Questions tagged [tensorflow]

0 votes
0 replies
FusedBatchNorm Operation in MobileNetV2
In MobileNetV2 model there is an operation named FusedBatchNorm. Can anyone explain how does the FusedBatchNorm operation work? 0 0 Answers...
asked 4 months ago
0 votes
0 replies
Low level API in tensorflow 2.0
I am very confused about the API in Tensorflow 2.0. How can I create a layer using the low-level API in Tensorflow 2.0? Using the high level API...
asked 4 months ago
0 votes
1 replies
Tensorflow: How to use Tensorflow's Dataset API to use TIFF images?
I'm currently using Tensorflow's Dataset API to load and preprocess images for training and testing. Here's a snipped of the code I'm using to do...
asked 4 months ago
0 votes
1 replies
input_tensor and output_tensor for TFLiteConverter.from_session gives TypeError (Tensor objects are only iterable when eager execution is enabled.)
I am trying to make tflite file of a model while using quantization aware training. I'm making and training the model in keras but I've run into...
-1 votes
1 replies
my trained model gives me wrong predict on input images
I'm making a program for arabic characters recognition using CNN on keras, and then i've trained the model trying different architectures eventho...
asked 4 months ago
0 votes
0 replies
Deploy Tensorflow object detection model on Nvidia TX2
I have object detection model trained using Tensorflow's object detection api and pretrained models. In installation process, especially at COCO...
0 votes
0 replies
Using subword information in OOV token from fasttext in word embedding layer (keras/tensorflow)
I have my own Fasttext model and trained with it a keras classification model with a word embedding layer. But, I wonder how I can make use of t...
1 votes
0 replies
How to combine autograph benefits with a training loop?
How can I improve the execution speed of the code below using autograph "magic"? Given code from here: https://www.tensorflow.org/alpha/tutorial...
asked 4 months ago
0 votes
0 replies
How to load other weights in tensorflow-hub.module?
I want to use ELMo in keras and for that i want to use tensorflow-hub, probably similar to (https://github.com/strongio/keras-elmo/blob/master/El...
asked 4 months ago
0 votes
0 replies
Why do I get a segmentation fault when I want to do multiple inferences with Tensorflow's C API?
I need to use Tensorflow's C API for doing inference on a trained neural network in frozen graph format. As I need to do it 34280 times (due to t...
asked 4 months ago
0 votes
0 replies
How to save model with lambda layer which referring to layers from another models?
I'm using tf.cond and tf.keras.layers.Lambda layer, which referring to layers from another models, for switch models by flag. This compound model...
asked 4 months ago
0 votes
1 replies
Is Tensorflow Dataset.from_generator deprecated in tensorflow 2.0 ? It throws tf.py_func deprecation error
When I create a tf dataset from generator and try to run tf2.0 code , It warns me with a deprication message. Code: import tensorflow as tf fr...
1 votes
1 replies
Xavier Initialization in Tensorflow 2
What is the alternative of tf.contrib.layers.xavier_initializer function in Tensorflow 2.0? The tf.contrib.layers.xavier_initializer is belong to...
asked 4 months ago
0 votes
0 replies
Different sizes (input_shape) in mobilenet_v2
Is it possible to use images with different sizes for learning mobilenet_v2, for example 450x610, 300x200, 670x540 at the same time? 0 0 Ans...
asked 4 months ago
0 votes
0 replies
How to elegantly debug nan/inf on gradients in a Tensorflow model with RNN?
Occassionally we may encounter some nan/inf in gradients during backprop on seq2seq Tensorflow models. How could we easily find the cause of such...
asked 4 months ago
0 votes
1 replies
How to kill tensorboard with Tensorflow2 (jupyter, Win)
sorry for the noob question, but how do I kill the Tensorflow PID? It says: Reusing TensorBoard on port 6006 (pid 5128), started 4 days, 18:03:...
asked 4 months ago
0 votes
1 replies
Create a RNN with Tensorflow 2
Im' trying to create a RNN with Python and Tensorflow 2a, but I'm really not sure about what I did... The prediction results are constants. What...
asked 4 months ago
-1 votes
1 replies
Softmax groups in a tensor by a group index from a second tensor
I am looking for a way to softmax parts of a tensor x by their group, which is denoted by a group index y[i], that exists for each element of x....
asked 4 months ago
0 votes
1 replies
Subtract each row of a tensor A by each row of a Tensor B
I want to subtract every row of Tensor A from every row in Tensor B. This is what I want (this does not produce the result): A = tf.constant([[1...
asked 4 months ago
0 votes
1 replies
Words missing from trained word2vec model vocabulary
I am currently working with python where I train a Word2Vec model using sentences that I provide. Then, I save and load the model to get the word...
asked 4 months ago
0 votes
1 replies
Tensorboard bounding box limit in eval of images
image1 image2 When looking at eval results on tensorboard training an object detection model, not all bounding boxes around every object is made....
0 votes
0 replies
'function' object has no attribute 'called'
when am try to export the pb file by the checkpoint of the fine tune model using %run export_inference_graph.py --input_type image_tensor --pipe...
0 votes
0 replies
I want to install Keras on R. How do I resolve this errors?
I want to install Keras on R. How do I resolve proxy errors? I installed the curl and httr packages, but it was useless. I tried every method I...
asked 4 months ago
0 votes
0 replies
OP_REQUIRES failed Invalid argument: transpose expects a vector of size 5. But input(1) is a vector of size 4
I have a TF Estimator model with code as follows : ''' estimator_model = keras.estimator.model_to_estimator(keras_model=model, model_dir=output...
0 votes
1 replies
Deeplab: how to separate segmentation of overlapping objects?
I'm using the tensorflow deeplab to dolphins segmentation, according to the following images. https://user-images.githubusercontent.com/307129/5...