This refers to one application's using multiple graphics-processing units, either in traditional (graphical) or general-purpose (GPGPU) applications.
Questions tagged [multi-gpu]
387 questions
0
votes
2 answers
DataParallel multi-gpu RuntimeError: chunk expects at least a 1-dimensional tensor
I am trying to run my model on multiple gpus using DataParallel by setting model = nn.DataParallel(model).cuda(), but everytime getting this error -
RuntimeError: chunk expects at least a 1-dimensional tensor (chunk at
…
Anubhav Garg
- 1
- 1
- 1
0
votes
0 answers
How use multi gpu?
I want to use the tensorflow multi gpu. Currently, there are two GPUs #0, #1, and if you run nvidia-smi and see GPU usage rate, only #0 GPU is used ~35%. I wonder how you can use both 0# and 1# 99%.
This code is added and executed before the…
MKH
- 11
- 4
0
votes
1 answer
tensorflow.GraphDef was modified concurrently during serialization
I use python and tensorflow,
My GPU is Tesla V100, has 4 GPUs, when I set:
os.environ['CUDA_VISIBLE_DEVICES'] = '0'
or
os.environ['CUDA_VISIBLE_DEVICES'] = '0,1'
or
os.environ['CUDA_VISIBLE_DEVICES'] = '0,1,2'
the code can run without any…
zhu
- 1
0
votes
1 answer
Tensorflow / keras multi_gpu_model is not splitted to more than one gpu
I'm encountered the problem, that I can not successfully split my training batches to more than one GPU. If multi_gpu_model from tensorflow.keras.utils is used, tensorflow allocates the full memory on all available (for example 2) gpus, but only the…
johni07
- 761
- 1
- 11
- 31
0
votes
1 answer
Tensorflow Multi-GPU loss
I am studying how to implement multi-GPU training on Tensorflow.
Now I am reading this source as recommended in the documentation.
As far as I understand, at line 178 variable loss accounts the loss for only one GPU (as the comment states). Thus, at…
lorenzop
- 525
- 4
- 19
0
votes
0 answers
Keras anomaly in its training time
I am using Keras in multi-gpu, with Tensorflow backend on 2 gpus. I am using a generator (keras.utils.Sequence) to load my data in batch mode (BS = 64). Therefore I am using the fit_generator class, providing it with my train and validation data and…
shoegazerstella
- 69
- 6
0
votes
1 answer
Pytorch 3-GPUs, just can only use 2 of them to train
I have three 1080TI, but when train I can only use 2 of them..
Code:
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model.cuda()
criterion = nn.CrossEntropyLoss().cuda()
optimizer_conv =…
王康年
- 11
- 2
0
votes
1 answer
Passing input in Tensorflow Mirrored Strategy distirbuted computing
So I am following the example code on tensorflow MirroredStrategy . However, I am getting the following error
raise ValueError('model_fn (%s) must include features argument.' % model_fn)
ValueError: model_fn (
screamingmamba
- 1
- 3
0
votes
0 answers
Pytorch using Multi GPU / accuracy is too low(10%)
if torch.cuda.is_available():
for epoch in range(epoch_num):
for i,(images, labels) in enumerate(trainloader):
images=images.to(device)
labels=labels.to(device)
optimizer.zero_grad()
#Forward Backward…
C. Eunbi
- 11
- 2
0
votes
0 answers
keras fine-tuning a pre-trained model does not change weights when using multi_gpu_model and layers.trainable = True
I'm loading the VGG16 pretrained model, adding a couple of dense layers and fine tuning the last 5 layers of the base VGG16. I'm training my model on mutliple gpus. I saved the model before and after training. The weights are the same inspite of…
Sharanya Arcot Desai
- 685
- 3
- 7
- 18
0
votes
0 answers
Why multi-gpu faster than single gpu in caffe training?
In the same hardware/software env, With the same net and solver, just differ in command line.
While command line is:
caffe-master/build/tools/caffe train --solver=solver_base.prototxt --gpu=6
It tasks about 50 seconds per 100 iters.
While command…
HiYuan
- 63
- 4
0
votes
2 answers
Saved multi-GPU trained model loaded into single-GPU; inconsistent results
I'm seeing strange results when loading a saved model that was trained on multiple GPUs into a single GPU model. I'm operating in a shared environment so I'm doing training on 4 GPUs but running tests using a single GPU.
What I'm seeing is the tests…
Chris Kirby
- 1
- 1
0
votes
0 answers
tf.layers.batch_normalization() gamma does not exist error when trained on multiple GPU
When I train Neural network on multiple GPU the batch normalization is not working, I use tf.layers.batch_normalization(), but when I run, I get following error:
ValueError: Variable batch_normalization_4/gamma does not exist, or was not created…
BryanF
- 111
- 1
- 5
0
votes
0 answers
How to do multiple gpu programming in tensorflow
I am using following code to do parallelization, however, I am not able to do that. It throws some random errors. I am using amazon aws for training. So when I run this training putty stops responding.
codes :
for d in…
user169703
- 66
- 1
- 5
0
votes
2 answers
When to set reuse=True for multi GPU training in tensorflow?
I am trying to train a network with tensorflow with multiple towers. I had set reuse = True for all the towers. But in the cifar10 multi gpu train of tensorflow tutorials, the reuse variable has set after the first tower was created:
with…
Mehraban
- 3,164
- 4
- 37
- 60