github/kata-containers
1
0
Fork 1
mirror of https://github.com/kata-containers/kata-containers.git synced 2025-08-08 11:27:29 +00:00
Code Issues Projects Releases Wiki Activity
stable-3.2
kata-containers/tests/metrics/machine_learning
History
Gabriela Cervantes 92fdaf9719 metrics: Use TensorFlow optimized image 
This PR replaces the ubuntu image for one which has TensorFlow optimized
for kata metrics.

Fixes #7866

Signed-off-by: Gabriela Cervantes <gabriela.cervantes.tellez@intel.com>
(cherry picked from commit 3a427795ea)
2023-09-21 13:39:40 +02:00
..
mobilenet_v1_bfloat16_fp32_dockerfile metrics: General improvements to mobilenet tensorflow test 2023-09-21 13:12:23 +02:00
pytorch_dockerfile tests: Add Pytorch Dockerfile 2023-07-12 16:34:17 +00:00
resnet50_fp32_dockerfile metrics: Use TensorFlow optimized image 2023-09-21 13:39:40 +02:00
resnet50_int8_dockerfile metrics: Use TensorFlow optimized image 2023-09-21 13:39:40 +02:00
tensorflow_nhwc_dockerfile metrics: Rename tensorflow scripts 2023-09-21 13:18:52 +02:00
pytorch.sh metrics: Use function from metrics common in pytorch script 2023-09-21 13:23:58 +02:00
README.md docs: Fix TensorFlow word across the document 2023-09-21 13:21:28 +02:00
tensorflow_mobilenet_v1_bfloat16_fp32.sh metrics: Remove unused variable in tensorflow mobilenet script 2023-09-21 13:22:47 +02:00
tensorflow_nhwc.sh metrics: Remove unused variable in tensorflow nhwc script 2023-09-21 13:25:56 +02:00
tensorflow_resnet50_fp32.sh metrics: Add TensorFlow ResNet50 FP32 benchmark 2023-09-21 13:25:03 +02:00
tensorflow_resnet50_int8.sh metrics: Add Dockerfile for ResNet50 int8 2023-09-21 13:21:13 +02:00

README.md

Kata Containers TensorFlow Metrics

Kata Containers provides a series of performance tests using the TensorFlow reference benchmarks (tf_cnn_benchmarks). The tf_cnn_benchmarks containers TensorFlow implementations of several popular convolutional models https://github.com/tensorflow/benchmarks/tree/master/scripts/tf_cnn_benchmarks.

Currently the TensorFlow benchmark on Kata Containers includes test for the AxelNet and ResNet50 models.

Running the test

Individual tests can be run by hand, for example:

$ cd metrics/machine_learning
$ ./tensorflow.sh 25 60

Kata Containers Pytorch Metrics

Based on a suite of Python high performance computing benchmarks that uses various popular Python HPC libraries using Python https://github.com/dionhaefner/pyhpc-benchmarks.

Running the Pytorch test

Individual tests can be run by hand, for example:

$ cd metrics/machine_learning
$ ./tensorflow.sh 40 100

Kata Containers TensorFlow MobileNet Metrics

MobileNets are small, low-latency, low-power models parameterized to meet the resource constraints of a variety of use cases. They can be built upon for classification, detection, embeddings and segmentation similar to how other popular large scale models, such as Inception, are used. MobileNets can be run efficiently on mobile devices with Tensorflow Lite.

Kata Containers provides a test for running MobileNet V1 inference using Intel-Optimized TensorFlow.

Running the TensorFlow MobileNet test

Individual test can be run by hand, for example:

$ cd metrics/machine_learning
$ ./tensorflow_mobilenet_benchmark.sh 25 60

Kata Containers TensorFlow ResNet50 Metrics

ResNet50 is an image classification model pre-trained on the ImageNet dataset. Kata Containers provides a test for running ResNet50 inference using Intel-Optimized TensorFlow.

Running the TensorFlow ResNet50 test

Individual test can be run by hand, for example:

$ cd metrics/machine_learning
$ ./tensorflow_resnet50_int8.sh 25 60