siakon89

{"lastUpload":"2021-04-11T06:50:47.971Z","extensionVersion":"v3.4.3"}

siakon89

predictor = estimator.deploy(
    initial_instance_count=1, 
    instance_type='ml.p2.xlarge',
    endpoint_name="no-elastic-inference-test"
)

predictor = estimator.deploy(
    initial_instance_count=1, 
    instance_type='ml.c5.large',
    accelerator_type='ml.eia1.medium',

siakon89

estimator = TensorFlow(
    entry_point='model.py',               # Script that has the model
    train_instance_type='ml.p3.2xlarge',  # The instance type for training
    train_instance_count=1,               # The number of instances to be spawned
    model_dir='/opt/ml/model',            # The location for the trained model
    hyperparameters={                     # Hyperparameters for the model
      'epochs': 250, 
      'batch_size': 128, 
      'learning_rate': 0.001
    },

siakon89

from smdebug.trials import create_trial
from smdebug import modes
import numpy as np
import matplotlib.pyplot as plt


# Get the tensors from S3
s3_output_path = estimator.latest_job_debugger_artifacts_path()

# Create a Trial https://github.com/awslabs/sagemaker-debugger/blob/master/docs/analysis.md#Trial

siakon89

from sagemaker.debugger import Rule, rule_configs

estimator = TensorFlow(
    entry_point='model.py',
    train_instance_type=train_instance_type,
    train_instance_count=1,
    model_dir=model_dir,
    hyperparameters=hyperparameters,
    role=sagemaker.get_execution_role(),
    base_job_name='tf-fashion-mnist',

siakon89

debugger_hook_config = DebuggerHookConfig(
    s3_output_path=f"s3://{bucket}/fashion-mnist/hook",
    collection_configs=[
        CollectionConfig(
            name="all_tensors",
            parameters={
                "include_regex": ".*",
                "save_steps": "1, 2, 3"
            }
        )

siakon89

hyperparameters = {'epochs': 10, 'batch_size': 256, 'learning_rate': 0.001}

estimator = TensorFlow(
    entry_point='model.py',
    train_instance_type='ml.p3.2xlarge',
    train_instance_count=1,
    model_dir='/opt/ml/model',
    hyperparameters=hyperparameters,
    role=sagemaker.get_execution_role(),
    base_job_name='tf-fashion-mnist',

siakon89

from sagemaker.debugger import CollectionConfig, DebuggerHookConfig

bucket = sess.default_bucket()

collection_config_biases = CollectionConfig(name='biases')
collection_config_weights = CollectionConfig(name='weights')
collection_config_metrics = CollectionConfig(name='metrics')

debugger_hook_config = DebuggerHookConfig(
    s3_output_path=f"s3://{bucket}/fashion-mnist/hook",

siakon89

sam build
sam package --output-template-file .aws-sam/packaged.yaml --s3-bucket <a bucket to save the template> --profile <your profile name>
sam deploy --template-file .aws-sam/packaged.yaml --stack-name imageTagging --capabilities CAPABILITY_IAM --profile <your profile name>

siakon89

import boto3
from boto3.s3.transfer import TransferConfig

BUCKET = "<your bucket from SAM template>"
REGION = "<your region>"
PROFILE = "<your profile>"


def upload_to_s3(file_name, key, bucket=BUCKET, region=REGION):