Questions tagged [tensorflow]

TensorFlow is an open source library for machine learning and machine intelligence. TensorFlow uses data flow graphs with tensors flowing along edges. For details, see https://www.tensorflow.org. TensorFlow is released under an Apache 2.0 License.

Filter by
Sorted by
Tagged with
0
votes
2answers
27 views

How can I use different loss functions for images with different labels in tensorflow?

I have two types of images, meaning two different sets of possible labels. How can I use two different loss functions based on which image I am currently training on with a batch size of 1? Thanks !
0
votes
1answer
202 views

How to get weightage of each factor for a prediction in a neural network classifier?

I am working on a model to predict which employee is going to resign from a firm. The dataset has columns like Date of Birth, Date of Joining, Department, Gender, Marital Status, Years at company etc. ...
0
votes
0answers
31 views

Feature engineering keywords for keras network

I am building a multi-label classifier, using the keras sequential API. My goal is to take an input of keywords 'cat', 'dog', 'rabbit' and apply a custom label. My ...
0
votes
1answer
30 views

Intermittent type error when running CNN inside Docker container

I've been working on a simple computer vision API with a few endpoints for grabbing useful information from ebay images. The API lives in a docker container that looks for the h5 files, downloads ...
0
votes
1answer
755 views

Loss and val loss too big

I have a random vector with known pdf x and i'm trying to predict some behaviour y. I created a dataset containing (X,y), where X is a vector of D realizations of the VA X and y = (y1,y2). That is, ...
0
votes
1answer
123 views

How is the loss between actual and predicted labels being calculated?

Quoting from this example of Predicting movie review with BERT and Tensorflow, ...
0
votes
1answer
2k views

How do I perform weighted loss in multiple outputs on a same model in Tensorflow?

How do I perform weighted loss in multiple outputs on a same model in Tensorflow? This means I am using a model that is intended to have 3 outputs. I did this ...
0
votes
0answers
213 views

An epoch step is taking too much time on GPU!

I am trying to train a CNN on phoneme recognition. My dataset is composed up of 50 Million sample, each sample has 120 feature (signal parameters). My input's shape is (11,120) and the target is a ...
0
votes
0answers
115 views

Keras: Struggle with loss-functions w/ additional parameters and merging them

I'm quite new to the field of data-science and fiddle around with some time-series-data, in my case stock-data. I want to learn a neural-net to forecast them, as everyone has tried to do some time I ...
0
votes
0answers
70 views

Why the accuracy is high on both training and validation set but very low on test set?

I'm using Tensorflow to train a classifier for image recognition, the model below is built via Keras. The original data is (50000, 3072), and reduced to (50000, 100) with PCA. The explained ratio is ...
0
votes
0answers
29 views

I can not find a pre-trained neural network model for object detection

A simple model is needed, but with good accuracy, mAP for detecting objects is preferably trained on COCO, preferably on ipynb (to run or train on google colab). The models from the Tensorflow (...
0
votes
0answers
33 views

How to change the neural network model to improve the accuracy of mAP without learning from 0, but retraining already trained model

how you can change the Faster R-CNN Resnet 101 or Faster R-CNN Inception Resnet V2 model to improve the accuracy of mAP detection without learning the model with 0, but retraining already trained ...
0
votes
0answers
115 views

conv2d output dimensions

the following code block applies 2 kernels of width-height (2, 2), stride 1, and same padding to x. I am struggling to get ...
0
votes
0answers
63 views

Tensorflow using float64 tensor

I have a feed-forward neural network with a customized cost function. Since my cost function has an exponential component, I need to handle very large numbers. By default, tensorflow uses float32 ...
0
votes
0answers
83 views

Price Prediction via Neural Network (Keras)

I am trying to do call price prediction my data set looks something like this ...
0
votes
1answer
29 views

How to load pre-processed .txt data and use it for training convolutional networks?

I have a lot of '.txt' (pre-processed) files related to C3D and I need to use this to train and test while using CNN model. How does one go about this? Most of data in each .txt files are formatted ...
0
votes
0answers
100 views
0
votes
0answers
111 views

Neural network with different input shapes

I'm currently designing the architecture of a neural network for the colorization of grayscale images. Later on it should be able to colorize images with different sizes and different aspect ratios. I ...
0
votes
0answers
206 views

Issues with using stateful in StackedRNN cells

I am having issues with using the 'stateful' feature when building stacked RNN using LSTMCell object. I am following the instructions on tensorflow on how to set 'stateful = True' by passing the '...
0
votes
0answers
988 views

How to map RGB image segmentation ground truth to classes/one-hot vectors in TensorFlow?

Is there a TensorFlow function which takes multicategorical ground truth for semantic image segmentations in the form of RGB images and outputs a tensor with a one-hot vector encoding of the ...
0
votes
1answer
623 views

Tensorflow/Keras, How to convert tf.feature_column into input tensors?

I have the following code to average embeddings for list of item-ids. (Embedding is trained on review_meta_id_input, and used as look up for pirors_input and for getting average embedding) ...
0
votes
0answers
218 views

Change one node type in Tensorflow graph

So I have a tensorflow graph saved in a .pb file, it works well on my machine, but when I run it on tensorflow v 1.3 it loads fine though it crashes with the ...
0
votes
1answer
163 views

how can LSTM output ever be negative with 0 initialization?

From the lstm equations, e.g. as they appear on p406 of the Deep Learning Book, it looks to me like initializing with zeros (as is common practice), must always produce a strictly positive output. ...
0
votes
1answer
2k views

Should I connect my two GPUs with SLI or not? (for Keras + TensorFlow)

I have two GPUs, NVIDIA GTX 1070 Ti. For using Keras with TensorFlow back-end, should I connect them with SLI or not? If not, then they will be treated separately, and one model will be trained on one ...
0
votes
1answer
4k views

Keras exception: Error when checking input: expected dense_input to have shape (2,) but got array with shape (1,)

I have an understanding of this error, it means that the input that I'm passing to the model is of a different dimension that what was expected. The error also states that the input that I'm passing ...
0
votes
1answer
392 views

Neural Network Sensordata as Input

I have a dataset consisting of sensor recordings about human movement. There are 22 classes of different movement like sitting or walking and 19 sensor values. Each recording of a movement has about ...
0
votes
1answer
64 views

Using text classification for system calls

I'm working on a project in which I should classify System calls sequences, my dataset is represented as sequences of integers (from 1 to 340). To do the classification I have inspired from Text ...
0
votes
1answer
116 views

does tensorflow use opencv to covert image to numpy array

Does tensorflow use opencv to covert image to numpy array ? How is feature extraction done in tensorflow ? How is it different from opencv model ?
0
votes
1answer
165 views

How to define a multi-dimensional neural network with keras

I have implemented a simple neural network with keras that takes an input of 50 values and returns a classification of '0' or '1'. I believe the model is expecting an input shape of (50, 1). I'd ...
0
votes
1answer
648 views

How to properly rotate image and labels for semantic segmentation data augmentation in Tensorflow?

What's a proper procedure for doing the image and label rotation for semantic segmentation in dataset augmentation using Tensorflow? Images I have seen the function tf.contib.image.rotate(), but ...
0
votes
1answer
85 views

Output of classifier.predict Tensorflow extract probabiltity

When I do a prediction with my DNN clasifier I get a dictionary like this. ...
0
votes
1answer
24 views

Restricting a weight vector (optimization parameter) to be in a certain domain using python ML library linear regression model

Sorry if the title is a bit long, but basically I'm trying to predict values $$ \hat{y}_i \in [-1,1]$$ using a simple model i.e. something like OLS or ridge regression, I'd like to know if anyone ...
0
votes
2answers
53 views

Metrics values are equal while training and testing a model

I'm working on a neural network model with python using Keras with TensorFlow backend. Dataset contains two sequences with a result which can be 1 or 0 and positives to negatives ratio in dataset is 1 ...
0
votes
1answer
513 views

How to increase accuracy of model from tensorflow model zoo?

Situation: My dataset is 70k images of people wearing clothes. Images are labeled: bbox position and class. There are 10 classes. I did 80:20 split. Categories are balanced with exception of one ...
0
votes
1answer
69 views

Why can distributed deep learning provide higher accuracy (lower error) than non-distributed one with the following cases?

Based on some papers which I read, distributed deep learning can provide faster training time. In addition, it also provides better accuracy or lower prediction error. What are the reasons? Question ...
0
votes
1answer
711 views

Can I use Tensorboard to manage-fjobs and visualize learning on non-Tensorflow algoithrms? (e.g. Scikit?)

I am still searching for a great tool that manages jobs and visualizes learning from my models. Tensorboard is obviously one option given it's massive support. But is it possible to organize jobs ...
0
votes
1answer
882 views

Tensorflow: Unable to Save checkpoint after every 2 global steps during training the SSD model object detection

python models/object_detection/train.py --logtostderr --train_dir=training/ --pipeline_config_path=training/ssd_inception_v2_coco.config INFO:tensorflow:Restoring parameters from /home/rahul/...
0
votes
1answer
1k views

Complex-Valued input to CNN [closed]

I want to train a CNN. However, my input is images of size 100*100 with complex numbers. I have runned the model, but it failed and the loss didn't decrease. Then I found out that my because my inputs ...
0
votes
1answer
2k views

how to use tensorflow graphs in multithread?valueerror:tensor a must be from the same graph as tensor b!

I am doing instance detect and image retrieval task by Keras and Tensorflow as backend. I plan to use multi thread to load two model, I load maskrcnn in a thread and load mobile net in another one. ...
0
votes
2answers
503 views

Should the cost function be zero using TensorFlow's sigmoid_cross_entropy_with_logits?

I'm building a CNN to make a binary classification (1 or zero). For this, I'm using the cost function sigmoid_cross_entropy_with_logits. But for some reason, the ...
0
votes
1answer
396 views

Invalid Argument Error when running simple Convolutional Neural Network [closed]

I am running a convolutional neural network with CSV files as training and test input. I am getting a strange error that I cannot solve. ...
0
votes
1answer
518 views

Accessing and Multiplying Individual Elements of a Layer's Output in Keras

Problem Description: I am trying to access the individual elements (i.e., scalar) from a softmax layer's output (with dimension (,2)) and multiply this with a tensor from another model, which has a ...
0
votes
2answers
152 views

How does backpropagation differ from reverse-mode autodiff

Going through this book, I am familiar with the following: For each training instance the backpropagation algorithm first makes a prediction (forward pass), measures the error, then goes through ...
0
votes
1answer
2k views

CUDA_ERROR_OUT_OF_MEMORY

I have a large network that is somewhat similar to Wavenet. Although (it seems) that my GPU has enough memory, I get an out of memory error on fitting (see logs below). Any idea? How can I ...
0
votes
1answer
439 views

Adding document vector doc2ec to seq2seq model

In general we use word2vec for word embedding in seq2seq model, is it possible to add the document vector from Doc2vec with the input words , I mean using the tag of the document as a word and its ...
0
votes
1answer
590 views

Does Tensorflow uses vectorization in its operators [closed]

I'm kinda new to tensorflow and just wanted to know if it already performs vectorization in its operators like multiplying matrices and so on. And if not if its doable. As an example let's say we ...
0
votes
1answer
200 views

How to use InceptionV3 without retraining

I know I can retrain Inception to label images, but can I just provide an image to an untrained Inception and get a label back? Is there an API for this? If there is, then why does Google open source ...
0
votes
1answer
1k views

Tensorflow skewed cost function

I am writing a TensorFlow program which is trying to categorize a heavily skewed dataset between two categories. One category is represented at 30x the rate of the other. ...
-1
votes
1answer
4k views

how can I used tf.argmax in multi-dimensions tensor?

I want to know who can I use the tf.argmax in multi-dimension tensor in the Tensorflow using convolution neural network. bellow is simple example to explain what I want exactly:: 1-D tensor example::...
-1
votes
1answer
86 views

Fine-tuning a Pre-trained model (Resnet50) do I need to validate it or just train it?

When fine-tuning a Resnet50 model should I do the standard train validation split and train like any normal CNN or should I just do the training and not the validation if I am going to use this model ...