Questions tagged [keras-layer]

The tag has no usage guidance.

0
votes
0answers
17 views

Concatenate input_layers in Keras raises AttributeError (Functional API) in R

I want to define a model in Keras, which has multiple inputs. Those inputs are similar in nature and should use the same layers in forward- and backward passes through the network. If I define a ...
2
votes
0answers
16 views

Custom RNN Layer with previous input and output time steps as Inputs

How can I extend the SimpleRNN or create a custom layer that could use not just the previous output y[n-1] but also the previous output y[n-2] and the previous inputs x[n-1], x[n-2]? def call(self, ...
1
vote
1answer
20 views

How to freeze specific layers of a model inside a model?

My keras model is made up of multiple models. Each "sub-model" has multiple layers. How do I call out the layers in the "sub-model" and set trainability / freeze specific layers?
0
votes
1answer
28 views

keras pre-trained ResNet50 target shape

I am trying to use ResNet50 Pretrained network for segmentation problem. I remove the last layer and add my desired layer. But when I try to fit, I get the following error: ValueError: Error when ...
1
vote
0answers
23 views

Mixing numerical and categorical data into keras sequential model with Dense layers

I have a training set in a Pandas dataframe, and I pass this data frame into model.fit() with df.values. Here is some information about the df: df.values.shape # (981, 5) df.values[0] # array([163, ...
0
votes
0answers
51 views

AttributeError: module 'tensorflow' has no attribute 'get_default_graph'

I am doing some task related to image captioning and I have loaded the weights of inception model like this model = InceptionV3(weights='imagenet') But I am getting error like this: ...
1
vote
1answer
21 views

What is the difference between dropout layer and dropout parameter in any keras layer

What is the difference between dropout layer and dropout parameter in any keras layer ,both are works for same purpose or different. like :model.add(LSTM(100, dropout=0.2, recurrent_dropout=0.2)) ...
0
votes
0answers
9 views

reuse pre-trained embedding model in a new model only to evaluate - keras

I am looking for a way to reuse a trained embedding function in a new model, but only to evaluate. For instance, def embedding(x): return Dense(2000)(x) input = Input((4,200)) embedded = [] ...
-1
votes
0answers
18 views

No such file or directory: train-scene classification'/train

I am doing image classification in google labs, I have files like this. ls/content/'train-scene classification'/ test.csv train/ train.csv I trying to improve my accuracy, so I have applied ...
0
votes
0answers
10 views

Keras Conv1D on DNA sequence

I would like to apply a 1D convolution on fixed size DNA sequence using keras. Dna sequence is 45 bases long. Each sequence has been one-hot encoded. There is one filter with kernel size = 3. See ...
0
votes
0answers
51 views

What is the meaning of ValueError in Keras? - 'Tensor conversion requested dtype complex64 for Tensor with dtype float32'

I have the following part of code - ... model.add(Generic_act_func()) print('layer 1 output') print(model.layers[1].output) #Here layers[1] is Generic_act_func() layer specified above model.add(...
-1
votes
2answers
68 views
+100

Custom pooling layer - minmax pooling - Keras - Tensorflow

I want to define my custom pooling layer, instead of returning the max values like the MaxPooling layer, it would output the k maximum values and the k minimum values. I am using Tensorflow as the ...
1
vote
2answers
21 views

Stacking functional models with python functions induce graph disconnection in Keras

I want to connect the model_1: x_a = Input((5,)) hidden_a = Dense(2)(x_a) hidden_b = Dense(2)(x_a) model_1 = Model(x_a, [hidden_a, hidden_b]) and model_2: x_b = Input((2,)) output = Dense(1)(x_b) ...
0
votes
0answers
31 views

Keras Error: Variable does not exist, or was not created with tf.get_variable()

import tensorflow.keras as ks def scope_error_test(): input_holder = tf.placeholder(dtype=tf.float32, shape=(None, 368, 368, 3), ...
1
vote
1answer
23 views

Error when checking target: expected concatenate_1 to have shape (1,) but got array with shape (851,)

I have some dimentional issue with a keras concatenation. It seems that the output array of the model (None, 851) is different from the dimention requiered in the error message. Here what I get: ...
0
votes
0answers
17 views

tf.Keras - resetting the sequential input shape into custom layer

I've build a custom layer that I would like to train my data over 6 layers. The first input into my layer works just fine, but the input to the second layer doesn't. This is because of the _features = ...
0
votes
1answer
23 views

Error when checking target: expected conv2d_29 to have 4 dimensions, but got array with shape (1255, 12)

I would like to train a deep learning model, where input image shape is (224,224,3) . And I would like to feed them into a u-net model. After training I get the error : Error when checking target: ...
0
votes
2answers
37 views

Keras/Tensorflow Input to RNN layers

I'm trying to build a RNN in Keras. I dont quite understand the required input format. I can build dense networks no problem, but I think that the RNN layers expect input dimension x batch x time step?...
0
votes
1answer
16 views

Need of specifying the input shape in Keras

I am seeking help from put the right form of input_shape, Let's say if I have the code like: G = ImageGenerator(batch_size,args,****kwargs) where G is a custom-made data_generator which specify ...
1
vote
1answer
50 views

Keras (Tensorflow) Reshape Layer input error

I have a reshape input error and i don't know why. The requested shape is 1058400, what is (1, 21168) multiplied with the batch size of 50. What I do not understand is the apparent input size of ...
1
vote
0answers
23 views

reading a pickle, IndexError: only integers, slices (`:`), ellipsis (`…`), numpy.newaxis (`None`) and integer or blean arrays are valid indices

train_features=load_photo_features('/home/feature/_0nX-El-ySo_83_93.pkl', train) reading a pickle file , throws an error like "IndexError: only integers, slices (:), ellipsis (...), numpy.newaxis (...
1
vote
0answers
20 views

how do I scatter a tensor with smaller shape to the bigger tensor?

I have an input tensor with shape (1,4,4,1), I want to concatenate this input with the output of conv layer with shape (1,28,28,1), but for this I want to put each value of the input to the new tensor ...
-2
votes
2answers
40 views

How do you know which layers and how many layers to use? [closed]

I am new to machine learning and have spent some time learning python. I have started to learn TensorFlow and Keras for machine learning and I literally have no clue nor any understanding of the ...
1
vote
0answers
28 views

preparing data with 6 time points in a sequential model: Keras for R

I have a question about pre-processing data in order to enter it into a sequential model in keras for R. I need to be sure that the input is correct for the model. The goal of the sequential model is ...
1
vote
0answers
27 views

Jacobian of decoder in VAE

I am new to both ML and this forum, so please be kind. What I would like to compute is the jacobian of the decoder part of the VAE w.r.t. the latent vectors. I have found the jacobian_batch function ...
0
votes
1answer
18 views

How apply Gridsearch on autoencoder model?

I want to apply GridSearchCV on the autoencoder model. The code of the atuoencoder and GridSearchCV is added below please tell me how I change this code to run GridSearchCV successfully. ...
0
votes
1answer
41 views

Keras 2D Dense Layer for Output

I am playing with a model which should take a 8x8 chess board as input, encoded as a 224x224 grey image, and then output a 64x13 one-hot-encoded logistic regression = probabilities of pieces on the ...
0
votes
0answers
22 views

Unstacking a Keras tensor as a list

I have a keras tensor of size [61,80,?] with the '?' being for batch size in custom layer. When I am trying to unstack this tensor as a list of 2D tensor of size[61,80] using tf.unstack,I am getting ...
-1
votes
0answers
15 views

how to add an inbuilt function in keras layer using lambda function

def dis(bin_image): res=distance_transform_cdt(temp) return res out = Conv2D(1, (1, 1), activation='sigmoid') (c9) outputs = lambda(dis)(c9) this is what i am trying to do but it's saying ...
0
votes
1answer
19 views

Dimensions for Keras RNN encoder-decoder architecture output

I've been unable to figure out the dimensions for an RNN encoder-decoder architecture. I understand how LSTMs work, but I am struggling to implement this one in Keras. After looking at the ...
2
votes
1answer
33 views

Keras' convolution layer on images coming form circular/cyclic domain

The Need Hello, I am experimenting the usage of CNNs on images which come from a cylindric domain, so I am interested to apply the Convolution layer in a circular (or cyclic) way. I mean a ...
0
votes
0answers
63 views

how can I wrap function using lambda layer?

I want to do some processes on one of my layer in autoencoder and then send it to next layer but I can not use the predefined functions in Keras like add ,... and I think I should use lambda with ...
0
votes
1answer
69 views

Tensorflow CuDNNLSTM keras error. TypeError: ('Keyword argument not understood:', 'time_major')

After I save a CuDNNLSTM model in tensorflow 1.13.1, I try to load it back immediately to use it. I get the following error TypeError: ('Keyword argument not understood:', 'time_major'). I looked ...
0
votes
0answers
14 views

is it possible to send the output of layer in CNN to several layer that works in parallel and group convolution node?

I am a beginner in Keras and maybe my question will be simple, but I need your answers. suppose I have a CNN. usually, we send the output of layer as an input to the next layer. now, I want to know is ...
0
votes
0answers
10 views

How do I add the value of cnn layers with special value during learning?

I have an autoencoder in keras and I want to do some processes on the encoder output layer before sending it to the next layer. the output of encoder part is a (1,28,28,1) tensor now I like to ...
0
votes
0answers
34 views

How do I load a keras saved model with custom Optimizer

I have compiled and trained a keras model with a custom optimizer. I saved the model but when I try to load the model, it throws an error stating ValueError: Unknown optimizer: MyOptimizer. I tried to ...
0
votes
0answers
29 views

Unable to stack keras 1D tensors to form 2D keras tensor

In my keras custom layer, i am trying to get the inner product of two same sized matrices and storing them in a 1D array. And then the 1D tensor have to be stacked as a 2D tensor. The code snippet is ...
-1
votes
1answer
36 views

how can i implement Gaussian noise attack in keras?

I try to implement the code of https://arxiv.org/abs/1810.07248 that is about watermarking using deep learning, but I need to apply some attacks like Gaussian noise during learning. I used ...
0
votes
1answer
41 views

Can I use a 3D input on a Keras Dense Layer?

As an exercise I need to use only dense layers to perform text classifications. I want to leverage words embeddings, the issue is that the dataset then is 3D (samples,words of sentence,embedding ...
1
vote
0answers
51 views

Merging Models with Keras' Functional API

I am new to deep learning and am teaching myself some NLP concepts. I am trying to understand this by working on Quora's sentence similarity scoring model as explained in this excellent tutorial here. ...
0
votes
1answer
15 views

what is the point of having a seperate build function in keras.layers

As per the following doc --> https://keras.io/layers/writing-your-own-keras-layers/ It is used for initializing the weight matrix.. but my doubt is why not just do this initialization in the call ...
0
votes
0answers
9 views

is it possible to select part of tensor for loss function?

I have an autoencoder with two inputs, image (28x28) and wtm(28x28). I add wtm with the output of the encoder part and send it to the decoder part. my main wtm was 4x4 but because I do not know how ...
0
votes
0answers
14 views

How can I calculate BER in keras?

I have an autoencoder and it has two outputs and I want to calculate the BER between input and output, but I have some questions. 1. for calculating do we have any function in python for doing it? ...
0
votes
1answer
41 views

nested model in lambda layer is not trainable

I have a model (i.e,spatial_model), which is nested inside the temporal model to construct a CNN-LSTM model. It seems that TimeDistributed layer does not accept two inputs while my spatial model needs ...
0
votes
0answers
21 views

How to fuse an additional feature in training deep neural networks?

I am training a Convolutional Neural Network (CNN) to classify Spectrogram images (frequency over time). These Spectrograms were created from some signals on specific times, therefore the time of ...
1
vote
1answer
351 views

TypeError: The added layer must be an instance of class Layer. Found: <keras.engine.input_layer.InputLayer object at 0x7fc6f1b92240>

I am trying to add vgg16 layers to Sequential model but getting the error mentioned in the Question Title from keras.applications.vgg16 import VGG16 from tensorflow.contrib.keras.api.keras.models ...
0
votes
1answer
68 views

What is the difference between SeparableConv2D and Conv2D layers?

I didn't find a clearly answer to this question online (sorry if it exists). I would like to understand the differences between the two functions (SeparableConv2D and Conv2D), step by step with, for ...
1
vote
0answers
49 views

How to implement low dimension embedding in TensorFlow

Assume I have an hign dimension vector v which dimension is (2000,1). And I want to get an embedding for v as following: f=E*v where E is the embedding matrix with size (30, 2000); now the new ...
0
votes
1answer
83 views

Keras Tokenization (fit on text)

When i am running this script--> tokenizer.fit_on_texts(df['text'].values) sequences = tokenizer.texts_to_sequences(df['text'].values) word_index = tokenizer.word_index print('Found %s unique tokens.'...
0
votes
1answer
61 views

Keras: How to load a model having two outputs and a custom loss function?

I have trained a Keras (with Tensorflow backend) model which has two outputs with a custom loss function. I need help in loading the model from disk using the custom_objects argument. When compiling ...

http://mssss.yulina-kosm.ru