All Questions
Tagged with tensorflow2.x or tensorflow2.0
5,975 questions
0
votes
0
answers
23
views
Appropriate GPU deepface
I am currently using only CPU(intel-13700K) for deepface to run face recognition.
I have achieved 15 images per second on my dataset; however, striving for more.
I am using a combination of ArcFace + ...
- 1
0
votes
0
answers
21
views
FileNotFoundError: [Errno 2] No such file or directory: 'patchelf' while building TensorFlow from source
Description:
I am trying to build TensorFlow 2.18 from source on CentOS 8 with the following environment:
Python Version: 3.10,
Bazel Version: 6.5.0,
CUDA/cuDNN Version: CUDA 12.5 / cuDNN 9.3,
GCC/...
0
votes
1
answer
57
views
Problem with tensorflow non linearity Apple M4 Chips
I have a problem with the result of non-linear(sigmoid) Neural Network Classification in tensorflow. I suspect is a problem with the M chip and my instalation but I tried several versions using ...
- 663
0
votes
1
answer
7
views
Keras metrics are aggregated over training?
Most TF keras models I see are implemented where at the end of a training step they calculate some metric and use metric.update_state(metric_value) to log the metric.
As far as I can understand this ...
1
vote
0
answers
40
views
InvalidArgumentError, Deep Learning with Python, Example 10.2.3
I'm trying example 10.2.3 from the book Deep Learning with Python by François Chollet. I'm getting an error on the line history = model.fit(...) using Tensorflow 2.17. However, I don't get an error if ...
- 10.1k
0
votes
0
answers
7
views
MirroredStrategy output varies depending on the visible GPUs
I'm trying to run the following code
import tensorflow as tf
def test_multi_gpu_communication():
# Check if multiple GPUs are available
gpus = tf.config.list_physical_devices('GPU')
if ...
- 1
0
votes
0
answers
17
views
Can't setup specific parameters of server_core through model_config_list
Im having an infinite loop when model_paths are not correct in model_config_list for tf-serving. This seems to be a known behaviour for tf-serving and it also seems to be two different ways of ...
- 1,278
0
votes
0
answers
14
views
Training Tensorflow from Bigquery table
I need to train a tf model with a huge table on Bigquery.
I try to follow the example in the documentation (https://www.tensorflow.org/io/tutorials/bigquery) but without success.
I can't figure out ...
- 1
1
vote
0
answers
17
views
How can I get model summary from a Object Detection API model?
I am working with https://github.com/tensorflow/models/tree/master/research/object_detection, where I have loaded the following model using pipeline config:
import tensorflow as tf
from ...
- 19
1
vote
1
answer
97
views
TensorFlow 2.14.0 Fails to Detect GPU on Google Colab
I'm trying to use Mask-RCNN with a GPU on Google Colab, as I need a powerful GPU that my local machine lacks. I’ve even subscribed to Google Colab Pro for this purpose. I've implemented a repository ...
- 13
0
votes
0
answers
40
views
Tensorflow serving keras bert model issue
I am trying to use tensorflow serving to serve a keras bert model, but I have problem to predict with rest api, below are informations. Can you please help me to resolve this problem.
predict output (...
- 193
1
vote
0
answers
30
views
ImageDataGenerator sending array of image, instead of file path in preprocessing function
I am trying to implement a Deep learning project in python, that
takes Image as input
detects one human
tells the color of cloths, that human wearing.
I have a folder which have all training images ...
- 21
0
votes
0
answers
38
views
TypeError: '_UserObject' object is not callable
I am posting a question because there was a problem in the process of creating and importing custom models from tensorflow.
This is the code that has a problem:
import os
os.environ['...
- 1
1
vote
1
answer
41
views
what kind of data is written to events.out.tfevents.1730092600.jolteon.215490.0.v2 from keras.callbacks.tensboard callbacks
Would like to know what kind of information is available and in what format is the file events.out.tfevents.1730092600.jolteon.215490.0.v2 written in.
This file is generated by :
tensorboard_callback =...
0
votes
0
answers
41
views
TensorFlow InvalidArgumentError: Concatenation dimension mismatch in ConcatOp - Shapes do not match
I keep getting this error:
---------------------------------------------------------------------------
InvalidArgumentError Traceback (most recent call last)
<ipython-input-50-...
0
votes
0
answers
26
views
Query on reinforcement learning model not converging
I was trying to build a reinforcement model to personalise email campaign (basis email variants, time of day etc.) and improve open rates.
For this, I created a DQN model and followed epsilon decay ...
0
votes
1
answer
27
views
Tensorflow - Training error at the end of each epoch in progbar.py (keras)
After changing laptops, the training of my transformer (a simple translator) fails at the end of the first epoch.
The code is pretty long but I use model.fit() and overridden train_step().
h = ...
- 7,361
0
votes
0
answers
19
views
Tensorflow Lite Dynamic Output shape
i have a tensorflow object detection model. when i load it with
tf.saved_model.load()
i see this signatures:
{'detection_scores': TensorShape([1, 100]), 'detection_boxes': TensorShape([1, 100, 4]), '...
0
votes
1
answer
53
views
Tensorflow executes slow on GPU - retracing issue?
I am trying to develop a transformer sequence to vector model but encounter performance issues.
I am working with a Tesla V100-PCIE-16GB. Whenever the model encounters an unseen sequence length, the (...
- 1
0
votes
0
answers
34
views
tf.math.segment_mean operation not working with XLA_GPU_JIT in google colab
I am using Google Colab pro with a T4 GPU for testing, my model is a GNN that is used to find relational data between chess pieces on a board. When I ran my code with a CPU it ran perfectly fine, but ...
- 23
0
votes
0
answers
35
views
Having issue with calculating SINR using TensorFlow, how numerically stable is TF when it comes to complex numbers?
I have code for training a neural network and the loss is negative SINR, for a system with multi group (G) and multi users (K). However when I calculate the sinr, something odd happens.
the SINR ...
0
votes
0
answers
25
views
Tensorflow probability - quantile is not implemented in Binomial but is documented
quantile() does not seem to be implemented in Binomial distribution but is documented here
To reproduce the issue please run :
import tensorflow_probability as tfp
import tensorflow as tf
tfd = tfp....
- 7,361
0
votes
1
answer
56
views
How do I handle a custom loss function with (1/(1-exp(-x))-1/x) in it?
I am working on a deep learning model with a ragged tensor where the custom loss function is related to:
f(x)+f(x+50)
and f(x)=1/(1-exp(-x))-1/x when x!=0, f(x)=0.5 when x=0.
f(x) is ranged between 0 ...
- 15
0
votes
0
answers
43
views
Is it possible to use Vertex AI exported AutoML Edge model(saved_model.pb stored in GCS ) file as pre-trained model for transfer learning
I trained an AutoML Edge model in Vertex AI using image dataset and I exported the trained model to GCS (saved_model.pb) and now i want to use this model as pre-trained model and train on new dataset ...
- 195
0
votes
0
answers
36
views
The tf.data.Dataset.from_generator() still returns full dataset in its entirety
The full dataset for my CNN model is too large for the 12GB vRAM the GPU has and model.fit() throws OOM error when all of them is loaded. To solve this OOM problem, data generator is implemented as ...
- 10.3k
0
votes
0
answers
30
views
Tensorflow creating CNN for Matrix Completion PYTHON 3.9.19
I have an issue with building a modell to fit my problem. First ill explain what the objective is:
This is just an example of what it looks like, they are points in a picture, and due to distortion ...
- 72
1
vote
0
answers
24
views
AttributeError: Can't get attribute 'TextPreprocessor', when deploying Flask app on Render
The problem is I have a flask app, which predict spam or ham text, by taking input from user.
I have two 'joblib' models with pipelines for text preprocessing and prediction (with tensorflow)
This is ...
- 21
0
votes
0
answers
35
views
A strange behavior about output_shapes in tf.data.Dataset.from_generator
A data generator is defined in a VGG16 transfer learning model. The input shape of image is (224, 224, 3). Here is the generator defined:
def preprocess_image_patches_image_embedding(image_paths, ...
- 10.3k
0
votes
0
answers
97
views
tensorflow Input shape error: Expected shape (None, 3), but input has incompatible shape (None, 3, 3)
After adding tf.dataset generator, the model.fit() starts to throw error about incompatible input shape. The model is a VGG16 transfer learning model with customized dense layer. The input is image of ...
- 10.3k
1
vote
1
answer
28
views
Issues implementing a custom layer in Tensorflow
I copied this code from tensorflow's documentation for implementing a custom layer:
import tensorflow as tf
class MyDenseLayer(tf.keras.layers.Layer):
def __init__(self, num_outputs):
super(...
- 11
0
votes
1
answer
41
views
How To Choose Values From Tensor Using Another Tensor In Tensorflow
I have 2 arrays (Tensors) with the same shape such as:
array1 = [[[0, 1, 0], [1, 0, 0]], [[0, 1, 0], [0, 0, 1]]]
array2 = [[[0.2, 0.7, 0.1], [0.2, 0.4, 0.4]], [[0.2, 0.8, 0.0], [0.5, 0.2, 0.3]]]
I ...
- 33
1
vote
2
answers
142
views
Tensorflow gpu error: Dst tensor is not initialized when model.fit()
When the vgg16 transfer learning model is being fitted. Tensorflow throws error. Every image patch is 224x224 RGB which is about 602KB (with float32. Max batch size is about 5000 by formula vRAM/patch ...
- 10.3k
0
votes
0
answers
66
views
ValueError: Invalid dtype: complex64 when using get_melspectrogram_layer in Kapre
I'm trying to use the get_melspectrogram_layer from the Kapre library in a Keras model. However, I'm encountering the following error:
ValueError: Exception encountered when calling STFT.call().
...
0
votes
0
answers
28
views
Tensorflow Conv1d kernel size for time series
I am really new at this topic, so please bear with me. I have the following problem:
I'm trying to use a CNN model to predict sleep stages with an ecg signal and some other features. The ecg singal is ...
- 1
0
votes
0
answers
66
views
Load a TensorFlow model trained on Windows Native environment to Windows WSL2 (Linux Ubuntu)
I have a TensorFlow functional model, model, that accepts two sets of inputs (time series and regular features) and produces the probabilities for 3 different classes. I trained this model on Windows ...
- 305
0
votes
0
answers
33
views
Donkeycar: donkey ui command raise RecursionError('maximum recursion depth exceeded') problem
I'm following this guide:
https://docs.donkeycar.com/guide/host_pc/setup_ubuntu/
since I have to install donkey ui on my laptop to train a deep learning autopilot.
The problem arise when I try to ...
- 1
-1
votes
1
answer
417
views
Tensorflow is not detecting my gpu even though i installed CUDA and CudNn
I've tried to install TensorFlow on my machine so I can perform Ai model training. I've installed the CUDA and cuDnn models following this tutorial but even then when I tried running
import tensorflow ...
0
votes
0
answers
19
views
TensorFlow - Replacement for normalize_utf8 to split accented characters
In order to split accented characters, I use tensorflow_text.normalize_utf8(x, 'NFKD')
Unfortunately, tensorflow_text is not avalaible on Windows after TF 2.10 in pip.
I am looking for a replacement ...
- 7,361
1
vote
0
answers
33
views
How to create graph for higher order derivative in Tensorflow
My target is to forcing gradients to be statistically indistinguishable across tasks in a multi-task model by exposing them into a discriminator (inspired by Adversarial Learning, training weights to ...
- 11
0
votes
0
answers
28
views
Getting Value Error while running the model prediction when it is deployed to Google Cloud Run
I'm getting the following error while running the model prediction through my streamlit app deployed on Google Cloud Run. But the same code is running perfectly locally and on Google colab with same ...
- 1
0
votes
0
answers
57
views
Cannot take the length of shape with unknown rank when fitting model
I am trying to train a deep audio classification model. When checking the shapes of the testing and training data it is (16, 6245, 257, 1), 16 being the number of items in the batch, 6245 being the ...
0
votes
1
answer
609
views
"ModuleNotFoundError: No module named 'tensorflow.keras' " in jupiter notebook
I want import in file ipynb:
from tensorflow.keras import layers
from tensorflow.keras import losses
My code looks like:
import matplotlib.pyplot as plt
import os
import re
import shutil
import ...
- 11
3
votes
1
answer
142
views
Persistent Issue Importing KerasClassifier
I’ve been facing a persistent issue with importing the KerasClassifier library in my Python environment, and I’m hoping to get some advice on how to resolve it.
I keep encountering an intermittent ...
1
vote
0
answers
24
views
Multi-Input network for image classification using TensorflowDataset gets none values not supported error
I have a dataset of tfrecords and mapped them to a Tensorflow dataset. The dataset structure has a dict structure as follows:
{ "image" : ImageDataTensor, "additional_features" : ...
- 23
0
votes
1
answer
28
views
"tf.summary.image" gives Invalid Argument Error for the standard example
I have recently started Tensorflow by trying to replicate some of the existing repos, and I have been stuck at writing summaries since quite some time now.
Specifically, I have been trying to port the ...
- 522
0
votes
0
answers
30
views
Efficiently finetune a VGG model in Colab
I am trying to fine-tune a VGG16 model for an ECG-image classification task with the resolution of 1700 * 2200
I haven't reshaped and reduced the size of the images because I thought some ...
- 357
0
votes
0
answers
21
views
Large Accuracy Difference Between TensorFlow Sequential and Subclassing API for CNN Model?
I've been working on a CNN model using both the Sequential API and the Subclassing API in TensorFlow. However, I'm seeing a significant difference in accuracy between the two implementations on the ...
0
votes
0
answers
18
views
Why are my tensorflow gradients None when using tf.python.ops.gradients?
I am struggling with tensorflow gradients. I am using a custom second order optimizer where I need to compute hessians. The library I have chosen uses tf.python.ops.gradients instead of the standard ...
- 11
0
votes
0
answers
87
views
How to load keras models into colab
Am tring to load a keras model into my colab environment using API's but on importing the model and unzipping it I keep getting errors that the model can not be read, the format is wrong
# To move my ...
- 1
0
votes
0
answers
13
views
Input empty but .tfrecord has data, so it doesnt get loaded into training model
I am trying to train an object detection model using my own dataset in google colab using the tensorflow object detection API. However, I keep getting this error message when i attempt to train my ...
