def test_ndarray_to_schema(self):
from lale.datasets.data_schemas import to_schema
from lale.type_checking import validate_schema
all_X, all_y = self._irisArr["X"], self._irisArr["y"]
assert not hasattr(all_X, "json_schema")
all_X_schema = to_schema(all_X)
validate_schema(all_X, all_X_schema, subsample_array=False)
assert not hasattr(all_y, "json_schema")
all_y_schema = to_schema(all_y)
validate_schema(all_y, all_y_schema, subsample_array=False)
all_X_expected = {
"type": "array",
"minItems": 150,
"maxItems": 150,
"items": {
"type": "array",
"minItems": 4,
"maxItems": 4,
"items": {
"type": "number"
},
},
}
all_y_expected = {
"type": "array",
"minItems": 150,
"maxItems": 150,
"items": {
"type": "integer"
},
}
self.maxDiff = None
self.assertEqual(all_X_schema, all_X_expected)
self.assertEqual(all_y_schema, all_y_expected)
def test_pandas_to_schema(self):
from lale.datasets.data_schemas import to_schema
from lale.type_checking import validate_schema
import pandas as pd
train_X, train_y = self._irisDf['X'], self._irisDf['y']
assert isinstance(train_X, pd.DataFrame)
assert not hasattr(train_X, 'json_schema')
train_X_schema = to_schema(train_X)
validate_schema(train_X, train_X_schema, subsample_array=False)
assert isinstance(train_y, pd.Series)
assert not hasattr(train_y, 'json_schema')
train_y_schema = to_schema(train_y)
validate_schema(train_y, train_y_schema, subsample_array=False)
train_X_expected = {
'type': 'array', 'minItems': 120, 'maxItems': 120,
'items': {
'type': 'array', 'minItems': 4, 'maxItems': 4,
'items': [
{'description': 'sepal length (cm)', 'type': 'number'},
{'description': 'sepal width (cm)', 'type': 'number'},
{'description': 'petal length (cm)', 'type': 'number'},
{'description': 'petal width (cm)', 'type': 'number'}]}}
train_y_expected = {
'type': 'array', 'minItems': 120, 'maxItems': 120,
'items': {'description': 'target', 'type': 'integer'}}
self.maxDiff = None
self.assertEqual(train_X_schema, train_X_expected)
self.assertEqual(train_y_schema, train_y_expected)
def test_ndarray_to_schema(self):
from lale.datasets.data_schemas import to_schema
from lale.type_checking import validate_schema
all_X, all_y = self._irisArr['X'], self._irisArr['y']
assert not hasattr(all_X, 'json_schema')
all_X_schema = to_schema(all_X)
validate_schema(all_X, all_X_schema, subsample_array=False)
assert not hasattr(all_y, 'json_schema')
all_y_schema = to_schema(all_y)
validate_schema(all_y, all_y_schema, subsample_array=False)
all_X_expected = {
'type': 'array',
'minItems': 150,
'maxItems': 150,
'items': {
'type': 'array',
'minItems': 4,
'maxItems': 4,
'items': {
'type': 'number'
}
}
}
all_y_expected = {
'type': 'array',
'minItems': 150,
'maxItems': 150,
'items': {
'type': 'integer'
}
}
self.maxDiff = None
self.assertEqual(all_X_schema, all_X_expected)
self.assertEqual(all_y_schema, all_y_expected)
def test_pandas_to_schema(self):
import pandas as pd
from lale.datasets.data_schemas import to_schema
from lale.type_checking import validate_schema
train_X, train_y = self._irisDf["X"], self._irisDf["y"]
assert isinstance(train_X, pd.DataFrame)
assert not hasattr(train_X, "json_schema")
train_X_schema = to_schema(train_X)
validate_schema(train_X, train_X_schema, subsample_array=False)
assert isinstance(train_y, pd.Series)
assert not hasattr(train_y, "json_schema")
train_y_schema = to_schema(train_y)
validate_schema(train_y, train_y_schema, subsample_array=False)
train_X_expected = {
"type": "array",
"minItems": 120,
"maxItems": 120,
"items": {
"type":
"array",
"minItems":
4,
"maxItems":
4,
"items": [
{
"description": "sepal length (cm)",
"type": "number"
},
{
"description": "sepal width (cm)",
"type": "number"
},
{
"description": "petal length (cm)",
"type": "number"
},
{
"description": "petal width (cm)",
"type": "number"
},
],
},
}
train_y_expected = {
"type": "array",
"minItems": 120,
"maxItems": 120,
"items": {
"description": "target",
"type": "integer"
},
}
self.maxDiff = None
self.assertEqual(train_X_schema, train_X_expected)
self.assertEqual(train_y_schema, train_y_expected)
def test_datasets_with_own_schemas(self):
from lale.datasets.data_schemas import to_schema
from lale.type_checking import validate_schema
for name in ['irisArr', 'irisDf', 'digits', 'housing', 'creditG', 'movies', 'drugRev']:
dataset = getattr(self, f'_{name}')
data_X, data_y = dataset['X'], dataset['y']
schema_X, schema_y = to_schema(data_X), to_schema(data_y)
validate_schema(data_X, schema_X, subsample_array=False)
validate_schema(data_y, schema_y, subsample_array=False)
def test_datasets_with_own_schemas(self):
from lale.datasets.data_schemas import to_schema
from lale.type_checking import validate_schema
for name in [
"irisArr",
"irisDf",
"digits",
"housing",
"creditG",
"movies",
"drugRev",
]:
dataset = getattr(self, f"_{name}")
data_X, data_y = dataset["X"], dataset["y"]
schema_X, schema_y = to_schema(data_X), to_schema(data_y)
validate_schema(data_X, schema_X, subsample_array=False)
validate_schema(data_y, schema_y, subsample_array=False)
def test_arff_to_schema(self):
from lale.datasets.data_schemas import to_schema
from lale.type_checking import validate_schema
train_X, train_y = self._creditG["X"], self._creditG["y"]
assert hasattr(train_X, "json_schema")
train_X_schema = to_schema(train_X)
validate_schema(train_X, train_X_schema, subsample_array=False)
assert hasattr(train_y, "json_schema")
train_y_schema = to_schema(train_y)
validate_schema(train_y, train_y_schema, subsample_array=False)
train_X_expected = {
"type": "array",
"minItems": 670,
"maxItems": 670,
"items": {
"type":
"array",
"minItems":
20,
"maxItems":
20,
"items": [
{
"description": "checking_status",
"enum": ["<0", "0<=X<200", ">=200", "no checking"],
},
{
"description": "duration",
"type": "number"
},
{
"description":
"credit_history",
"enum": [
"no credits/all paid",
"all paid",
"existing paid",
"delayed previously",
"critical/other existing credit",
],
},
{
"description":
"purpose",
"enum": [
"new car",
"used car",
"furniture/equipment",
"radio/tv",
"domestic appliance",
"repairs",
"education",
"vacation",
"retraining",
"business",
"other",
],
},
{
"description": "credit_amount",
"type": "number"
},
{
"description":
"savings_status",
"enum": [
"<100",
"100<=X<500",
"500<=X<1000",
">=1000",
"no known savings",
],
},
{
"description": "employment",
"enum":
["unemployed", "<1", "1<=X<4", "4<=X<7", ">=7"],
},
{
"description": "installment_commitment",
"type": "number"
},
{
"description":
"personal_status",
"enum": [
"male div/sep",
"female div/dep/mar",
"male single",
"male mar/wid",
"female single",
],
},
{
"description": "other_parties",
"enum": ["none", "co applicant", "guarantor"],
},
{
"description": "residence_since",
"type": "number"
},
{
"description":
"property_magnitude",
"enum": [
"real estate",
"life insurance",
"car",
"no known property",
],
},
{
"description": "age",
"type": "number"
},
{
"description": "other_payment_plans",
"enum": ["bank", "stores", "none"],
},
{
"description": "housing",
"enum": ["rent", "own", "for free"]
},
{
"description": "existing_credits",
"type": "number"
},
{
"description":
"job",
"enum": [
"unemp/unskilled non res",
"unskilled resident",
"skilled",
"high qualif/self emp/mgmt",
],
},
{
"description": "num_dependents",
"type": "number"
},
{
"description": "own_telephone",
"enum": ["none", "yes"]
},
{
"description": "foreign_worker",
"enum": ["yes", "no"]
},
],
},
}
train_y_expected = {
"type": "array",
"minItems": 670,
"maxItems": 670,
"items": {
"description": "class",
"enum": ["good", "bad"]
},
}
self.maxDiff = None
self.assertEqual(train_X_schema, train_X_expected)
self.assertEqual(train_y_schema, train_y_expected)
def test_arff_to_schema(self):
from lale.datasets.data_schemas import to_schema
from lale.type_checking import validate_schema
train_X, train_y = self._creditG['X'], self._creditG['y']
assert hasattr(train_X, 'json_schema')
train_X_schema = to_schema(train_X)
validate_schema(train_X, train_X_schema, subsample_array=False)
assert hasattr(train_y, 'json_schema')
train_y_schema = to_schema(train_y)
validate_schema(train_y, train_y_schema, subsample_array=False)
train_X_expected = {
'type': 'array', 'minItems': 670, 'maxItems': 670,
'items': {
'type': 'array', 'minItems': 20, 'maxItems': 20,
'items': [
{'description': 'checking_status', 'enum': [
'<0', '0<=X<200', '>=200', 'no checking']},
{'description': 'duration', 'type': 'number'},
{'description': 'credit_history', 'enum': [
'no credits/all paid', 'all paid',
'existing paid', 'delayed previously',
'critical/other existing credit']},
{'description': 'purpose', 'enum': [
'new car', 'used car', 'furniture/equipment',
'radio/tv', 'domestic appliance', 'repairs',
'education', 'vacation', 'retraining', 'business',
'other']},
{'description': 'credit_amount', 'type': 'number'},
{'description': 'savings_status', 'enum': [
'<100', '100<=X<500', '500<=X<1000', '>=1000',
'no known savings']},
{'description': 'employment', 'enum': [
'unemployed', '<1', '1<=X<4', '4<=X<7', '>=7']},
{'description': 'installment_commitment', 'type': 'number'},
{'description': 'personal_status', 'enum': [
'male div/sep', 'female div/dep/mar', 'male single',
'male mar/wid', 'female single']},
{'description': 'other_parties', 'enum': [
'none', 'co applicant', 'guarantor']},
{'description': 'residence_since', 'type': 'number'},
{'description': 'property_magnitude', 'enum': [
'real estate', 'life insurance', 'car',
'no known property']},
{'description': 'age', 'type': 'number'},
{'description': 'other_payment_plans', 'enum': [
'bank', 'stores', 'none']},
{'description': 'housing', 'enum': [
'rent', 'own', 'for free']},
{'description': 'existing_credits', 'type': 'number'},
{'description': 'job', 'enum': [
'unemp/unskilled non res', 'unskilled resident',
'skilled', 'high qualif/self emp/mgmt']},
{'description': 'num_dependents', 'type': 'number'},
{'description': 'own_telephone', 'enum': ['none', 'yes']},
{'description': 'foreign_worker', 'enum': ['yes', 'no']}]}}
train_y_expected = {
'type': 'array', 'minItems': 670, 'maxItems': 670,
'items': {'description': 'class', 'enum': [0, 1]}}
self.maxDiff = None
self.assertEqual(train_X_schema, train_X_expected)
self.assertEqual(train_y_schema, train_y_expected)
