RoaldSchuring

import argparse
import pandas as pd
import os
from sklearn.externals import joblib
from sklearn.neighbors import NearestNeighbors
import numpy as np
import subprocess
import sys

def install(package):

RoaldSchuring

def predict_fn(input_data, model):
    input_data_reshaped = input_data.reshape(1, -1)
    distance, indice = model.kneighbors(input_data_reshaped, 11)
    distance_list = distance[0].tolist()[1:]
    indice_list = indice[0].tolist()[1:]
    nearest_neighbors = [distance_list, indice_list]
    print('predict_fn output is', nearest_neighbors)
    return np.array(nearest_neighbors)

RoaldSchuring

from sagemaker_containers.beta.framework import worker, encoders
from six import BytesIO

def _npy_dumps(data):
    # Serializes a numpy array into a stream of npy-formatted bytes.
    buffer = BytesIO()
    np.save(buffer, data)
    return buffer.getvalue()

def output_fn(prediction_output, accept):

RoaldSchuring

def model_fn(model_dir):
    clf = joblib.load(os.path.join(model_dir, "model.joblib"))
    return clf

RoaldSchuring

from sagemaker.sklearn.estimator import SKLearn

script_path = 'sklearn_nearest_neighbors.py'
sess = sagemaker.Session()

# run the Scikit-Learn script
sklearn = SKLearn(
    entry_point=script_path,
    train_instance_type="ml.m5.large",
    role=role,

RoaldSchuring

recommendations = predictor.predict(sample_vector)
print(recommendations)

RoaldSchuring

RoaldSchuring

RoaldSchuring

train_data = 's3://{}/wine-corpus.txt'.format(bucket)
s3_output_location = 's3://{}/output'.format(bucket)

region_name = boto3.Session().region_name
container = sagemaker.amazon.amazon_estimator.get_image_uri(region_name, "blazingtext", "latest")
print('Using SageMaker BlazingText container: {} ({})'.format(container, region_name))

RoaldSchuring

sess = sagemaker.Session()

# define the specifications of the sagemaker training instance
bt_model = sagemaker.estimator.Estimator(container,
                                         role, 
                                         train_instance_count=2, 
                                         train_instance_type='ml.c4.2xlarge',
                                         train_volume_size = 5,
                                         train_max_run = 360000,
                                         input_mode= 'File',