</details>

touch .env

GITHUB_USERNAME="CHANGE_THIS_TO_YOUR_USERNAME"# ← CHANGE THIS

Create a repository by following these instructions: [Create a new repository](https://github.com/new) → name it`Made-With-ML` → Toggle`Add a README file` (**very important** as this creates a`main` branch) → Click`Create repository` (scroll down)

git remote set-url origin https://github.com/GITHUB_USERNAME/Made-With-ML.git# <-- CHANGE THIS to your username

"title":"Transfer learning with transformers",

"description":"Using transformers for transfer learning on text classification tasks."

}' http://127.0.0.1:8000/predict

export HOLDOUT_LOC="https://raw.githubusercontent.com/GokuMohandas/Made-With-ML/main/datasets/holdout.csv"

curl -X POST -H"Content-Type: application/json" -d'{

}' http://127.0.0.1:8000/evaluate

While the application is running, we can use it via cURL, Python, etc.:

curl -X POST -H"Content-Type: application/json" -d'{

}' http://127.0.0.1:8000/predict

Once the application is running, we can use it via cURL, Python, etc.:

python3 -m pytest --cov madewithml --cov-report html

load_dotenv()

ROOT_DIR=Path(__file__).parent.parent.absolute()

LOGS_DIR=Path(ROOT_DIR,"logs")

LOGS_DIR.mkdir(parents=True,exist_ok=True)

EFS_DIR=Path(f"/efs/shared_storage/madewithml/{os.environ.get('GITHUB_USERNAME','')}")

MODEL_REGISTRY=Path("/tmp/mlflow")

MODEL_REGISTRY=Path(f"{EFS_DIR}/mlflow")

Path(MODEL_REGISTRY).mkdir(parents=True,exist_ok=True)

MLFLOW_TRACKING_URI="file://"+str(MODEL_REGISTRY.absolute())

mlflow.set_tracking_uri(MLFLOW_TRACKING_URI)

best_checkpoint=get_best_checkpoint(run_id=run_id)

predictor=TorchPredictor.from_checkpoint(best_checkpoint)

preprocessor=predictor.get_preprocessor()

sample_df=pd.DataFrame([{"title":title,"description":description,"tag":"other"}])

parser.add_argument("--run_id",help="run ID to use for serving.")

parser.add_argument("--threshold",type=float,default=0.9,help="threshold for `other` class.")

args=parser.parse_args()

ray.init()

ray.init(runtime_env={"env_vars": {"GITHUB_USERNAME":os.environ["GITHUB_USERNAME"]}})

serve.run(ModelDeployment.bind(run_id=args.run_id,threshold=args.threshold))

frommadewithmlimportdata,models,utils

frommadewithml.configimportMLFLOW_TRACKING_URI,logger

frommadewithml.configimportEFS_DIR,MLFLOW_TRACKING_URI,logger

app=typer.Typer()

)

run_config=RunConfig(

callbacks=[mlflow_callback],

checkpoint_config=checkpoint_config,

)

run_config=RunConfig(callbacks=[mlflow_callback],checkpoint_config=checkpoint_config,storage_path=EFS_DIR)

ds=data.load_data(dataset_loc=dataset_loc,num_samples=train_loop_config["num_samples"])

if__name__=="__main__":# pragma: no cover, application

ifray.is_initialized():

ray.shutdown()

ray.init()

ray.init(runtime_env={"env_vars": {"GITHUB_USERNAME":os.environ["GITHUB_USERNAME"]}})

app()

frommadewithmlimportdata,train,utils

frommadewithml.configimportMLFLOW_TRACKING_URI,logger

frommadewithml.configimportEFS_DIR,MLFLOW_TRACKING_URI,logger

app=typer.Typer()

experiment_name=experiment_name,

save_artifact=True,

)

run_config=RunConfig(

callbacks=[mlflow_callback],

checkpoint_config=checkpoint_config,

)

run_config=RunConfig(callbacks=[mlflow_callback],checkpoint_config=checkpoint_config,storage_path=EFS_DIR)

initial_params=json.loads(initial_params)

if__name__=="__main__":# pragma: no cover, application

ifray.is_initialized():

ray.shutdown()

ray.init()

ray.init(runtime_env={"env_vars": {"GITHUB_USERNAME":os.environ["GITHUB_USERNAME"]}})

app()