def get_clean_data():
ds = TabularDatasetFactory.from_delimited_files(path=_DATA_URL)
return clean_data(ds)
from azureml.data.dataset_factory import TabularDatasetFactory
from azureml.core.run import Run
from sklearn.ensemble import RandomForestRegressor
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import OneHotEncoder
from sklearn.metrics import mean_absolute_error
import argparse
import os
import numpy as np
import joblib
import pandas as pd
# Imports dataset
mbdataset = TabularDatasetFactory.from_delimited_files("https://raw.githubusercontent.com/czofficial/nd00333-capstone/4a6a4924bdd4c6a188aeb24e0c282bae11c8933b/mercedes.csv")
df = mbdataset.to_pandas_dataframe()
# Cleans dataset
def clean_data(df):
x_df = df
x_df = pd.get_dummies(df, columns=['model', 'transmission', 'fuelType'])
y_df = x_df.pop("price")
return x_df,y_df
x, y = clean_data(df)
# Splits dataset into train and test
import numpy as np
from sklearn.metrics import mean_squared_error
import joblib
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import OneHotEncoder
import pandas as pd
from azureml.core.run import Run
from azureml.data.dataset_factory import TabularDatasetFactory
# TODO: Create TabularDataset using TabularDatasetFactory
# Data is located at:
#
url = "https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv"
ds = TabularDatasetFactory.from_delimited_files(url)
# TODO: Split data into train and test sets.
### YOUR CODE HERE ###a
run = Run.get_context()
def clean_data(data):
# Dict for cleaning data
months = {"jan":1, "feb":2, "mar":3, "apr":4, "may":5, "jun":6, "jul":7, "aug":8, "sep":9, "oct":10, "nov":11, "dec":12}
weekdays = {"mon":1, "tue":2, "wed":3, "thu":4, "fri":5, "sat":6, "sun":7}
# Clean and one hot encode data
def retrieve_cleaned_data():
ds = TabularDatasetFactory.from_delimited_files(path="https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv")
x, y = clean_data(ds)
return train_test_split(x, y, test_size = 0.3, random_state = 0)
def main():
# Add arguments to script
parser = argparse.ArgumentParser()
parser.add_argument('--learning_rate',
type=float,
default=0.03,
help="Learning rate param for lgbm")
parser.add_argument('--max_depth',
type=int,
default=10,
help="Limit the tree depth explicitly")
parser.add_argument('--num_leaves',
type=int,
default=255,
help="Control the complexity of the tree model")
parser.add_argument('--min_data_in_leaf',
type=int,
default=3,
help="Large value can avoid growing too deep a tree")
parser.add_argument('--num_iterations',
type=int,
default=500,
help="Number of boosting iterations")
args = parser.parse_args()
run.log("learning-rate:", np.float(args.learning_rate))
run.log("max_depth:", np.int(args.max_depth))
run.log("num_leaves", np.int(args.num_leaves))
run.log("min_data_in_leaf", np.int(args.min_data_in_leaf))
run.log("num_iterations", np.int(args.num_iterations))
factory = TabularDatasetFactory()
train_data_path = "https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv"
valid_data_path = "https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_validate.csv"
train_ds = factory.from_delimited_files(train_data_path)
valid_ds = factory.from_delimited_files(valid_data_path)
X_train, y_train = clean_data(train_ds)
X_valid, y_valid = clean_data(valid_ds)
d_train = lgbm.Dataset(X_train, label=y_train)
lgbm_params = {}
lgbm_params['learning_rate'] = args.learning_rate
lgbm_params['boosting_type'] = 'gbdt'
lgbm_params['objective'] = 'binary'
lgbm_params['metric'] = 'binary_logloss'
lgbm_params['max_depth'] = args.max_depth
lgbm_params['num_leaves'] = args.num_leaves
lgbm_params['min_data_in_leaf'] = args.min_data_in_leaf
lgbm_params['colsample_bytree'] = 1.0,
model = lgbm.train(lgbm_params, d_train, args.num_iterations)
accuracy = accuracy_score(
model.predict(X_valid).round(0).astype(int), y_valid)
run.log("Accuracy", np.float(accuracy))
os.makedirs('outputs', exist_ok=True)
joblib.dump(model, 'outputs/bankmarketing-lgbm-model.joblib')
import os
import numpy as np
from sklearn.metrics import mean_squared_error
import joblib
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import OneHotEncoder
import pandas as pd
from azureml.core.run import Run
from azureml.data.dataset_factory import TabularDatasetFactory
# TODO: Create TabularDataset using TabularDatasetFactory
# Data is located at:
# "https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv"
path = "https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv"
ds = TabularDatasetFactory.from_delimited_files(path) ### YOUR CODE HERE ###
x, y = clean_data(ds)
# TODO: Split data into train and test sets.
### YOUR CODE HERE ###a
x_train, x_test, y_train, y_test = train_test_split(x,
y,
test_size=0.3,
random_state=42)
run = Run.get_context()
def clean_data(data):
# Dict for cleaning data
help="Maximum number of iterations to converge")
args = parser.parse_args()
run.log("Regularization Strength:", np.float(args.C))
run.log("Max iterations:", np.int(args.max_iter))
model = LogisticRegression(C=args.C,
max_iter=args.max_iter).fit(x_train, y_train)
AUC_weighted = model.score(x_test, y_test)
run.log("AUC_weighted", np.float(AUC_weighted))
if __name__ == '__main__':
# importing the dataset for use
dataset = TabularDatasetFactory.from_delimited_files(
"https://raw.githubusercontent.com/ujjwalbb30/nd00333-capstone/ujjwalbb30-patch-1/heart_failure_clinical_records_dataset.csv"
)
# converting the dataset imported to pandas dataframe
df = dataset.to_pandas_dataframe()
x, y = df.drop(columns=['DEATH_EVENT']), df['DEATH_EVENT']
# TODO: Split data into train and test sets.
x_train, x_test, y_train, y_test = train_test_split(x,
y,
test_size=0.33,
random_state=42)
run = Run.get_context()
main()
import argparse
import os
import numpy as np
from sklearn.metrics import mean_squared_error
import joblib
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import OneHotEncoder
import pandas as pd
from azureml.core.run import Run
from azureml.data.dataset_factory import TabularDatasetFactory
# TODO: Create TabularDataset using TabularDatasetFactory
# Data is located at:
# "https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv"
location = "https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv"
ds = TabularDatasetFactory.from_delimited_files(path=location)
run = Run.get_context()
def clean_data(data):
# Dict for cleaning data
months = {
"jan": 1,
"feb": 2,
"mar": 3,
"apr": 4,
"may": 5,
"jun": 6,
"jul": 7,
"aug": 8,
import argparse
import os
import joblib
import numpy as np
from azureml.data.dataset_factory import TabularDatasetFactory
from azureml.core.run import Run
from sklearn.model_selection import train_test_split
# Select the data URL
data_url = 'https://archive.ics.uci.edu/ml/machine-learning-databases/glass/glass.data'
# create an output folder for the model file
if "outputs" not in os.listdir():
os.mkdir("./outputs")
ds = TabularDatasetFactory.from_delimited_files(path=data_url, header=False)
def clean_data(data):
# Transform the TabularDataset into a pandas dataframe, removing NA's
x_df = data.to_pandas_dataframe().dropna()
# Rename the columns to it's right labels
x_df.columns = [
'id', 'RI', 'Na', 'Mg', 'Al', 'Si', 'K', 'Ca', 'Ba', 'Fe', 'Type'
]
# Make the target variable the right type
x_df.Type = x_df.Type.astype('category')
# Remove the id column
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import OneHotEncoder
import pandas as pd
from azureml.core.run import Run
from azureml.data.dataset_factory import TabularDatasetFactory
# TODO: Create TabularDataset using TabularDatasetFactory
# Data is located at:
# "https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv"
### YOUR CODE HERE ###
web_path = [
'https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv'
]
ds = TabularDatasetFactory.from_delimited_files(path=web_path, separator=',')
run = Run.get_context()
def clean_data(data):
# Dict for cleaning data
months = {
"jan": 1,
"feb": 2,
"mar": 3,
"apr": 4,
"may": 5,
"jun": 6,
"jul": 7,
"aug": 8,
from sklearn.linear_model import LogisticRegression
import argparse
import os
import numpy as np
from sklearn.metrics import mean_squared_error
import joblib
from sklearn.model_selection import train_test_split
import pandas as pd
from azureml.core.run import Run
from azureml.data.dataset_factory import TabularDatasetFactory
# Load data
ds = TabularDatasetFactory.from_delimited_files(
"https://raw.githubusercontent.com/GowthamiWudaru/heart-Disease-Prediction-With-Azure/main/heartDisease.csv"
)
df = ds.to_pandas_dataframe()
y = df['num']
x = df.drop(['num'], axis=1)
x_train, x_test, y_train, y_test = train_test_split(x,
y,
test_size=0.3,
random_state=42)
run = Run.get_context()
def main():
# Add arguments to script
parser = argparse.ArgumentParser()
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import LabelEncoder
import pandas as pd
from azureml.core.run import Run
from azureml.data.dataset_factory import TabularDatasetFactory
from interpret.ext.blackbox import TabularExplainer
from azureml.interpret import ExplanationClient
OUTPUT_DIR = './outputs/'
os.makedirs(OUTPUT_DIR, exist_ok=True)
run = Run.get_context()
client = ExplanationClient.from_run(run)
ds = TabularDatasetFactory.from_delimited_files("https://raw.githubusercontent.com/ketcx/ml-ops-exercise/master/data/hotel_bookings_clean.csv", validate=True, include_path=False, infer_column_types=True, set_column_types=None, separator=',', header=True, partition_format=None, support_multi_line=False, empty_as_string=False)
def transform(dataframe):
le = LabelEncoder() # Label Enconder From Sklearn
# Select all categorcial features
categorical_features = list(dataframe.columns[dataframe.dtypes == object])
# Apply Label Encoding on all categorical features
return dataframe[categorical_features].apply(lambda x: le.fit_transform(x))
def clean_data(data):
"""
Work with clean dataset.
"""
#dropping unnecessary columns
unwantedcolumnlist = ["gender", "MultipleLines", "PaymentMethod", "tenure"]
df = df2.copy()
df = df.drop(unwantedcolumnlist, axis=1)
#Let's convert all the categorical variables into dummy variables
df_dummies = pd.get_dummies(df)
y = df_dummies[['Churn']]
x = df_dummies.drop(columns=['Churn'])
return x, y
ds = TabularDatasetFactory.from_delimited_files(
path=
"https://raw.githubusercontent.com/arfa-t/ML-Nanodegree/main/Capstone%20Project/churn.csv"
)
x, y = clean_data(ds)
#train test split
from sklearn.model_selection import train_test_split
x_train, x_test, y_train, y_test = train_test_split(x,
y,
test_size=0.30,
shuffle=True)
from sklearn.ensemble import RandomForestClassifier
from sklearn.linear_model import LogisticRegression
from sklearn import svm
from sklearn.svm import SVC
import argparse
import joblib
import os
import numpy as np
from sklearn.metrics import auc, roc_curve
import joblib
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import OneHotEncoder
import pandas as pd
from azureml.core.run import Run
from azureml.data.dataset_factory import TabularDatasetFactory
path = "https://www.kaggle.com/andrewmvd/heart-failure-clinical-data/download"
#this
ds = TabularDatasetFactory.from_delimited_files(path,
validate=True,
include_path=False,
infer_column_types=True,
set_column_types=None,
separator=',',
header=True,
partition_format=None)
df = ds.to_pandas()
X, y = df.loc[:, ~df.columns.isin(["DEATH_EVENT"])], df.loc[:, "DEATH_EVENT"]
X_train, X_test, y_train, y_test = train_test_split(X,
y,
test_size=0.3,
random_state=123)
run = Run.get_context()
def main():
parser = argparse.ArgumentParser()
x_df = x_df.join(education)
x_df["month"] = x_df.month.map(months)
x_df["day_of_week"] = x_df.day_of_week.map(weekdays)
x_df["poutcome"] = x_df.poutcome.apply(lambda s: 1
if s == "success" else 0)
y_df = x_df.pop("y").apply(lambda s: 1 if s == "yes" else 0)
path = "https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv"
ds = TabularDatasetFactory.from_delimited_files(path,
validate=True,
include_path=False,
infer_column_types=True,
set_column_types=None,
separator=',',
header=True,
partition_format=None,
support_multi_line=False,
empty_as_string=False,
encoding='utf8')
x, y = clean_data(ds)
# TODO: Split data into train and test sets.
x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.33)
run = Run.get_context()
import argparse
import os
import numpy as np
from sklearn.metrics import mean_squared_error
import joblib
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import OneHotEncoder
import pandas as pd
from azureml.core.run import Run
from azureml.data.dataset_factory import TabularDatasetFactory
# TODO: Create TabularDataset using TabularDatasetFactory
# Data is located at:
# "https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv"
ds = TabularDatasetFactory.from_delimited_files("https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv",validate=True, include_path=False, infer_column_types=True, set_column_types=None, separator=',', header=True, partition_format=None, support_multi_line=False, empty_as_string=False)
def clean_data(data):
# Dict for cleaning data
months = {"jan":1, "feb":2, "mar":3, "apr":4, "may":5, "jun":6, "jul":7, "aug":8, "sep":9, "oct":10, "nov":11, "dec":12}
weekdays = {"mon":1, "tue":2, "wed":3, "thu":4, "fri":5, "sat":6, "sun":7}
# Clean and one hot encode data
x_df = data.to_pandas_dataframe().dropna()
jobs = pd.get_dummies(x_df.job, prefix="job")
x_df.drop("job", inplace=True, axis=1)
x_df = x_df.join(jobs)
x_df["marital"] = x_df.marital.apply(lambda s: 1 if s == "married" else 0)
x_df["default"] = x_df.default.apply(lambda s: 1 if s == "yes" else 0)
x_df["housing"] = x_df.housing.apply(lambda s: 1 if s == "yes" else 0)
x_df["loan"] = x_df.loan.apply(lambda s: 1 if s == "yes" else 0)
import os
import numpy as np
from sklearn.metrics import mean_squared_error
import joblib
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import OneHotEncoder
import pandas as pd
from azureml.core.run import Run
from azureml.data.dataset_factory import TabularDatasetFactory
# TODO: Create TabularDataset using TabularDatasetFactory
# Data is located at:
# "https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv"
path_to_data = "https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv"
ds = TabularDatasetFactory.from_delimited_files(path=path_to_data)
run = Run.get_context()
def clean_data(data):
# Dict for cleaning data
months = {"jan":1, "feb":2, "mar":3, "apr":4, "may":5, "jun":6, "jul":7, "aug":8, "sep":9, "oct":10, "nov":11, "dec":12}
weekdays = {"mon":1, "tue":2, "wed":3, "thu":4, "fri":5, "sat":6, "sun":7}
# Clean and one hot encode data
x_df = data.to_pandas_dataframe().dropna()
jobs = pd.get_dummies(x_df.job, prefix="job")
x_df.drop("job", inplace=True, axis=1)
x_df = x_df.join(jobs)
x_df["marital"] = x_df.marital.apply(lambda s: 1 if s == "married" else 0)
x_df["default"] = x_df.default.apply(lambda s: 1 if s == "yes" else 0)
from sklearn.tree import DecisionTreeClassifier
from sklearn.model_selection import train_test_split
# Azure imports e.g., Workspace,...
from azureml.core.run import Run
from azureml.data.dataset_factory import TabularDatasetFactory
# VRK: Create TabularDataset using TabularDatasetFactory
# create nsl-kdd network train data
nsl_kdd_webpath = [
'https://raw.githubusercontent.com/venkataravikumaralladi/AzureMLCapstoneProject/main/input/KDDTrain.csv'
]
#create network analysis data set in tabular format using TabularDatasetFactory
nsl_kdd_dataset = TabularDatasetFactory.from_delimited_files(path=nsl_kdd_webpath)
class NSLKDDFeatureAnalysis:
# class variables
network_data_column_names = [
'duration', 'protocol_type', 'service',
'flag', 'src_bytes', 'dst_bytes',
'land', 'wrong_fragment', 'urgent',
'hot', 'num_failed_logins', 'logged_in',
'num_compromised', 'root_shell', 'su_attempted',
'num_root', 'num_file_creations', 'num_shells',
'num_access_files', 'num_outbound_cmds', 'is_hot_login',
'is_guest_login',
from sklearn.linear_model import LogisticRegression
import argparse
import os
import numpy as np
from sklearn.metrics import mean_squared_error
import joblib
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import OneHotEncoder
import pandas as pd
from azureml.core.run import Run
from azureml.data.dataset_factory import TabularDatasetFactory
data_loc = 'https://archive.ics.uci.edu/ml/machine-learning-databases/00537/sobar-72.csv'
ds = TabularDatasetFactory.from_delimited_files(data_loc)
#Save model for current iteration
run = Run.get_context()
x_df = ds.to_pandas_dataframe().dropna()
y_df = x_df.pop('ca_cervix')
# TODO: Split data into train and test sets.
x_train, x_test, y_train, y_test = train_test_split(x_df,
y_df,
test_size=0.2,
random_state=0)
def main():
from sklearn.linear_model import LogisticRegression
import argparse
import os
import numpy as np
from sklearn.metrics import mean_squared_error
import joblib
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import OneHotEncoder
import pandas as pd
from azureml.core.run import Run
from azureml.data.dataset_factory import TabularDatasetFactory
web_path = "https://raw.githubusercontent.com/ciccioska/Udacity-AzureML-Capstone-Project/master/healthcare-dataset-stroke-data.csv"
ds = TabularDatasetFactory.from_delimited_files(web_path)
x_df = ds.to_pandas_dataframe().dropna()
x_df["work_type"] = x_df.work_type.apply(lambda s: 1 if s == "Private" else 0
if s == "Self-employed" else -1)
x_df["bmi"] = x_df.bmi.apply(lambda s: 0 if s == "N/A" else s)
x_df["gender"] = x_df.gender.apply(lambda s: 1 if s == "Male" else 0)
x_df["Residence_type"] = x_df.Residence_type.apply(lambda s: 1
if s == "Urban" else 0)
x_df["smoking_status"] = x_df.smoking_status.apply(
lambda s: 1 if s == "smokes" else -1 if s == "never smoked" else 2
if s == "formerly smoked" else 0)
print(x_df)
y_df = x_df.pop("stroke")
import joblib
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import OneHotEncoder
import pandas as pd
from azureml.core.run import Run
from azureml.data.dataset_factory import TabularDatasetFactory
import joblib
run = Run.get_context()
# TODO: Create TabularDataset using TabularDatasetFactory
# Data is located at:
# "https://archive.ics.uci.edu/ml/machine-learning-databases/00519/heart_failure_clinical_records_dataset.csv"
ds = TabularDatasetFactory.from_delimited_files("https://archive.ics.uci.edu/ml/machine-learning-databases/00519/heart_failure_clinical_records_dataset.csv")
def clean_data(data):
x_df = data.to_pandas_dataframe()
y_df = x_df.pop("DEATH_EVENT")
return (x_df, y_df)
x, y = clean_data(ds)
# TODO: Split data into train and test sets.
x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.25, random_state=0)
def _create_tabular(self, parameters, validate):
source_type = self._json_utility.try_get_value(
parameters, self._prop_source_type, None,
lambda v: v in self._valid_source_types,
'Property "{}" must be one of {}.'.format(
self._prop_source_type, self._valid_source_types))
if source_type == 'sql_query':
query = self._get_query(parameters)
if is_dataprep_installed():
return TabularDatasetFactory.from_sql_query(query, validate)
return self._create_dataset_from_blocks(
[_Block.craft_read_sql_block(query)], TabularDataset)
path = self._get_path(parameters)
include_path = self._json_utility.try_get_value(
parameters, self._prop_include_path, self._default_include_path)
partition_format = self._json_utility.try_get_value(
parameters, self._prop_partition_format, None)
if source_type == 'parquet_files':
if is_dataprep_installed():
return TabularDatasetFactory.from_parquet_files(
path,
validate,
include_path,
partition_format=partition_format)
return self._create_dataset_from_blocks([
_Block.craft_get_file_block(path),
_Block.craft_read_parquet_block(),
_Block.craft_partition_format_block(partition_format)
if partition_format else None,
_Block.craft_drop_path_column_block()
if not include_path else None
], TabularDataset)
if source_type == 'json_lines_files':
if is_dataprep_installed():
return TabularDatasetFactory.from_json_lines_files(
path,
validate,
include_path,
partition_format=partition_format)
return self._create_dataset_from_blocks([
_Block.craft_get_file_block(path),
_Block.craft_read_json_lines_block(),
_Block.craft_partition_format_block(partition_format)
if partition_format else None,
_Block.craft_drop_path_column_block()
if not include_path else None
], TabularDataset)
if source_type == 'delimited_files':
infer_column_types = self._json_utility.try_get_value(
parameters, self._prop_infer_column_types,
self._default_infer_column_types)
separator = self._json_utility.try_get_value(
parameters, self._prop_separator, self._default_separator)
header = self._json_utility.try_get_value(parameters,
self._prop_header,
self._default_header)
if is_dataprep_installed():
return TabularDatasetFactory.from_delimited_files(
path,
validate,
include_path=include_path,
partition_format=partition_format,
infer_column_types=infer_column_types,
separator=separator,
header=header)
if infer_column_types:
_raise_dataprep_missing_error(
'Cannot infer column types',
self._error_utility.get_error_message(
'setting {} to false'.format(
self._prop_infer_column_types)))
return self._create_dataset_from_blocks([
_Block.craft_get_file_block(path),
_Block.craft_read_delimited_block(separator, header),
_Block.craft_partition_format_block(partition_format)
if partition_format else None,
_Block.craft_drop_path_column_block()
if not include_path else None
], TabularDataset)
raise RuntimeError('Unexpected code path for source_type: ' +
source_type)
# Remove any incomplete entries from the dataset.
df.dropna(inplace=True)
indices_to_keep = ~df.isin([np.nan, np.inf, -np.inf]).any(1)
x_df = df[indices_to_keep]
y_df = x_df.pop("result")
x_df.reset_index()
return x_df, y_df
# Import player data CSV
ds = TabularDatasetFactory.from_delimited_files(
path=
"https://oracleselixir-downloadable-match-data.s3-us-west-2.amazonaws.com/2020_LoL_esports_match_data_from_OraclesElixir_20210126.csv"
)
player_data = ds.to_pandas_dataframe()
# Process dataframe
x, y = clean_data(player_data)
x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.2)
run = Run.get_context()
def main():
# Add arguments
parser = argparse.ArgumentParser()
education = pd.get_dummies(x_df.education, prefix="education")
x_df.drop("education", inplace=True, axis=1)
x_df = x_df.join(education)
x_df["month"] = x_df.month.map(months)
x_df["day_of_week"] = x_df.day_of_week.map(weekdays)
x_df["poutcome"] = x_df.poutcome.apply(lambda s: 1 if s == "success" else 0)
y_df = x_df.pop("y").apply(lambda s: 1 if s == "yes" else 0)
return x_df, y_df
# TODO: Create TabularDataset using TabularDatasetFactory
# Data is located at:
# "https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv"
data_path= "https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv"
ds =TabularDatasetFactory.from_delimited_files(path=data_path, separator=",")
X, y = clean_data(ds)
# TODO: Split data into train and test sets.
### YOUR CODE HERE ###
X_train, X_test,y_train,y_test= train_test_split(X, y, test_size=0.33, random_state=42)
run = Run.get_context()
def main():
from sklearn.linear_model import LogisticRegression
import argparse
import os
import numpy as np
from sklearn.metrics import mean_squared_error
import joblib
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import OneHotEncoder
import pandas as pd
from azureml.core.run import Run
from azureml.data.dataset_factory import TabularDatasetFactory
dataset_path = "https://raw.githubusercontent.com/fati-ma/nd00333-capstone/master/heart_failure_clinical_records_dataset%5B1%5D.csv"
ds = TabularDatasetFactory.from_delimited_files(path=dataset_path)
x_df = ds.to_pandas_dataframe().dropna()
run = Run.get_context()
y_df = x_df.pop("DEATH_EVENT")
x_train, x_test, y_train, y_test = train_test_split(x_df,
y_df,
test_size=0.3,
random_state=123)
def main():
# Add arguments to script
parser = argparse.ArgumentParser()
import os
import numpy as np
from sklearn.metrics import mean_squared_error
import joblib
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import OneHotEncoder
import pandas as pd
from azureml.core.run import Run
from azureml.data.dataset_factory import TabularDatasetFactory
# TODO: Create TabularDataset using TabularDatasetFactory
# Data is located at:
# "https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv"
DATA_URL = "https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv"
ds = TabularDatasetFactory.from_delimited_files(DATA_URL)
def clean_data(data):
# Dict for cleaning data
months = {
"jan": 1,
"feb": 2,
"mar": 3,
"apr": 4,
"may": 5,
"jun": 6,
"jul": 7,
"aug": 8,
"sep": 9,
"oct": 10,
from sklearn.linear_model import LogisticRegression
import argparse
import os
import numpy as np
import joblib
from sklearn.model_selection import train_test_split
from sklearn.metrics import roc_auc_score
import pandas as pd
from azureml.core.run import Run
from azureml.data.dataset_factory import TabularDatasetFactory
url = "https://raw.githubusercontent.com/benbot8/azure_capstone/f54d4da8ea6c2cd5c9082b6703271c25d5f16b40/starter_file/train.csv"
df = TabularDatasetFactory.from_delimited_files(path=url)
#df = pd.read_csv("train.csv")
def clean_data(data):
#clean data and convert categorical to indicator variables
x_df = data.to_pandas_dataframe().dropna()
x_df.reset_index(drop=True, inplace=True)
x_df.drop(['state', 'account_length', 'area_code'], axis=1, inplace=True)
x_df['international_plan'] = x_df.international_plan.apply(
lambda s: 1 if s == "yes" else 0)
x_df['voice_mail_plan'] = x_df.voice_mail_plan.apply(lambda s: 1
if s == "yes" else 0)
x_df['churn'] = x_df.churn.apply(lambda s: 1 if s == "yes" else 0)
x_df.rename(columns={"churn": "y"}, inplace=True)
y_df = x_df.pop("y")
return x_df, y_df
# Azure imports e.g., Workspace,...
from azureml.core.run import Run
from azureml.data.dataset_factory import TabularDatasetFactory
# VRK: Create TabularDataset using TabularDatasetFactory
# Data is located at:
# "https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv"
# create bank marketing tabular dataset from public azure ml.
bankmarketing_webpath = [
'https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv'
]
#create bankmarketing data set in tabular format using TabularDatasetFactory
bankmarketing_dataset = TabularDatasetFactory.from_delimited_files(
path=bankmarketing_webpath)
def clean_data(data):
# Dict for cleaning data
months = {
"jan": 1,
"feb": 2,
"mar": 3,
"apr": 4,
"may": 5,
"jun": 6,
"jul": 7,
"aug": 8,
"sep": 9,
"oct": 10,
x_df["day_of_week"] = x_df.day_of_week.map(weekdays)
x_df["poutcome"] = x_df.poutcome.apply(lambda s: 1
if s == "success" else 0)
y_df = x_df.pop("y").apply(lambda s: 1 if s == "yes" else 0)
return (x_df, y_df
) #adding missed return (df with both features and targets)
# TODO: Create TabularDataset using TabularDatasetFactory
# Data is located at:
# "https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv"
#creating a tabularDataset
ds = TabularDatasetFactory.from_delimited_files(
'https://automlsamplenotebookdata.blob.core.windows.net/automl-sample-notebook-data/bankmarketing_train.csv'
) ### YOUR CODE HERE ###
#cleaning data
x, y = clean_data(ds)
# TODO: Split data into train and test sets.
#spliting data --> 80-20
x_train, x_test, y_train, y_test = train_test_split(x,
y,
test_size=0.2,
random_state=42)
run = Run.get_context()
def main():
import numpy as np
import pandas as pd
from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split
from azureml.core.run import Run
import argparse
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import roc_auc_score
from azureml.data.dataset_factory import TabularDatasetFactory
#-----------------------------------------------------------------------------
#Import data (out of Azure):
#data = pp.load_data('https://www.openml.org/data/get_csv/16826755/phpMYEkMl')
#Import data (Within Azure):
data = TabularDatasetFactory.from_delimited_files(
'https://www.openml.org/data/get_csv/16826755/phpMYEkMl')
dataset = data.to_pandas_dataframe()
#-----------------------------------------------------------------------------
#Clean Data
#Define clean data function
def clean_data(df):
df = pp.replace_question_marks(df)
df['cabin'] = df['cabin'].apply(pp.get_first_cabin)
df['title'] = df['name'].apply(pp.get_title)
# cast numerical variables as floats
