nn-Meter
A DNN inference latency prediction toolkit for accurately modeling and predicting the latency on diverse edge devices.
File Explorer
Download Latest Version (.zip)- codeql-analysis.yml
- integration-test.yml
- integration-test_nni_based_torch.yml
- integration-test_onnx_based_torch.yml
- build_kernel_latency_predictor.md
- customize_predictor.md
- overview.md
- prepare_backend.md
- test_fusion_rules.md
- fusion_rule.png
- hw-nas.png
- irgraph.png
- predict_performance.png
- hardware-aware-model-design.md
- input_models.md
- usage.md
- openvino_requirements.txt
- requirements.txt
- requirements_builder.txt
- dataset.md
- kernel.md
- ops.md
- overview.md
- quick_start.md
- build_kernel_latency_predictor.ipynb
- test_fusion_rule.ipynb
- 1.quick_start.ipynb
- 2.nn-Meter_for_latency_prediction.ipynb
- 3.nn-Meter_benchmark_dataset.ipynb
- nn-meter_dataset_for_gnn.ipynb
- nn-meter_predictor_for_bench_dataset.ipynb
- nn-meter_predictor_for_different_model_format.ipynb
- README.md
- mobilenetv3small_0.json
- mobilenetv3small_0.onnx
- mobilenetv3small_0.pb
- __init__.py
- build_tf_models.py
- build_torch_models.py
- generate_testcase.py
- interface.py
- test_fusion_rule.py
- utils.py
- __init__.py
- utils.py
- __init__.py
- frozenpb_patcher.py
- patcher_wrapper.py
- __init__.py
- converters.py
- restart.py
- __init__.py
- openvino_backend.py
- openvino_profiler.py
- vpu.py
- __init__.py
- cpu.py
- gpu.py
- tflite_backend.py
- tflite_profiler.py
- __init__.py
- interface.py
- __init__.py
- conv.csv
- dwconv.csv
- fc.csv
- pooling.csv
- utils.py
- __init__.py
- config_sampler.py
- finegrained_sampler.py
- generator.py
- prior_distribution_sampler.py
- utils.py
- __init__.py
- build_predictor.py
- extract_feature.py
- predictor_lib.py
- utils.py
- __init__.py
- __init__.py
- blocks.py
- operators.py
- utils.py
- __init__.py
- blocks.py
- operators.py
- utils.py
- __init__.py
- interface.py
- __init__.py
- config_manager.py
- nn_meter_builder.py
- utils.py
- backend_openvino_config.yaml
- backend_tflite_config.yaml
- ruletest_config.yaml
- predictorbuild_config.yaml
- predictors.yaml
- __init__.py
- bench_dataset.py
- gnn_dataloader.py
- __init__.py
- frozenpb_converter.py
- frozenpb_parser.py
- protobuf_helper.py
- shape_fetcher.py
- shape_inference.py
- __init__.py
- constants.py
- converter.py
- utils.py
- __init__.py
- converter.py
- opset_map.py
- __init__.py
- utils.py
- __init__.py
- add-relu_fusionunit.json
- bn-relu_fusionunit.json
- channelshuffle_fusionunit.json
- conv-bn-relu_fusionunit.json
- conv-bn_fusionunit.json
- dwconv-bn-relu_fusionunit.json
- elewise_fusionunit.json
- gap_fusionunit.json
- hswish_fusionunit.json
- se_fusionunit.json
- utils.py
- __init__.py
- constants.py
- fusion_aware_graph.py
- ir_tools.py
- match_helper.py
- union_find.py
- __init__.py
- kernel_detector.py
- rule_reader.py
- rule_splitter.py
- __init__.py
- extract_feature.py
- kernel_predictor.py
- predict_by_kernel.py
- utils.py
- __init__.py
- nn_meter_predictor.py
- utils.py
- __init__.py
- builder.py
- interface.py
- predictor.py
- registry.py
- __init__.py
- config_manager.py
- evaluation.py
- graph_tool.py
- import_package.py
- path.py
- pyutils.py
- utils.py
- __init__.py
- __init__.py
- _utils.py
- alexnet.py
- densenet.py
- googlenet.py
- inception.py
- mnasnet.py
- mobilenet.py
- resnet.py
- shufflenetv2.py
- squeezenet.py
- utils.py
- vgg.py
- reference_result.txt
- reference_result_nni_based_torch.txt
- reference_result_onnx_based_torch.txt
- tensorflow_file.pb
- test_latency_predictor.py
- test_latency_predictor_torch.py
- __init__.py
- backend.py
- kernel.py
- operators.py
- testcase.py
- test_fusion_rule_detector.py
- test_module_register.py
- test_nn_modules.py
- test_predictor_builder.py
- README.md
- .flake8
- .gitattributes
- .gitignore
- CODE_OF_CONDUCT.md
- LICENSE
- README.md
- SECURITY.md
- setup.py
- SUPPORT.md
# Installation Guide
1. Get the code
git clone https://github.com/microsoft/nn-Meter
Downloads the entire project code from GitHub to your computer.
cd nn-Meter
Moves into the project folder you just downloaded.
2. Official Install Script
Easy RecommendedPrerequisites
- Python 3 Python is required to use pip.
pip install nn-meter
Installs the package published on PyPI directly β no need to clone the source.
pip install .
Installs the package published on PyPI directly β no need to clone the source.
After installing, open a new terminal and run the program's version command (e.g. --version) to confirm it worked.
Pulled directly from this repo's README.
3. Python
EasyPrerequisites
pip install nn-meter
Installs the package published on PyPI directly β no need to clone the source.
pip install .
Installs the package published on PyPI directly β no need to clone the source.
If it runs without errors and prints output in the terminal, it worked.
Pulled directly from this repo's README.
// repository documentation
Was this content helpful?
(0 ratings)