Skip to content

Create CLI interface for tf-cloud #283

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 9 commits into
base: master
Choose a base branch
from
Open
Show file tree
Hide file tree
Changes from 1 commit
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions src/python/dependencies.py
Original file line number Diff line number Diff line change
Expand Up @@ -26,6 +26,8 @@ def make_required_install_packages():
"tensorboard>=2.3.0",
"tensorflow>=1.15.0,<3.0",
"tensorflow_datasets<3.1.0",
"colorama==0.4.4",
"typer[all]==0.3.2",
]


Expand Down
76 changes: 76 additions & 0 deletions src/python/tensorflow_cloud/core/cli.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,76 @@
from typing import Optional

import typer

from .run import run
from .run import remote
from . import docker_config as docker_config_module
from .machine_config import COMMON_MACHINE_CONFIGS

app = typer.Typer()

@app.command("remote")
def remote_command():
"""
To know is you code is running remote with TF Cloud.
"""
if remote():
typer.echo("Running remotely")
typer.Exit()

return typer.echo("Running Locally")

@app.command("run", help="Run your code in a remote cloud environment with TF Cloud.")
def run_command(
entry_point: Optional[str] = typer.Argument(..., help="File path to the python file or iPython notebook that contains the TensorFlow code"),
requirements_txt: Optional[str] = typer.Option(None, help="File path to requirements.txt file containing additional pip dependencies if any"),
image_uri: Optional[str] = typer.Option(None, help="Docker image URI for the Docker image being built"),
parent_image: Optional[str] = typer.Option(None, help="Parent Docker image to use. Example value - 'gcr.io/my_gcp_project/deep_learning:v2' If a parent Docker image is not provided here, we will use a [TensorFlow Docker image](https://www.tensorflow.org/install/docker) as the parent image."),
cache_from: Optional[str] = typer.Option(None, help="Docker image URI to be used as a cache when building the new Docker image. This is especially useful if you are iteratively improving your model architecture/training code. If this parameter is not provided, then we will use `image` URI as cache."),
image_build_bucket: Optional[str] = typer.Option(None, help="GCS bucket name to be used for building a Docker image via [Google Cloud Build](https://cloud.google.com/cloud-build/). If it is not specified, then your local Docker daemon will be used for Docker build."),
distribution_strategy: str = typer.Option("auto", help="Tensorflow distribution strategy based on the machine config"),
chief_config: str = typer.Option("auto", help="`MachineConfig` that represents the configuration for the chief worker in a distribution cluster. Choose between (CPU, K80_1X, K80_4X, K80_8X, P100_1X, P100_4X, P4_1X, P4_4X, V100_1X, V100_4X, T4_1X, T4_4X, TPU)"),
worker_config: str = typer.Option("auto", help="`MachineConfig` that represents the configuration for the general workers in a distribution cluster. Choose between (CPU, K80_1X, K80_4X, K80_8X, P100_1X, P100_4X, P4_1X, P4_4X, V100_1X, V100_4X, T4_1X, T4_4X, TPU)"),
worker_count: int = typer.Option(0, help="Represents the number of general workers in a distribution cluster."),
entry_point_args: str = typer.Option(None, help="Command line arguments to pass to the `entry_point` program. Not implemented yet."),
stream_logs: bool = typer.Option(False, help="Boolean flag which when enabled streams logs back from the cloud job."),
job_labels: str = typer.Option(None, help="Labels to organize jobs. You can specify up to 64 key-value pairs in lowercase letters and numbers, where the first character must be lowercase letter. For more details see https://cloud.google.com/ai-platform/training/docs/resource-labels. Not implemented yet.")
):
entry_point_args = None
job_labels = None

docker_config = docker_config_module.DockerConfig(image=image_uri,
parent_image=parent_image,
cache_from=cache_from,
image_build_bucket=image_build_bucket)

if chief_config != "auto":
try:
chief_config = COMMON_MACHINE_CONFIGS[chief_config]
except KeyError:
typer.BadParameter("You need to choose a between (CPU, K80_1X, K80_4X, K80_8X, P100_1X, P100_4X, P4_1X, P4_4X, V100_1X, V100_4X, T4_1X, T4_4X, TPU) options")

if worker_config != "auto":
try:
worker_config = COMMON_MACHINE_CONFIGS[worker_config]
except KeyError:
typer.BadParameter("You need to choose a between (CPU, K80_1X, K80_4X, K80_8X, P100_1X, P100_4X, P4_1X, P4_4X, V100_1X, V100_4X, T4_1X, T4_4X, TPU) options")

info = run(
entry_point=entry_point,
requirements_txt=requirements_txt,
docker_config=docker_config,
distribution_strategy=distribution_strategy,
chief_config=chief_config,
worker_config=worker_config,
worker_count=worker_count,
entry_point_args=entry_point_args,
stream_logs=stream_logs,
job_labels=job_labels
)

typer.echo(f"Job id: {info['job_id']}")
typer.echo(f"Docker image URI: {info['docker_image']}")

if __name__ == '__main__':
app()