Stack Overflow
Stack Overflow

Questions tagged [multi-gpu]

This refers to one application's using multiple graphics-processing units, either in traditional (graphical) or general-purpose (GPGPU) applications.

387 questions
0
votes
1 answer

Freeze layers with the multi_gpu_model in Keras

I'm trying to fine-tune a modified InceptionV3 model in Keras. I follow the example "Fine-tune InceptionV3 on a new set of classes" on this page. So I first trained the top dense layers that were added to the InceptionV3 base model with the…
chaohuang
  • 3,965
  • 4
  • 27
  • 35
0
votes
0 answers

Data Parallelism for RNN in tensorflow

Recently, I have used tensorflow to develop an NMT system. I tried to train this system on multi-gpus using data-parallelism method to speed up it. I follow the standard data-parallelism way widely used in tensorflow. For example, if we want to run…
user4193910
  • 1
  • 1
0
votes
1 answer

Tensorflow: Multi-GPU training cannot make all GPU running at the same time

I have a machine that has 3x 1080 GPU. Below are the code of the training: dynamic_learning_rate = tf.placeholder(tf.float32, shape=[]) model_version = tf.constant(1, tf.int32) with tf.device('/cpu:0'): with…
Max Shek-wai Chu
  • 21
  • 7
0
votes
0 answers

Multi GPU CIFAR10 example taking more time as the number of GPU is increasing. I am using eight Tesla K80 GPU

I am running Multi GPU CIFAR10 example. I observed that as I am increasing the number of GPU in the example, time taken to train is increasing. nvidia-smi -l 1 command shows the expected utilization and behavior of the GPUs, but the time taken to…
Kumar Deepankar
  • 1
  • 1
0
votes
0 answers

TensorFlow model doesn't use GPU

I'm trying to run the following code to train a neural network: import tensorflow as tf import pickle import numpy as np import nltk from nltk.tokenize import word_tokenize from nltk.stem import WordNetLemmatizer num_gpus = 2 lemmatizer =…
aL_eX
  • 1,453
  • 2
  • 15
  • 30
0
votes
1 answer

Multiprocessing for Python parallelization error - "function' object is not iterable"

We have NVIDIA Tesla K80 GPU accelerator computing in our data center with the following characteristics: Intel(R) Xeon(R) CPU E5-2670 v3 @2.30GHz, 48 CPU processors, 128GB RAM, 12 CPU coresrunning under Linux 64-bit. I am running the following code…
Desta Haileselassie Hagos
  • 23,140
  • 7
  • 48
  • 53
0
votes
1 answer

tensorflow multi gpu tower error: loss = tower_loss(scope) . ValueError: Variable tower_1/loss/xentropy_mean/avg/ does not exist

When I use multi gpu in tensorflow, and Errors came out as follows: Traceback (most recent call last): File "multi_gpu_train.py", line 290, in tf.app.run() File…
Mao Shu
  • 11
  • 1
0
votes
0 answers

tensorflow with multi-gpu and tf.RandomShuffleQueue

I am trying to modify the code of mask rcnn to run it on multi-gpu, based on the sample of cifar10, the most part of code is below One image and ground truth infomation is read from TFRecords file as below image, ih, iw, gt_boxes, gt_masks,…
D. Tony
  • 1
  • 1
0
votes
1 answer

tensorflow slim multi-GPU can't work

Currently I use tensorflow slim to train the model from scrach. If I just follow the instruction here https://github.com/tensorflow/models/tree/master/slim#training-a-model-from-scratch, everything is OK. However, I want to use multi GPU, so I set…
happenzZ
  • 1
  • 2
0
votes
1 answer

Does tensorflow convnet only duplicate model across multiple GPUs?

I am currently running a Tensorflow convnet for image recognition and I am considering of buying new GPUs to enable more complex graphs, batch size, and input dimensions. I have read posts like this that do not recommend using AWS GPU instances to…
Jialer Chew
  • 349
  • 1
  • 10
0
votes
1 answer

Couple 2 different GPU cards for mini batching

I just bought a GTX 1080Ti and I wanted to know if I can use both my old GTX 1070 and GTX 1080Ti in parallel for mini batching with either TensorFlow or PyTorch. My main concern is: Would the GTX 1070 bottleneck the GTX 1080Ti or the power of each…
E-Kami
  • 2,529
  • 5
  • 30
  • 50
0
votes
0 answers

Multiple CUDA program resource Allocation on Multi-GPU Server

So I am currently looking to buy a server for complex numerical computations using CUDA code. In short I am trying to decide if I want to spend the money on having multiple GPUs. I know as of CUDA 4.0, multi-GPU computation using a single CUDA code…
John
  • 1
  • 4
0
votes
1 answer

Tensorflow seems to be using two GPUs but one GPU seems not be doing anything

I just build a system with two GTX 680 GPUs. To test my system I'm running cifar10_multi_gpu_train.py, training CIFAR10 using Tensorflow. Tensorflow creates two Tensorflow devices based on the GPUs (last two lines): $ python…
Visionscaper
  • 3,979
  • 1
  • 23
  • 26
0
votes
1 answer

Tensorflow. One set of GPUs on same machine and same model work well, another gets OOM error

I am using multiple GPUs (num_gpus = 4) for training one model with multiple towers. The model is training well on one set of GPUs: CUDA_VISIBLE_DEVICES = 0,1,2,3 while it gets OOM problem during the first graph evaluation with CUDA_VISIBLE_DEVICES…
Falls
  • 1
0
votes
1 answer

How to use multiple GPUs effectively when training deep networks?

I am using a machine which has 2 GPUs Titan Black to train my deep learning model which has 3 layers (3x3, 3x3 and 5x5). The training runs pretty well but when I watch nvidia-smi (watch every 1 sec), I realized that my program uses only one GPU for…
Tai Christian
  • 654
  • 1
  • 10
  • 21
1 2 3
…
25 26