OJ Sales Simulated

from azureml.core.workspace import Workspace
ws = Workspace.from_config()
datastore = ws.get_default_datastore()

from azureml.opendatasets import OjSalesSimulated

Adatok olvasása az Azure Open Datasetsből

# Create a Data Directory in local path
import os

oj_sales_path = "oj_sales_data"

if not os.path.exists(oj_sales_path):
    os.mkdir(oj_sales_path)

# Pull all of the data
oj_sales_files = OjSalesSimulated.get_file_dataset()

# or pull a subset of the data
oj_sales_files = OjSalesSimulated.get_file_dataset(num_files=10)

oj_sales_files.download(oj_sales_path, overwrite=True)

Az egyes adathalmazok feltöltése a Blob Storage-ba

Feltöltjük az adatokat a Blobba, és létrehozzuk a FileDatasetet a csv-fájlok mappájából.

target_path = 'oj_sales_data'

datastore.upload(src_dir = oj_sales_path,
                target_path = target_path,
                overwrite = True, 
                show_progress = True)

A fájladatkészlet létrehozása

Meg kell határoznunk az adatok elérési útját a FileDataset létrehozásához.

from azureml.core.dataset import Dataset

ds_name = 'oj_data'
path_on_datastore = datastore.path(target_path + '/')

input_ds = Dataset.File.from_files(path=path_on_datastore, validate=False)

A fájladatkészlet regisztrálása a munkaterületre

Regisztrálni szeretnénk az adathalmazt a munkaterületünkön, hogy a folyamat bemeneteként hívhassuk az előrejelzéshez.

registered_ds = input_ds.register(ws, ds_name, create_new_version=True)
named_ds = registered_ds.as_named_input(ds_name)

# This is a package in preview.
# You need to pip install azureml-opendatasets in Databricks cluster. https://learn.microsoft.com/azure/data-explorer/connect-from-databricks#install-the-python-library-on-your-azure-databricks-cluster
# Download or mount OJ Sales raw files Azure Machine Learning file datasets.
# This works only for Linux based compute. See https://learn.microsoft.com/azure/machine-learning/service/how-to-create-register-datasets to learn more about datasets.

from azureml.opendatasets import OjSalesSimulated

ojss_file = OjSalesSimulated.get_file_dataset()
ojss_file

ojss_file.to_path()

# Download files to local storage
import os
import tempfile

mount_point = tempfile.mkdtemp()
ojss_file.download(mount_point, overwrite=True)

# Mount files. Useful when training job will run on a remote compute.
import gzip
import struct
import pandas as pd
import numpy as np

# load compressed OJ Sales Simulated gz files and return numpy arrays
def load_data(filename, label=False):
    with gzip.open(filename) as gz:
        gz.read(4)
        n_items = struct.unpack('>I', gz.read(4))
        if not label:
            n_rows = struct.unpack('>I', gz.read(4))[0]
            n_cols = struct.unpack('>I', gz.read(4))[0]
            res = np.frombuffer(gz.read(n_items[0] * n_rows * n_cols), dtype=np.uint8)
            res = res.reshape(n_items[0], n_rows * n_cols)
        else:
            res = np.frombuffer(gz.read(n_items[0]), dtype=np.uint8)
            res = res.reshape(n_items[0], 1)
    return pd.DataFrame(res)

import sys
mount_point = tempfile.mkdtemp()
print(mount_point)
print(os.path.exists(mount_point))
print(os.listdir(mount_point))

if sys.platform == 'linux':
  print("start mounting....")
  with ojss_file.mount(mount_point):
    print(os.listdir(mount_point))  
    train_images_df = load_data(os.path.join(mount_point, 'train-tabular-oj-ubyte.gz'))
    print(train_images_df.info())