diff options
author | 2018-07-03 12:06:44 -0700 | |
---|---|---|
committer | 2018-07-03 12:06:44 -0700 | |
commit | 5625ba11cd9fd68e29d23d874d9e73a93aabab8f (patch) | |
tree | b43c209898808ada9759ac6be02993386f90075e /tensorflow/docs_src | |
parent | 6b1d90270bec3d1259b50e84d2f693d675947ff9 (diff) |
Update text to tflite performance page
Diffstat (limited to 'tensorflow/docs_src')
-rw-r--r-- | tensorflow/docs_src/mobile/tflite/performance.md | 34 |
1 files changed, 15 insertions, 19 deletions
diff --git a/tensorflow/docs_src/mobile/tflite/performance.md b/tensorflow/docs_src/mobile/tflite/performance.md index 29b087bea7..fee777f8d0 100644 --- a/tensorflow/docs_src/mobile/tflite/performance.md +++ b/tensorflow/docs_src/mobile/tflite/performance.md @@ -1,27 +1,23 @@ -# Performance Benchmark numbers +# Performance -This document contains the performance benchmark numbers for running a few well +This document lists TensorFlow Lite performance benchmarks when running well known models on some Android and iOS devices. -The benchmark numbers were generated by running the [TFLite benchmark -binary](https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/lite/tools/benchmark) -on Android and running the [iOS benchmark -app](https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/lite/tools/benchmark/ios) -on iOS. +These performance benchmark numbers were generated with the +[Android TFLite benchmark binary](https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/lite/tools/benchmark) +and the [iOS benchmark app](https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/lite/tools/benchmark/ios). -# Android benchmarks +# Android performance benchmarks -When running Android benchmarks, the CPU affinity is set to use big cores on the -device to reduce variance (see -[details](https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/lite/tools/benchmark#reducing-variance-between-runs-on-android)). +For Android benchmarks, the CPU affinity is set to use big cores on the device to +reduce variance (see [details](https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/lite/tools/benchmark#reducing-variance-between-runs-on-android)). -Models are assumed to have been downloaded from the link, unzipped and pushed to -`/data/local/tmp/tflite_models` folder. The benchmark binary is built according -to instructions listed -[here](https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/lite/tools/benchmark#on-android). -and is assumed to have been pushed to `/data/local/tmp`. +It assumes that models were download and unzipped to the +`/data/local/tmp/tflite_models` directory. The benchmark binary is built +using [these instructions](https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/lite/tools/benchmark#on-android) +and assumed in the `/data/local/tmp` directory. -The following command was used to run the benchmark: +To run the benchmark: ``` adb shell taskset ${CPU_MASK} /data/local/tmp/benchmark_model \ @@ -32,7 +28,7 @@ adb shell taskset ${CPU_MASK} /data/local/tmp/benchmark_model \ --use_nnapi=false ``` -where `${GRAPH}` is the name of model and `${CPU_MASK}` is the CPU affinity +Here, `${GRAPH}` is the name of model and `${CPU_MASK}` is the CPU affinity chosen according to the following table: Device | CPU_MASK | @@ -120,7 +116,7 @@ Pixel xl | 0c | # iOS benchmarks -For running iOS benchmarks, the [benchmark +To run iOS benchmarks, the [benchmark app](https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/lite/tools/benchmark/ios) was modified to include the appropriate model and `benchmark_params.json` was modified to set `num_threads` to 1. |