search
HomeBackend DevelopmentGolangImport cycle in tensorflow protobuf

张量流 Protobuf 中的导入周期

Question content

I am trying to write client code to communicate with tensorflow server. I need golang protobufs compiled for tensorflow and tensorflow_serving. None of this comes easy, I did it through this. Basically, use buf to generate them. This is buf yaml:

version: v1
managed:
  enabled: true
  optimize_for: code_size

  # go
  go_package_prefix:
    default: "some/path"

plugins:
  - plugin: buf.build/protocolbuffers/go
    out: gen/proto/go

Run successfully, but running application log:

 package command-line-arguments
     imports my-package/internal/infer
     imports my-package/internal/infer/tensorflow_serving/apis
     imports my-package/internal/infer/tensorflow/core/protobuf
     imports my-package/internal/infer/tensorflow/compiler/xla/stream_executor
     imports my-package/internal/infer/tensorflow/compiler/xla
     imports my-package/internal/infer/tensorflow/compiler/xla/service
     imports my-package/internal/infer/tensorflow/compiler/xla: import cycle not allowed

Please note that everything under tensorflow and tensorflow_serving is compiled directly from the original repository.

It surprises me that something as widely used as tensorflow should have an import cycle, but maybe it does. How can I solve this problem?


Correct answer


tl;dr

The root cause is that the repository https://www.php.cn/link/1a16abf2a3149fc7cd6083687cce01c2 really didn't organize the original files correctly (or at least didn't make it go-friendly).

The following two files cause an import loop in go (xla->xla/service->xla):

  • tensorflow/compiler/xla/xla.proto

    • import "tensorflow/compiler/xla/service/hlo.proto"
  • tensorflow/compiler/xla/service/hlo.proto

    • import "tensorflow/compiler/xla/xla_data.proto"

Since xla_data.proto does not import any other files, we can move it into its own package to break the import loop. We can do this using buf's override function. This is the final buf.gen.yaml file:

version: v1
managed:
  enabled: true
  go_package_prefix:
    default: example.com/mymodule/internal
  override:
    go_package:
      # move the generated xla_data.pb.go file into package xla/data to break the import cycle.
      tensorflow/compiler/xla/xla_data.proto: 'example.com/mymodule/internal/tensorflow/compiler/xla/data'
plugins:
  - name: go
    out: internal
    opt:
      - module=example.com/mymodule/internal

  - name: go-grpc
    out: internal
    opt:
      - module=example.com/mymodule/internal

Complete settings for compiling tensorflow proto files using buf

This is the final directory structure:

├── buf.gen.yaml
├── buf.work.yaml
├── buf.yaml
├── go.mod
├── go.sum
├── internal
│   ├── tensorflow
│   └── tensorflow_serving
└── testdata
    ├── serving
    └── tensorflow

buf.gen.yaml: See the "tl;dr" section.

buf.work.yaml

version: v1
directories:
  - testdata/serving
  - testdata/tensorflow

buf.yaml

version: v1
breaking:
  use:
    - file
lint:
  use:
    - default

This is my environment:

$ go version
go version go1.20.3 linux/amd64
$ buf --version
1.17.0
$ protoc --version
libprotoc 3.12.4
$ protoc-gen-go --version
protoc-gen-go v1.30.0
$ protoc-gen-go-grpc --version
protoc-gen-go-grpc 1.3.0
$ git version
git version 2.37.2

Now execute the following command in the root directory of this directory:

$ go mod init example.com/mymodule
$ go get google.golang.org/grpc
$ git clone https://www.php.cn/link/1a16abf2a3149fc7cd6083687cce01c2.git testdata/tensorflow
$ git clone https://github.com/tensorflow/serving.git testdata/serving
$ buf generate
$ go build ./...

Comments:

    The
  1. tensorflow repositories are cloned into the testdata directory so that go build will ignore them.
  2. The installation program will generate files in the internal directory. You can modify the buf.gen.yaml files to place them wherever you want.
  3. go build ./... will not report any errors. But I'm not sure if the generated file is valid.

The above is the detailed content of Import cycle in tensorflow protobuf. For more information, please follow other related articles on the PHP Chinese website!

Statement
This article is reproduced at:stackoverflow. If there is any infringement, please contact admin@php.cn delete
Choosing Between Golang and Python: The Right Fit for Your ProjectChoosing Between Golang and Python: The Right Fit for Your ProjectApr 19, 2025 am 12:21 AM

Golangisidealforperformance-criticalapplicationsandconcurrentprogramming,whilePythonexcelsindatascience,rapidprototyping,andversatility.1)Forhigh-performanceneeds,chooseGolangduetoitsefficiencyandconcurrencyfeatures.2)Fordata-drivenprojects,Pythonisp

Golang: Concurrency and Performance in ActionGolang: Concurrency and Performance in ActionApr 19, 2025 am 12:20 AM

Golang achieves efficient concurrency through goroutine and channel: 1.goroutine is a lightweight thread, started with the go keyword; 2.channel is used for secure communication between goroutines to avoid race conditions; 3. The usage example shows basic and advanced usage; 4. Common errors include deadlocks and data competition, which can be detected by gorun-race; 5. Performance optimization suggests reducing the use of channel, reasonably setting the number of goroutines, and using sync.Pool to manage memory.

Golang vs. Python: Which Language Should You Learn?Golang vs. Python: Which Language Should You Learn?Apr 19, 2025 am 12:20 AM

Golang is more suitable for system programming and high concurrency applications, while Python is more suitable for data science and rapid development. 1) Golang is developed by Google, statically typing, emphasizing simplicity and efficiency, and is suitable for high concurrency scenarios. 2) Python is created by Guidovan Rossum, dynamically typed, concise syntax, wide application, suitable for beginners and data processing.

Golang vs. Python: Performance and ScalabilityGolang vs. Python: Performance and ScalabilityApr 19, 2025 am 12:18 AM

Golang is better than Python in terms of performance and scalability. 1) Golang's compilation-type characteristics and efficient concurrency model make it perform well in high concurrency scenarios. 2) Python, as an interpreted language, executes slowly, but can optimize performance through tools such as Cython.

Golang vs. Other Languages: A ComparisonGolang vs. Other Languages: A ComparisonApr 19, 2025 am 12:11 AM

Go language has unique advantages in concurrent programming, performance, learning curve, etc.: 1. Concurrent programming is realized through goroutine and channel, which is lightweight and efficient. 2. The compilation speed is fast and the operation performance is close to that of C language. 3. The grammar is concise, the learning curve is smooth, and the ecosystem is rich.

Golang and Python: Understanding the DifferencesGolang and Python: Understanding the DifferencesApr 18, 2025 am 12:21 AM

The main differences between Golang and Python are concurrency models, type systems, performance and execution speed. 1. Golang uses the CSP model, which is suitable for high concurrent tasks; Python relies on multi-threading and GIL, which is suitable for I/O-intensive tasks. 2. Golang is a static type, and Python is a dynamic type. 3. Golang compiled language execution speed is fast, and Python interpreted language development is fast.

Golang vs. C  : Assessing the Speed DifferenceGolang vs. C : Assessing the Speed DifferenceApr 18, 2025 am 12:20 AM

Golang is usually slower than C, but Golang has more advantages in concurrent programming and development efficiency: 1) Golang's garbage collection and concurrency model makes it perform well in high concurrency scenarios; 2) C obtains higher performance through manual memory management and hardware optimization, but has higher development complexity.

Golang: A Key Language for Cloud Computing and DevOpsGolang: A Key Language for Cloud Computing and DevOpsApr 18, 2025 am 12:18 AM

Golang is widely used in cloud computing and DevOps, and its advantages lie in simplicity, efficiency and concurrent programming capabilities. 1) In cloud computing, Golang efficiently handles concurrent requests through goroutine and channel mechanisms. 2) In DevOps, Golang's fast compilation and cross-platform features make it the first choice for automation tools.

See all articles

Hot AI Tools

Undresser.AI Undress

Undresser.AI Undress

AI-powered app for creating realistic nude photos

AI Clothes Remover

AI Clothes Remover

Online AI tool for removing clothes from photos.

Undress AI Tool

Undress AI Tool

Undress images for free

Clothoff.io

Clothoff.io

AI clothes remover

AI Hentai Generator

AI Hentai Generator

Generate AI Hentai for free.

Hot Tools

MinGW - Minimalist GNU for Windows

MinGW - Minimalist GNU for Windows

This project is in the process of being migrated to osdn.net/projects/mingw, you can continue to follow us there. MinGW: A native Windows port of the GNU Compiler Collection (GCC), freely distributable import libraries and header files for building native Windows applications; includes extensions to the MSVC runtime to support C99 functionality. All MinGW software can run on 64-bit Windows platforms.

SAP NetWeaver Server Adapter for Eclipse

SAP NetWeaver Server Adapter for Eclipse

Integrate Eclipse with SAP NetWeaver application server.

Dreamweaver Mac version

Dreamweaver Mac version

Visual web development tools

EditPlus Chinese cracked version

EditPlus Chinese cracked version

Small size, syntax highlighting, does not support code prompt function

Safe Exam Browser

Safe Exam Browser

Safe Exam Browser is a secure browser environment for taking online exams securely. This software turns any computer into a secure workstation. It controls access to any utility and prevents students from using unauthorized resources.