Added test data locally

.gitattributes

@@ -1,3 +1,6 @@
 # Set the default behavior, in case people don't have core.autocrlf set.
 * text=auto
-docs/** linguist-documentation
+docs/** linguist-documentation
+*.csv.tar.gz filter=lfs diff=lfs merge=lfs -text
+*.json.tar.gz filter=lfs diff=lfs merge=lfs -text
+*.txt.tar.gz filter=lfs diff=lfs merge=lfs -text

.github/workflows/build-python.yml

@@ -115,6 +115,7 @@ jobs:
         uses: actions/checkout@v4
         with:
           fetch-depth: 0
+          lfs: true
 
       - name: Setup PDM
         uses: pdm-project/setup-pdm@v4

docs/reference/notebooks/amazon_review_classification_sklearn.ipynb

@@ -135,8 +135,8 @@
     "TARGET_NAME = \"isHelpful\"\n",
     "\n",
     "# Paths.\n",
-    "DATA_URL = \"ftp://sys.giskard.ai/pub/unit_test_resources/amazon_review_dataset/reviews.json\"\n",
-    "DATA_PATH = Path.home() / \".giskard\" / \"amazon_review_dataset\" / \"reviews.json\""
+    "DATA_URL = \"https://github.com/Giskard-AI/giskard/raw/a501a92b215bf730f6647be4d118a0c1ce67eb6e/tests/test_data/amazon_reviews.json.tar.gz\"\n",
+    "DATA_PATH = Path.home() / \".giskard\" / \"amazon_review_dataset\" / \"reviews.json.tar.gz\""
    ]
   },
   {

docs/reference/notebooks/drug_classification_sklearn.ipynb

@@ -76,12 +76,12 @@
     "\n",
     "import numpy as np\n",
     "import pandas as pd\n",
-    "from sklearn.svm import SVC\n",
     "from imblearn.over_sampling import SMOTE\n",
+    "from imblearn.pipeline import Pipeline as PipelineImb\n",
     "from sklearn.metrics import accuracy_score\n",
-    "from sklearn.preprocessing import OneHotEncoder\n",
     "from sklearn.model_selection import train_test_split\n",
-    "from imblearn.pipeline import Pipeline as PipelineImb\n",
+    "from sklearn.preprocessing import OneHotEncoder\n",
+    "from sklearn.svm import SVC\n",
     "\n",
     "from giskard import Dataset, Model, scan, testing"
    ]
@@ -121,8 +121,8 @@
     "NA_TO_K_CATEGORIES = ['<10', '10-20', '20-30', '>30']\n",
     "\n",
     "# Paths.\n",
-    "DATA_URL = \"ftp://sys.giskard.ai/pub/unit_test_resources/drug_classification_dataset/drug200.csv\"\n",
-    "DATA_PATH = Path.home() / \".giskard\" / \"drug_classification_dataset\" / \"drug200.csv\""
+    "DATA_URL = \"https://github.com/Giskard-AI/giskard/raw/abac244d63e5deac5f42009f26971ee9ee6592a9/tests/test_data/drug_classification.csv.tar.gz\"\n",
+    "DATA_PATH = Path.home() / \".giskard\" / \"drug_classification_dataset\" / \"drug200.csv.tar.gz\""
    ]
   },
   {

docs/reference/notebooks/hotel_text_regression.ipynb

@@ -67,6 +67,7 @@
    "outputs": [],
    "source": [
     "from pathlib import Path\n",
+    "from typing import Iterable\n",
     "from urllib.request import urlretrieve\n",
     "\n",
     "import pandas as pd\n",
@@ -76,7 +77,6 @@
     "from sklearn.model_selection import train_test_split\n",
     "from sklearn.pipeline import Pipeline\n",
     "from sklearn.preprocessing import FunctionTransformer\n",
-    "from typing import Iterable\n",
     "\n",
     "from giskard import Model, Dataset, scan, testing"
    ]
@@ -107,8 +107,8 @@
     "TARGET_COLUMN_NAME = \"Reviewer_Score\"\n",
     "\n",
     "# Paths.\n",
-    "DATA_URL = \"ftp://sys.giskard.ai/pub/unit_test_resources/hotel_text_regression_dataset/Hotel_Reviews.csv\"\n",
-    "DATA_PATH = Path.home() / \".giskard\" / \"hotel_text_regression_dataset\" / \"Hotel_Reviews.csv\""
+    "DATA_URL = \"https://github.com/Giskard-AI/giskard/raw/abac244d63e5deac5f42009f26971ee9ee6592a9/tests/test_data/hotel_reviews.csv.tar.gz\"\n",
+    "DATA_PATH = Path.home() / \".giskard\" / \"hotel_text_regression_dataset\" / \"Hotel_Reviews.csv.tar.gz\""
    ]
   },
   {

docs/reference/notebooks/ieee_fraud_detection_adversarial_validation.ipynb

@@ -118,7 +118,11 @@
     "IDX_LABEL = 'TransactionID'\n",
     "\n",
     "# Paths.\n",
-    "DATA_URL = \"ftp://sys.giskard.ai/pub/unit_test_resources/fraud_detection_classification_dataset/{}\"\n",
+    "TEST_IDENTITY_DATA_URL = \"https://github.com/Giskard-AI/giskard/raw/abac244d63e5deac5f42009f26971ee9ee6592a9/tests/test_data/test_identity.csv.tar.gz\"\n",
+    "TEST_TRANSACTION_DATA_URL = \"https://github.com/Giskard-AI/giskard/raw/abac244d63e5deac5f42009f26971ee9ee6592a9/tests/test_data/test_transaction.csv.tar.gz\"\n",
+    "TRAIN_IDENTITY_DATA_URL = \"https://github.com/Giskard-AI/giskard/raw/abac244d63e5deac5f42009f26971ee9ee6592a9/tests/test_data/train_identity.csv.tar.gz\"\n",
+    "TRAIN_TRANSACTION_DATA_URL = \"https://github.com/Giskard-AI/giskard/raw/abac244d63e5deac5f42009f26971ee9ee6592a9/tests/test_data/train_transaction.csv.tar.gz\"\n",
+    "\n",
     "DATA_PATH = Path.home() / \".giskard\" / \"fraud_detection_classification_dataset\""
    ]
   },
@@ -165,9 +169,9 @@
     "\n",
     "\n",
     "def fetch_dataset():\n",
-    "    files_to_fetch = [\"train_transaction.csv\", \"train_identity.csv\", \"test_transaction.csv\", \"test_identity.csv\"]\n",
+    "    files_to_fetch = [TEST_IDENTITY_DATA_URL, TEST_TRANSACTION_DATA_URL, TRAIN_IDENTITY_DATA_URL, TRAIN_TRANSACTION_DATA_URL]\n",
     "    for file_name in files_to_fetch:\n",
-    "        fetch_from_ftp(DATA_URL.format(file_name), DATA_PATH / file_name)\n",
+    "        fetch_from_ftp(file_name, DATA_PATH / file_name)\n",
     "\n",
     "\n",
     "# Define data-types of transactions features.\n",
@@ -225,11 +229,11 @@
     "def read_set(_type):\n",
     "    \"\"\"Read both transactions and identity data.\"\"\"\n",
     "    print(f\"Reading transactions data...\")\n",
-    "    _df = pd.read_csv(os.path.join(DATA_PATH, f'{_type}_transaction.csv'),\n",
+    "    _df = pd.read_csv(os.path.join(DATA_PATH, f'{_type}_transaction.csv.tar.gz'),\n",
     "                      index_col=IDX_LABEL, dtype=DATA_TYPES_TRANSACTION, nrows=250)\n",
     "\n",
     "    print(f\"Reading identity data...\")\n",
-    "    _df = _df.join(pd.read_csv(os.path.join(DATA_PATH, f'{_type}_identity.csv'),\n",
+    "    _df = _df.join(pd.read_csv(os.path.join(DATA_PATH, f'{_type}_identity.csv.tar.gz'),\n",
     "                               index_col=IDX_LABEL, dtype=DATA_TYPES_ID))\n",
     "    return _df\n",
     "\n",

docs/reference/notebooks/medical_transcript_classification_sklearn.ipynb

@@ -95,6 +95,7 @@
    "source": [
     "import string\n",
     "from pathlib import Path\n",
+    "from typing import Iterable\n",
     "from urllib.request import urlretrieve\n",
     "\n",
     "import nltk\n",
@@ -107,7 +108,6 @@
     "from sklearn.model_selection import train_test_split\n",
     "from sklearn.pipeline import Pipeline\n",
     "from sklearn.preprocessing import FunctionTransformer\n",
-    "from typing import Iterable\n",
     "\n",
     "from giskard import Dataset, Model, scan, testing"
    ]
@@ -154,8 +154,8 @@
     "RANDOM_SEED = 8888\n",
     "\n",
     "# Data.\n",
-    "DATA_URL = \"ftp://sys.giskard.ai/pub/unit_test_resources/medical_transcript_classification_dataset/mtsamples.csv\"\n",
-    "DATA_PATH = Path.home() / \".giskard\" / \"medical_transcript_classification_dataset\" / \"mtsamples.csv\""
+    "DATA_URL = \"https://github.com/Giskard-AI/giskard/raw/abac244d63e5deac5f42009f26971ee9ee6592a9/tests/test_data/medical_transcript_classification_dataset_mtsamples.csv.tar.gz\"\n",
+    "DATA_PATH = Path.home() / \".giskard\" / \"medical_transcript_classification_dataset\" / \"mtsamples.csv.tar.gz\""
    ]
   },
   {

docs/reference/notebooks/tripadvisor_sentiment_classification.ipynb

@@ -75,6 +75,7 @@
     "import string\n",
     "from dataclasses import dataclass\n",
     "from pathlib import Path\n",
+    "from typing import Union, List\n",
     "from urllib.request import urlretrieve\n",
     "\n",
     "import nltk\n",
@@ -85,7 +86,6 @@
     "from torch.utils.data import DataLoader\n",
     "from torch.utils.data import TensorDataset\n",
     "from transformers import DistilBertForSequenceClassification, DistilBertTokenizer\n",
-    "from typing import Union, List\n",
     "\n",
     "from giskard import Dataset, Model, scan, testing"
    ]
@@ -123,9 +123,9 @@
     "STOP_WORDS = set(stopwords.words('english'))\n",
     "RANDOM_SEED = 0\n",
     "\n",
-    "DATA_URL = \"ftp://sys.giskard.ai/pub/unit_test_resources/tripadvisor_reviews_dataset/{}\"\n",
+    "DATA_URL = \"https://github.com/Giskard-AI/giskard/raw/abac244d63e5deac5f42009f26971ee9ee6592a9/tests/test_data/{}\"\n",
     "DATA_PATH = Path.home() / \".giskard\" / \"tripadvisor_reviews_dataset\"\n",
-    "DATA_FILE_NAME = \"tripadvisor_hotel_reviews.csv\""
+    "DATA_FILE_NAME = \"tripadvisor_hotel_reviews.csv.tar.gz\""
    ]
   },
   {

tests/fixtures/amazon_review__binary_classification.py

@@ -1,5 +1,4 @@
 import string
-from pathlib import Path
 
 import numpy as np
 import pandas as pd
@@ -12,7 +11,7 @@
 
 from giskard import Dataset
 from giskard.models.sklearn import SKLearnModel
-from tests.url_utils import fetch_from_ftp
+from tests import path
 
 # Constants.
 RANDOM_SEED = 0
@@ -24,14 +23,7 @@
 FEATURE_COLUMN_NAME = "reviewText"
 
 # Data.
-DATA_URL = "ftp://sys.giskard.ai/pub/unit_test_resources/amazon_review_dataset/reviews.json"
-DATA_PATH = Path.home() / ".giskard" / "amazon_review_dataset" / "reviews.json"
-
-
-def download_data(**kwargs) -> pd.DataFrame:
-    fetch_from_ftp(DATA_URL, DATA_PATH)
-    _df = pd.read_json(DATA_PATH, lines=True, **kwargs)
-    return _df
+DATA_PATH = path("test_data/amazon_reviews.json.tar.gz")
 
 
 def preprocess_data(df: pd.DataFrame) -> pd.DataFrame:
@@ -58,7 +50,7 @@ def preprocess_data(df: pd.DataFrame) -> pd.DataFrame:
 
 @pytest.fixture(scope="session")
 def amazon_review_raw_data() -> pd.DataFrame:
-    return preprocess_data(download_data(nrows=5000))
+    return preprocess_data(pd.read_json(DATA_PATH, lines=True, nrows=5000))
 
 
 @pytest.fixture()

tests/fixtures/drug_classification__multiclass_classification.py

@@ -1,7 +1,5 @@
 from typing import List, Tuple
 
-from pathlib import Path
-
 import pandas as pd
 import pytest
 from imblearn.over_sampling import SMOTE
@@ -11,11 +9,10 @@
 
 from giskard import Dataset
 from giskard.models.sklearn import SKLearnModel
-from tests.url_utils import fetch_from_ftp
+from tests import path
 
 # Data.
-DATA_URL = "ftp://sys.giskard.ai/pub/unit_test_resources/drug_classification_dataset/drug200.csv"
-DATA_PATH = Path.home() / ".giskard" / "drug_classification_dataset" / "drug200.csv"
+DATA_PATH = path("test_data/drug_classification.csv.tar.gz")
 
 # Constants.
 TARGET_NAME = "Drug"
@@ -49,12 +46,7 @@ def _bin_na_to_k(_df: pd.DataFrame) -> pd.DataFrame:
 
 @pytest.fixture(scope="session")
 def drug_classification_raw_data() -> pd.DataFrame:
-    # Download data.
-    fetch_from_ftp(DATA_URL, DATA_PATH)
-
-    # Load and wrap data.
-    raw_data = bin_numerical(pd.read_csv(DATA_PATH))
-    return raw_data
+    return bin_numerical(pd.read_csv(DATA_PATH))
 
 
 @pytest.fixture()

tests/fixtures/fraud_detection__binary_classification.py

@@ -1,19 +1,16 @@
 from typing import List, Tuple
 
-from pathlib import Path
-
 import pandas as pd
 import pytest
 from lightgbm import LGBMClassifier
 from pandas.api.types import union_categoricals
 from sklearn.model_selection import train_test_split
 
 from giskard import Dataset, Model
-from tests.url_utils import fetch_from_ftp
+from tests import path
 
 # Data.
-DATA_URL = "ftp://sys.giskard.ai/pub/unit_test_resources/fraud_detection_classification_dataset/{}"
-DATA_PATH = Path.home() / ".giskard" / "fraud_detection_classification_dataset"
+DATA_URL = path("test_data/{}")
 
 # Constants.
 TARGET_COLUMN = "isTest"
@@ -85,20 +82,12 @@
 ]
 
 
-def fetch_dataset():
-    files_to_fetch = ["train_transaction.csv", "train_identity.csv", "test_transaction.csv", "test_identity.csv"]
-    for file_name in files_to_fetch:
-        fetch_from_ftp(DATA_URL.format(file_name), DATA_PATH / file_name)
-
-
 def read_set(_type, nrows=150):
     """Read both transactions and identity data."""
-    fetch_dataset()
-
     _df = pd.read_csv(
-        DATA_PATH / f"{_type}_transaction.csv", index_col=IDX_LABEL, dtype=DATA_TYPES_TRANSACTION, nrows=nrows
+        DATA_URL / f"{_type}_transaction.csv", index_col=IDX_LABEL, dtype=DATA_TYPES_TRANSACTION, nrows=nrows
     )
-    _df = _df.join(pd.read_csv(DATA_PATH / f"{_type}_identity.csv", index_col=IDX_LABEL, dtype=DATA_TYPES_ID))
+    _df = _df.join(pd.read_csv(DATA_URL / f"{_type}_identity.csv", index_col=IDX_LABEL, dtype=DATA_TYPES_ID))
 
     return _df
 

tests/fixtures/hotel_text__regression.py

@@ -1,7 +1,5 @@
 from typing import Iterable
 
-from pathlib import Path
-
 import pandas as pd
 import pytest
 from sklearn.ensemble import GradientBoostingRegressor
@@ -11,11 +9,10 @@
 
 from giskard import Dataset
 from giskard.models.sklearn import SKLearnModel
-from tests.url_utils import fetch_from_ftp
+from tests import path
 
 # Data.
-DATA_URL = "ftp://sys.giskard.ai/pub/unit_test_resources/hotel_text_regression_dataset/Hotel_Reviews.csv"
-DATA_PATH = Path.home() / ".giskard" / "hotel_text_regression_dataset" / "Hotel_Reviews.csv"
+DATA_PATH = path("test_data/hotel_reviews.csv.tar.gz")
 
 # Constants.
 FEATURE_COLUMN_NAME = "Full_Review"
@@ -33,8 +30,6 @@ def load_data(**kwargs) -> pd.DataFrame:
 
 @pytest.fixture(scope="session")
 def hotel_text_raw_data():
-    fetch_from_ftp(DATA_URL, DATA_PATH)
-
     raw_data = load_data(nrows=105)[[FEATURE_COLUMN_NAME, TARGET_COLUMN_NAME]]
     return raw_data
 

tests/fixtures/medical_transcript_multiclass_classification.py

@@ -1,7 +1,6 @@
 from typing import Iterable
 
 import string
-from pathlib import Path
 
 import pandas as pd
 import pytest
@@ -12,7 +11,7 @@
 
 from giskard import Dataset
 from giskard.models.sklearn import SKLearnModel
-from tests.url_utils import fetch_from_ftp
+from tests import path
 
 # Constants.
 LABELS_LIST = [
@@ -29,13 +28,11 @@
 LANGUAGE = "english"
 
 # Paths.
-DATA_URL = "ftp://sys.giskard.ai/pub/unit_test_resources/medical_transcript_classification_dataset/mtsamples.csv"
-DATA_PATH = Path.home() / ".giskard" / "medical_transcript_classification_dataset" / "mtsamples.csv"
+DATA_PATH = path("test_data/medical_transcript_classification_dataset_mtsamples.csv.tar.gz")
 
 
 def load_data() -> pd.DataFrame:
     # Download dataset.
-    fetch_from_ftp(DATA_URL, DATA_PATH)
     df = pd.read_csv(DATA_PATH)
 
     # Drop useless columns.

tests/fixtures/tripadvisor_text_classification_torch.py

@@ -3,7 +3,6 @@
 import re
 import string
 from dataclasses import dataclass
-from pathlib import Path
 
 import numpy as np
 import pandas as pd
@@ -13,12 +12,10 @@
 from transformers import DistilBertForSequenceClassification, DistilBertTokenizer
 
 from giskard import Dataset, Model, models
-from tests.url_utils import fetch_from_ftp
+from tests import path
 
 # Data
-DATA_URL = "ftp://sys.giskard.ai/pub/unit_test_resources/tripadvisor_reviews_dataset/{}"
-DATA_PATH = Path.home() / ".giskard" / "tripadvisor_reviews_dataset"
-DATA_FILE_NAME = "tripadvisor_hotel_reviews.csv"
+DATA_PATH = path("test_data/tripadvisor_hotel_reviews.csv.tar.gz")
 
 # Constants
 PRETRAINED_WEIGHTS_NAME = "distilbert-base-uncased"
@@ -115,8 +112,7 @@ def text_preprocessor(df: pd.DataFrame) -> pd.DataFrame:
 
 def load_dataset() -> pd.DataFrame:
     # Download dataset
-    fetch_from_ftp(DATA_URL.format(DATA_FILE_NAME), DATA_PATH / DATA_FILE_NAME)
-    df = pd.read_csv(DATA_PATH / DATA_FILE_NAME, nrows=MAX_NUM_ROWS)
+    df = pd.read_csv(DATA_PATH, nrows=MAX_NUM_ROWS)
     # Obtain labels for our task.
     df[TARGET_COLUMN_NAME] = df.Rating.apply(lambda x: create_label(x))
     df.drop(columns="Rating", inplace=True)