Transformers documentation

导出为 TFLite

You are viewing main version, which requires installation from source. If you'd like regular pip install, checkout the latest stable version (v4.40.2).
Hugging Face's logo
Join the Hugging Face community

and get access to the augmented documentation experience

to get started

导出为 TFLite

TensorFlow Lite 是一个轻量级框架,用于资源受限的设备上,如手机、嵌入式系统和物联网(IoT)设备,部署机器学习模型。TFLite 旨在在计算能力、内存和功耗有限的设备上优化和高效运行模型。模型以一种特殊的高效可移植格式表示,其文件扩展名为 .tflite

🤗 Optimum 通过 exporters.tflite 模块提供将 🤗 Transformers 模型导出至 TFLite 格式的功能。请参考 🤗 Optimum 文档 以获取支持的模型架构列表。

要将模型导出为 TFLite 格式,请安装所需的依赖项:

pip install optimum[exporters-tf]

请参阅 🤗 Optimum 文档 以查看所有可用参数,或者在命令行中查看帮助:

optimum-cli export tflite --help

运行以下命令,以从 🤗 Hub 导出模型的检查点(checkpoint),以 google-bert/bert-base-uncased 为例:

optimum-cli export tflite --model google-bert/bert-base-uncased --sequence_length 128 bert_tflite/

你应该能在日志中看到导出进度以及生成的 model.tflite 文件的保存位置,如下所示:

Validating TFLite model...
	-[✓] TFLite model output names match reference model (logits)
	- Validating TFLite Model output "logits":
		-[✓] (1, 128, 30522) matches (1, 128, 30522)
		-[x] values not close enough, max diff: 5.817413330078125e-05 (atol: 1e-05)
The TensorFlow Lite export succeeded with the warning: The maximum absolute difference between the output of the reference model and the TFLite exported model is not within the set tolerance 1e-05:
- logits: max diff = 5.817413330078125e-05.
 The exported model was saved at: bert_tflite

上面的示例说明了从 🤗 Hub 导出检查点的过程。导出本地模型时,首先需要确保将模型的权重和分词器文件保存在同一目录(local_path)中。在使用 CLI(命令行)时,将 local_path 传递给 model 参数,而不是 🤗 Hub 上的检查点名称。

< > Update on GitHub