getitune-running-inference

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Running inference with getitune

使用getitune运行推理

getitune

runs inference through

engine.predict()

(per-item predictions) and

engine.test()

(metrics on the test subset). The same calls work whether the engine holds a PyTorch model or an exported OpenVINO/ONNX model — the backend is selected from what you pass to

model=

Run everything from

library/

getitune

通过

engine.predict()

（逐样本预测）和

engine.test()

（测试子集指标计算）运行推理。无论引擎加载的是PyTorch模型还是导出的OpenVINO/ONNX模型，这些调用都能正常工作——后端会根据你传入

model=

的内容自动选择。

请在

library/

目录下运行所有命令。

PyTorch inference (trained model)

PyTorch推理（训练后模型）

python

from getitune.engine import create_engine

engine = create_engine(
    model="efficientnet_b0",
    data="/path/to/dataset",
)
test_metrics = engine.test()      # metrics on the test subset
predictions = engine.predict()    # predictions on the test subset

python

from getitune.engine import create_engine

engine = create_engine(
    model="efficientnet_b0",
    data="/path/to/dataset",
)
test_metrics = engine.test()      # 测试子集的指标计算
predictions = engine.predict()    # 测试子集的预测结果

OpenVINO / ONNX inference (exported model)

OpenVINO / ONNX推理（导出后模型）

python

from getitune.engine import create_engine

python

from getitune.engine import create_engine

OpenVINO IR — pass the .xml

OpenVINO IR — 传入.xml文件路径

ov_engine = create_engine(model="/path/to/exported_model.xml", data="/path/to/dataset") ov_engine.test() ov_engine.predict()

ONNX — pass the .onnx

ONNX — 传入.onnx文件路径

onnx_engine = create_engine(model="/path/to/exported_model.onnx", data="/path/to/dataset") onnx_engine.test() onnx_engine.predict()


Passing an `.xml` or `.onnx` path builds an `OVEngine`, which loads the model via
[ModelAPI](https://github.com/open-edge-platform/model_api).

onnx_engine = create_engine(model="/path/to/exported_model.onnx", data="/path/to/dataset") onnx_engine.test() onnx_engine.predict()


传入`.xml`或`.onnx`路径会构建一个`OVEngine`，它将通过[ModelAPI](https://github.com/open-edge-platform/model_api)加载模型。

Workflow

工作流程

Pick the model surface. Use a model name/checkpoint for PyTorch inference, or an exported
```
.xml
```
/
```
.onnx
```
for deployed inference.
- Done when:
```
create_engine(...)
```
  returns the expected engine type.
Point
data=
at a dataset with a test subset (see
```
getitune-preparing-datasets
```
).
- Done when:
```
engine.test()
```
  runs without a data/format error.
Run
test()
for metrics or
predict()
for per-item outputs.
- Done when: metrics are produced, or predictions are returned for each item.
Compare backends when validating an export. PyTorch vs OpenVINO/ONNX metrics should closely match (small numeric drift is expected).
- Done when: exported-model metrics are within tolerance of the PyTorch model.

选择模型形式：使用模型名称/checkpoint进行PyTorch推理，或使用导出的
```
.xml
```
/
```
.onnx
```
文件进行部署式推理。
- 完成标志：
```
create_engine(...)
```
  返回预期的引擎类型。
将
data=
指向包含测试子集的数据集（参考
```
getitune-preparing-datasets
```
）。
- 完成标志：
```
engine.test()
```
  运行时无数据/格式错误。
运行
test()
获取指标或
predict()
获取逐样本输出。
- 完成标志：生成指标，或返回每个样本的预测结果。
验证导出模型时对比不同后端：PyTorch与OpenVINO/ONNX的指标应高度匹配（允许微小数值偏差）。
- 完成标志：导出模型的指标与PyTorch模型的指标在误差容忍范围内。

CLI

命令行界面（CLI）

bash

undefined

bash

undefined

from library/

在library/目录下执行

getitune predict --data_root /path/to/dataset --model efficientnet_b0 getitune test --data_root /path/to/dataset --model /path/to/exported_model.xml

undefined

getitune predict --data_root /path/to/dataset --model efficientnet_b0 getitune test --data_root /path/to/dataset --model /path/to/exported_model.xml

undefined

getitune-running-inference

Original

Translation

Running inference with getitune

使用getitune运行推理

PyTorch inference (trained model)

PyTorch推理（训练后模型）

OpenVINO / ONNX inference (exported model)

OpenVINO / ONNX推理（导出后模型）

OpenVINO IR — pass the .xml

OpenVINO IR — 传入.xml文件路径

ONNX — pass the .onnx

ONNX — 传入.onnx文件路径

Workflow

工作流程

CLI

命令行界面（CLI）

from library/

在library/目录下执行

Related skills

相关技能