benchmarks
benchmarks copied to clipboard
tensorflow
A benchmark framework for Tensorflow
After taking the latest benchmarks, we noticed a drop in performance on models inception3 and resnet152. Testing with TensorFlow r1.5 on 32xP100 GPUs (8 servers), imagenet data, batch size 64....
On slow filesystems, calling glob twice can have a significant performance penalty. Only one call is necessary.
The environment： centos7.3 cuda8.0 cudnn60 Tesla P40 tensorflow_gpu1.3 ``` [root@Tensorflow tf_cnn_benchmarks]# python /home/benchmarks-tf_benchmark_stage/scripts/tf_cnn_benchmarks.py --num_gpus=8 --batch_size=128 --model=resnet50 --variable_update=parameter_server TensorFlow: 1.3 Model: resnet50 Mode: `training` Batch size: 1024 global 128 per device...
Hi, I'm trying to train resnet56 on CIFAR-10 with the following param. However, each time I start the run, it creates a log file of size 1.2G or 2.4G. Somehow...
@reedwm Hi, I am in trouble during using the following code. ""' for i, (g, v) in enumerate(grads): apply_gradient_op = opt.apply_gradients([(g, v)]) barrier = self.benchmark_cnn.add_sync_queues_and_barrier( 'replicate_variable_%s' % i, [apply_gradient_op]) """...
System information： OS Platform: ubuntu 16.04 TensorFlow : install from source Python version: Python 2.7.5 1. Run with the command: `python tf_cnn_benchmarks.py --num_batches 100 --display_every 1 --num_gus 8 --model resnet50...
After I pull and merge the latest commit, I got the `ImportError`. I attached the error log as below: ``` Traceback (most recent call last): File "tf_cnn_benchmarks.py", line 26, in...
Inside `preprocessing.py` you are using inception preprocessing to train images and vgg preprocessing for evaluation, according to slim. It's a little bit confusing. If you want to train a different...
At the time of writing, DenseNet is implemented in this benchmark with what is described as the "naive" implementation in [Memory-Efficient Implementation of DenseNets](http://arxiv.org/abs/1707.06990). This will under perform compared to...
This issue can be taken as a feature-request or a request related to documentation. The high-performance benchmarking example is a good effort. However the code is very fused (combining distributed...
