converge with continous Actionspace

I'm currently a bit confused. I implemented an Actor-Critic Network and depending on the setup ist either begins to converge a little but produces values far from right. Or it produces the nearly the ...
-1
votes
1answer
28 views

Is there any algorithm to determine the relevance of a word to a text, which the word actually does not actual occur in the text [on hold]

Working on a problem on identifying the relevancy of a term (word) to a text. The current algorithm like TF-IDF, TextRank, or LDA requires the word actually existed in the corpus. However that may ...
-3
votes
0answers
14 views

cannot capture a good frame of faces from IP camera [on hold]

everyone. I am working on some research about getting frames from IP camera, and then detect and recognize faces. There is some implements on the last two steps. And my first problem is to judge if a ...
0
votes
0answers
8 views

framework caffe does not work when testing

I am trying to use caffe test in order to check out the accuracy of caffennet. Pretty simple, isn't it?Really NOT. It doesn't work at all.Below I describe the steps I am following (To my mind ...
0
votes
0answers
33 views

Matplotlib shows warning and doesn't show the output images

I am learning keras and tensorflow and I am working on the cifar10 dataset from keras. I wish to plot the images of it but the code shows some warning and I am unable to see the output. Please help me ...
0
votes
0answers
9 views

CNN Validation Accuracy 1.0 and K-fold

I'm using ImageDataGenerator of Keras to make the dataset more popular and K-Fold cross validation.I can't believe that my validation accuracy is getting higher every epochs and the final answer is ...
-1
votes
0answers
12 views

How tf.transpose calculates the output shape for 5 D tensor?

I have a tensor having shape of [batch_size x 1152 x 10 x 8 x 1].There is a code here : Capsule Net ImplementationI need to know what is the output shape of the tf.transpose() operation on this ...
-1
votes
0answers
27 views

How to improve accuracy of a multi layer percepton (simple neural network)

I have been learning about neural networks and end up in the google crash course. I followed up the tutorial and got stuck in the multi_class_classification.I tried running the example model in my ...
-2
votes
0answers
12 views

Analyzing the keywords or the words that decides the sentence sentiment [on hold]

Sentence: My mobile is best because it has large screen and good speakers. This sentence is a positive. I'm getting the sentiment of the sentence by using RNN and LSTM.But i want to know why that ...
-1
votes
0answers
7 views

ImportError: No module named 'resnets_utils'

I'm totally new on deep learning and I'm trying to implement a ResNet on keras and I got this error ImportError: No module named 'resnets_utils'can anyone please tell how to
1
vote
2answers
22 views

How to add constant tensor in Keras?

What I'm trying to do is to add a constant tensor to output of network:inputs=Input(shape=(config.N_FRAMES_IN_SEQUENCE, config.IMAGE_H, config.IMAGE_W, config.N_CHANNELS))cnn=VGG16(include_top=...
0
votes
1answer
17 views

How to get the accuracy of classifier on test data in DeepLearning

I am trying to use DL4J for deep learning and have provided the training data with the labels. I am then trying to send a test data by assigning a dummy label. Without providing a dummy label, it ...
-3
votes
0answers
14 views

The opposite result of external and internal data

I'm testing the internal test data(about 750 data set) and then checking for performance(about accuracy 88%) and testing the external data(about 180 data set).The problem is that the performance is ...
0
votes
0answers
24 views

How to get probabilities when using Pytorch's densenet?

I want to do a binary classification and I used the DenseNet from Pytorch.Here is my predict code:densenet=torch.load(model_path)densenet.eval()output=densenet(input)print(output)And here ...
-3
votes
0answers
10 views

how to generate bounding box for attention map

I have got the attention map of images,but i want to crop the attention region as new samples to process. now, i'm confusing about how to generate bounding box of attention region of an image.in ...
-2
votes
0answers
17 views

Training DeepLab on custom data without a pre-trained model [on hold]

I was able to run DeepLab V3 on my own dataset, but would like to ask about how I can do so without the pre-trained model.Thanks a lot.
0
votes
0answers
11 views

H2o (R) deep Learning Binary classifcation

I hope you should answer to me. I'm trying to build a neural network binary classifier with R and with h2o package. Since as activaction function i can choose one beetwen Rectifier, Tanh and Maxhout, ...
1
vote
0answers
17 views

How to implement a message propagation block for label-relation (Keras)

I am trying to build a network architecture to classify fashion images I found in this paper https://arxiv.org/pdf/1806.09445.pdf using keras + tensorflow (like the authors in the paper).The paper ...
0
votes
0answers
21 views

AlexNet paper does not match dimensions they reported?

I am following the paper where AlexNet was introduced, and the dimensions they report just don't match with the figure they attached.The output of the first conv layer (which is 96 11x11x3 ...
0
votes
0answers
11 views

Pre-trained Weight Initialization Problem with Skip Connections in Keras

I am trying to implement pre-trained VGG-16 for weight initializing along with Skip Connections. Although model's summary shows correct number of parameters, but weights are not initialized properly. ...
0
votes
0answers
16 views

ValueError: Tensor(“Reshape_4:0”, shape=(200, 1), dtype=float32) must be from the same graph as Tensor(“w1_rx:0”, shape=(100, 200), dtype=float32_ref)

I'm trying to code a GRU in tensorflow. I am using two functions, one for initializing the weights and the other for theGRU equations. But it gives me the following error:ValueError: Tensor("...
-1
votes
0answers
20 views

How to let an ai evaluate a table [on hold]

in a game I want to find out which tradingroutes gain the most profit. For this I thought of an AI because I've watched a lot of deep learning tutorials with brain.js and tensorflow. I thought I ...
-4
votes
0answers
34 views

Predict amount of money inside shops [on hold]

I'd like to predict how much money I should get to shop for the next day basing on historical data (about two years). My goal is to not remain without money the next day, and trying to keep this ...
0
votes
0answers
16 views

How to implement keras custom metrics for multi-class multi-label accuracy (Jaccard index)

Keras Custom Accuracy for Jaccard index from SemEval-2018 Task 1 Task E-cI am implementing Jaccard index from SemEval-2018 Task 1 Task E-c, it's a multi-class multi-label problem(metrics definition ...
1
vote
1answer
21 views

TypeError: cannot unpack non-iterable int object

Im trying to make my first CNN using pyTorch and am following online help and code already people wrote. i am trying to reproduce their results. I'm using the Kaggle Dogs Breed Dataset for this and ...
-4
votes
0answers
22 views

Binary classifier(Fire detection) [on hold]

I want to create 2 binary classifier model1) Fire detection2) Smoke detectionCan you please suggest which will be better to train LBP cascade(It require annotated data) or SVM/Logistic ...
0
votes
2answers
32 views

Neuron freezing in Tensorflow

I need to implement neurons freezing in CNN for a deep learning research,I tried to find any function in the Tensorflow docs, but I didn't find anything.How can I freeze specific neuron when I ...
-4
votes
0answers
20 views

Show matching items in clothing [on hold]

how do I recommend complimentary matches for an item that a user has selected?For example when an user selects a shirt, we have to show the matching products like pant , shoe , watches..etc.What are ...
0
votes
0answers
15 views

Stuck in Training Pet Model in Google Cloud

I have been following this tutorial.For the training Pets dataset in Google Cloud. I have done every step and my training command looks like the following :-gcloud ml-engine jobs submit training `...
1
vote
2answers
24 views

What does keras normalize axis argument does?

I am a beginner in deep learning and I am working upon the mnist dataset in keras.I used normalization astf.keras.utils.normalize(x_train, axis=1)I don't understand what does the axis argument ...
0
votes
0answers
10 views

Makefile:554: recipe for target '.build_release/lib/libcaffe.so.1.0.0-rc3' failed

I am compiling Caffe framework to run faster RCNN.I did 'make all j4' and got error:collect2: error: ld returned 1 exit statusMakefile:554: recipe for target '.build_release/lib/libcaffe....
0
votes
1answer
27 views

Modify layer parameters in Keras

I am interested in updating existing layer parameters in Keras (not removing a layer and inserting a new one instead, rather just modifying existing parameters).I will give an example of a function ...
0
votes
0answers
26 views

CNN with attention layer for multi-class classification

Number of label - 1000Each document (restricting to 600 words else padding) can contain 0or at-least 1 labelWord representation of 300D for each word - use of pretrained glove(standford)Applied ...
-3
votes
0answers
28 views

How can I classify these aerial images?

I want to use R to classify high res aerial images (4 band tiffs). The images are of residential food gardens in Portland. I want to train a model to be able to identify if there is a food garden ...
0
votes
0answers
8 views

Input Tensor with no shape in saved SqueezeNet .pb model

I am trying to run inference using a saved SqueezeNet model from official Tensorflow lite docs. I am required to pass input and output tensor from the model as in the code below. The input tensor for ...
-3
votes
0answers
16 views

machine learning algorithm to predict next animal behavior

What are the appropriate programming languages to train a model for predictions of next movement of an animal?I use R language to train a model.And I need a suitable mathematical model also. I use ...
-1
votes
0answers
10 views

detach().numpy() for pytorch float tensor turn 3 digit float numbers to integers

flloat_torch.detach().numpy() change the 3 digit float numbers in the tensor to integers.for example: 102.3000 changes to 102.What should I do?Thank you very much
-3
votes
0answers
22 views

better way for reshaping the images to patches with tensor flow

I want to know what is the best method to reshape the input images to patches required for convolutional networks.Right now i have found two methods.First method:Create a single tensor-flow ...
2
votes
0answers
26 views

A strange loss descent(looks like a 'cliff') appeared in training 3d Unet with weighted cross-entropy loss or dice score loss

The figure of dice loss:The figure of weighted cross-entropy loss:The optimizer is Adam, lr=0.0002, beta1=0.5, beta2=0.999. Does anyone have the same problem as me? Can you tell me the solution ...
-1
votes
0answers
38 views

Reading video frames for 3D-CNN

I am planning to train a 3D-CNN based autoencoder for anomaly detection.But, i have never worked with video dataset before.I have a dataset which consists of 300 videos.But i am confused that how to ...
7
votes
3answers
111 views

Why do I fail to predict y=x**4 with Keras? (y=x**3 works)

I manage to predict y=x**2 and y=x**3, but equations like y=x**4 or y=x**5 or y=x**7 converge only to inaccurate lines?What do I do wrong? What could I improve?import numpy as npfrom keras.layers ...
0
votes
1answer
32 views

How to fetch data to train my neural network model in Kaggle?

I am trying to build a convolution neural network which classifies between cats and dogs,to train that network I am using Kaggle as a tool ,I have uploaded all the data sets consisting of 8000 ...
0
votes
0answers
24 views

What are the ways to calculate the error rate of a deep Convolutional Neural Network when the network produces different results using the same data?

I am new to the object recognition field. Here I am asking the ways to calculate the error rate of a deep CNN used in the object recognition field. Hopefully, these ways are broadly accepted in the ...
1
vote
1answer
38 views

Concatenating features of two pooling layers

I am trying to design a Bi-Directional LSTM model and I want to concatenate features after Max pooling and Average pooling layers.I have this for my model:from keras.layers import Dense, Embedding...
0
votes
1answer
53 views

Why do I fail to predict a linear equation (Y=2*x) with Keras?

I tried to predict a linear equation (Y=2*x) with Keras, but there it failed.With a sigmoid activation function I get rectangular predictions, with ReLu I get NaN´s.What is the cause? How could I ...
0
votes
1answer
19 views

'Tensor' object has no attribute '_keras_history' Keras with no Tensorflow tensor

This code:a=Input(ish)for i in range(a.shape[1]):x=Conv2D(filters=50, kernel_size=3, padding='same', activation=rl)(a[:,i])x=MaxPooling2D(pool_size=2)(x)x=Dropout(0.5)(x)x=Conv2D(...
0
votes
0answers
25 views

How to reuse variables scope for batch normalization

I want to use same network twice. So, I need to reuse variables.I am using batch normalization function given below and it gives me error.This problem is similar to https://github.com/tensorflow/...
0
votes
0answers
10 views

What is the difference between Stacked auto encoders and Deep belief networks?

I was reading the research paper titled "A Study on Deep Neural Networks Framework"It describes DBNs and Stacked autoencoders. But after reading the description of both it seems as if both are the ...

153050per page
angop.ao, elkhabar.com, noa.al, afghanpaper.com, bbc.com, time.com, cdc.gov, nih.gov, xnxx.com, github.com,