English | 简体中文
Before running demo, the following two steps need to be confirmed:
-
- Hardware and software environment meets the requirements. Please refer to Environment requirements for FastDeploy.
-
- Install FastDeploy Python whl package, please refer to FastDeploy Python Installation.
This document shows an inference example on the CPU using the PaddleClas classification model MobileNetV2 as an example.
import fastdeploy as fd
model_url = "https://bj.bcebos.com/fastdeploy/models/mobilenetv2.tgz"
fd.download_and_decompress(model_url, path=".")option = fd.RuntimeOption()
option.set_model_path("mobilenetv2/inference.pdmodel",
"mobilenetv2/inference.pdiparams")
# **** CPU Configuration ****
option.use_cpu()
option.use_ort_backend()
option.set_cpu_thread_num(12)
# Initialise runtime
runtime = fd.Runtime(option)
# Get model input name
input_name = runtime.get_input_info(0).name
# Constructing random data for inference
results = runtime.infer({
input_name: np.random.rand(1, 3, 224, 224).astype("float32")
})
print(results[0].shape)When loading is complete, you will get the following output information indicating the initialized backend and the hardware devices.
[INFO] fastdeploy/fastdeploy_runtime.cc(283)::Init Runtime initialized with Backend::OrtBackend in device Device::CPU.