All Questions
26,813
questions
0
votes
0
answers
18
views
Unable to import theano backend
Im getting this error everytime i execute any jupyter cell importing tensorflow or it dependencies. Take a look at the 'value error' below.
value error: Unable to import backend : theano
I tried ...
0
votes
0
answers
20
views
How can I save a Keras file with custom layer so that it keeps the trained weights
Inspired by https://keras.io/examples/vision/captcha_ocr/ I tried to build a similar CaptchaAI that can solve 6 instead of 5 characters (as in the linked example).
The whole model and training data ...
-1
votes
0
answers
16
views
ResourceExhaustedError: OOM when allocating tensor with shape[17656320,1024] and type float
This question has been asked a couple of times and I tried this, this, this etc, but couldn't solve my problem. I have just 4 images of approximately 1MB which I'm trying to fit into a deep learning ...
0
votes
0
answers
17
views
ValueError: Arguments clone_function and input_tensors are only supported for Sequential models or Functional models
I am using Quantization aware training. I refer to the lstm code on the Internet and want to put QAT into lstm, but I encountered a ValueError.
ValueError Traceback (...
0
votes
0
answers
11
views
TF/Keras why did MaxPooling3D return a tuple of one tensor instead of a singleton tensor in this snippet?
tensorflow 2.15 backend
inp = layers.Input(batch_shape=batch_shape)
print('Input {}'.format(str(inp.shape)))
Input (1, 7, 60, 60, 1)
x = inp
x = layers.Dropout(0.2)(x)
x = layers....
0
votes
0
answers
10
views
How to preprocess observations for a DQN agent in Stable Baselines3?
I'm working on a reinforcement learning project using the Stable Baselines3 library to train a DQN agent on the Breakout-v4 environment from OpenAI Gym. I've written some code to preprocess the ...
0
votes
0
answers
34
views
Processing Frames instead of jpg images for Deep Learning Model
I've been trying to implement live translation with a sign language cnn model i made with Keras and OpenCV and im running into an issue with converting the frames to a format the model is able to ...
0
votes
0
answers
76
views
AttributeError: module 'pydot' has no attribute 'InvocationException'
Problem:
I want to use plot_model from keras.utils to draw a beautiful AI model layers map;
The tool plot_model requires "pydot" and "graphviz";
I installed "pydot" ...
0
votes
0
answers
21
views
Keras CNN model gives different output predictions for the same input
Hi I have trained a CNN model with keras similar to the example model they use on their website but with slightly smaller layers and an extra dropout layer at the end. The model build functions looks ...
-1
votes
1
answer
24
views
How can TruePositive be a decimal number in keras.metrics.TruePositives?
I am trying to train a CNN model on an image dataset, but I am stuck getting decimal values for TruePositives, TrueNegatives, FalsePositives, and FalseNegatives. How can that be possible?
ERROR ...
0
votes
0
answers
17
views
tf.tile in sparse format in Tensorflow
How do I perform following operation in sparse implementation?
multiple = tf.constant([a,b],tf.int32)
T_tiled = tf.tile(T, multiple)
0
votes
0
answers
17
views
Parallel predictions on the same Keras Model
I have a .keras file (which I will use to load my model) and a huge dataset X (~5m rows) which I want to call model.predict(X) on to retrieve my predictions.
Is there any way to utilise ...
-1
votes
0
answers
20
views
ValueError: `inputs` argument cannot be empty. Received: inputs=[] [closed]
I use the latest tensorflow version and get this error
this is the code
import os
os.environ["TF_CPP_MIN_LOG_LEVEL"] = "2"
import tensorflow as tf
from tensorflow import keras
...
-3
votes
0
answers
37
views
How to make predictions from trained keras model without keras
I have a 1D CNN that I created and trained with keras, and I've saved the weights in an h5 file and the architecture in a JSON file. I'd now like to be able to read the architecture and weights and ...
-1
votes
1
answer
34
views
'tuple' object has no attribute 'rank'
This is a FabricNet sample implementation (official code). As it was in the research paper, the input image size is 120x120x3. I tried the below code, but it throws the error 'tuple' object has no ...