All Questions
52 questions
0
votes
0
answers
39
views
Unable to install and use torch through CUDA and VSCode
I have set up Anaconda environment and downloaded CUDA and CuDNN in it. I also have tensorflow installed via conda (not tensorflow-gpu) along with keras and scikit-learn. Once that is done, I am ...
3
votes
1
answer
1k
views
Hashed cross-product transformation in PyTorch
I want to implement a hashed cross product transformation like the one Keras uses:
>>> layer = keras.layers.HashedCrossing(num_bins=5, output_mode='one_hot')
>>> feat1 = np.array([1, ...
- 706
0
votes
0
answers
2k
views
Converting Tensorflow Tensor to PyTorch Tensor
I have been trying to convert a Tensorflow tensor to a Pytorch tensor. I have turned run eagerly to true. I tried:
keras_array = K.eval(input_layer)
numpy_array = np.array(keras_array)
pytorch_tensor ...
- 7
-1
votes
1
answer
93
views
Library housing CNN shape calculation in a function?
I find myself continually re-implementing the same free function for a convolutional neural network's output shape, given hyperparameters. I am growing tired of re-implementing this function and ...
- 2,571
0
votes
1
answer
615
views
numpy vs pytorch precision
I've a numpy matrix a defined as:
>>> a
>>> array([[ 1.920941165 , 0.9518795607, 1.5358781432],
[-0.2418292026, 0.0851087409, -0.2760766872],
[-0.4161812806, 0....
0
votes
3
answers
113
views
Taking mean of all rows in a numpy matrix grouped by values based on another numpy matrix
I have a matrix A of size NXN with float values and another boolean matrix B of size NXN
For every row, I need to find the mean of all values in A belonging to indices where True is the corresponding ...
-1
votes
1
answer
1k
views
getting AttributeError: 'numpy.ndarray' object has no attribute 'dim' when converting tensorflow code to pytorch
I was translating my TensorFlow code to PyTorch and suddenly faced this error.
What am I doing wrong here?
AttributeError Traceback (most recent call last)
<...
0
votes
1
answer
318
views
Tensor slicing: tensorflow vs pytorch
I was testing this simple slicing operation in TF and PyTorch which should match in both
import tensorflow as tf
import numpy as np
import torch
tf_x = tf.random.uniform((4, 64, 64, 3))
pt_x = torch....
- 4,170
-2
votes
1
answer
910
views
tf.bitcast equivalent in pytorch?
This question is different from tf.cast equivalent in pytorch?.
bitcast do bitwise reinterpretation(like reinterpret_cast in C++) instead of "safe" type conversion.
This operation is useful ...
- 697
0
votes
1
answer
127
views
Gathering entries in a matrix based on a matrix of column indices (tensorflow/numpy)
A little example to demonstrate what I need
I have a question about gathering in tensorflow. Let's say I have a tensor of values (that I care about for some reason):
test1 = tf.round(5*tf.random....
- 448
0
votes
1
answer
603
views
AttributeError: 'Tensor' object has no attribute 'close'
this is the code I am working with:
I searched for a similiar error do i need to change the .close() ?
Can anyone help please?
0
votes
1
answer
422
views
torch dataset error -- 'numpy.int32' is not callable
I'm preparing a set of medical imaging volumes and segmentation masks to be input into a multi-label segmentation neural network for training. I am recieving the following error message when I attempt ...
0
votes
1
answer
749
views
Pytorch/Tensorflow: Compute gradient of Mixture of Gaussians log density
I have a mixture of three Gaussians and would like to compute the gradient of the log-density using Pytorch or Tensorflow. How can I do that?
from numpy import eye, log
from scipy.stats import ...
- 3,581
0
votes
1
answer
214
views
How to load huge time series windows dataset without memory errors?
I want to convert a typical time series dataset of about 1 million lines into 100-item windows with 50% overlap. Note that it's a multivariate one, so for example given 8 features and 1000 windows ...
- 423
0
votes
2
answers
660
views
Match pytorch scatter output in tensorflow
How can I do the same operation in tensorflow?
tensor = np.random.RandomState(42).uniform(size=(2, 4, 2)).astype(np.float32)
tensor = torch.from_numpy(tensor)
index = tensor.max(-1, keepdim=True)[1]
...
- 1,637