benchmarks
benchmarks copied to clipboard
tensorflow
A benchmark framework for Tensorflow
The meaning of the flag "use_datasets" is confusing. Below two flags are flags "use_datasets" and one more which is related to it named "datasets_use_prefetch". flags.DEFINE_boolean('use_datasets', True, 'Enable use of datasets...
Currently, we [regenerate random labels](https://github.com/tensorflow/benchmarks/blob/bbf7790ba3e44561e6ec7544ed10fe161ed1aea4/scripts/tf_cnn_benchmarks/benchmark_cnn.py#L1810) every step. Instead, we should store the labels in a local variable initialized with `tf.random_uniform`, so we don't have to call `tf.random_uniform` every step. Calling...
On PS node, I get ``` E tensorflow/core/framework/dataset.cc:76] The Encode() method is not implemented for datasetvariantwrapper objects``` The docker image tag is `tensorflow/tensorflow:nightly-gpu-py3`
Hi I'm running TensorFlow benchmark on a single Motherboard Pro WS C246-ACE(Processor: i9-i9900K, DRAM: 64GB), just trying out the performance without using GPUs. But the performance is quick low. 2.5~3.5...
I used the command provided in #310. I repetitively obtain the same error (single or multiple GPUs): ```python 2019-02-17 16:42:14.539440: W tensorflow/core/framework/op_kernel.cc:1401] OP_REQUIRES failed at function_ops.cc:47 : Invalid argument: Argument...
The COCOPreprocessor.create_dataset wants a "num_threads" parameter, while COCOPreprocessor.minibatch did not pass "params.datasets_num_private_threads". By default, the dataset will use inter-thread pool for preprocessing, which leads to bad performance when inter-thread-num is...
Hi all, I run `tf_cnn_benchmarks.py` and checkout branch to `cnn_tf_v1.13_compatible`, When the number of gpu is greater than 4, the performance is significantly reduced, and gpu utilization does not exceed...
estimators support `TF_CONFIG` environment [link](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/contrib/learn/python/learn/estimators/run_config.py#L81) We can use TF_CONFIG to specify the cluster spec in distribute tensorflow training. Will we support `TF_CONFIG` environment variable in benchmarks ?
This is regarding the issue https://github.com/tensorflow/benchmarks/issues/313.
When I ran tf_cnn_benchmarks with and without horovod, I got different evaluation sequences. Without horovod: python tf_cnn_benchmarks.py --data_dir ${HOME}/mldl/data/imagenet --model resnet50 --batch_size 128 --num_epochs 2 --eval_during_training_every_n_epochs 1 --num_gpus 6 I...
