benchmarks
benchmarks copied to clipboard
tensorflow
A benchmark framework for Tensorflow
I realize that there are now runtime MKL flags for KMP_AFFINITY, KMP_BLOCKTIME, and KMP_SETTINGS. However, I often set these variables in my environment via `export KMP_BLOCKTIME=0`. Currently, the benchmark script...
I have observed repeatedly that running tf_cnn_benchmarks on one machine vs. running it in distributed replicated mode with the same settings (same global mini-batch size) leads to different convergence. Command...
How benchmark achieves data parallelism in distributed training. It seems that there is no tf.dataset.shard in the code data_utils.py. How to guarantee that different workers get different data at the...
### Envs: - Python: 2.7 - TensorFlow: 1.8.0 - CUDA: 9.0 - CuDNN: 7.0 - benchmarks commit id: fc993da280312ab65210e7e80bb6fa7f7489182e - benchmarks commit date: Wed May 16 16:54:07 2018 - CPU:...
I started a distributed train with 16 worker (4 gpus per worker) and the worker0 appeared to hang after print "Running warm up" . I checked all the worker, the...
https://github.com/tensorflow/benchmarks/search?q=google-cloud&unscoped_q=google-cloud This repository contains a setup.py or requirements.txt file that depends on the `google-cloud` package. This package is deprecated. If you wish to continue to use this package, please pin...
It doesn't work in --summary_verbosity=3 --save_summaries_steps=20 --train_dir=mydir. IMAGES page shows "No image data was found". I think the developer here can help this issue. Thanks very much.
https://github.com/tensorflow/benchmarks/blob/2389369f6b5c9d3241676a728b450e47482966c0/scripts/tf_cnn_benchmarks/benchmark_cnn.py#L1579 why not change global_step update op for #199 ? Make sure only chief worker can add global_step like tf.train.SyncReplicasOptimizer @alsrgv @reedwm
I am trying to store information for"loss" and accuracy for tensorboard for each iteration. However, could not figure out which file to look into for this. Any comments.
