datfid 0.1.12__tar.gz

datfid-0.1.12/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 DATFID
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

datfid-0.1.12/PKG-INFO

@@ -0,0 +1,87 @@
+Metadata-Version: 2.4
+Name: datfid
+Version: 0.1.12
+Summary: SDK to access the DATFID API hosted on Hugging Face Spaces
+Author: DATFID
+Author-email: igor.schapiro@datfid.com
+License: MIT
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3
+Requires-Python: >=3.7
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: requests>=2.31.0
+Requires-Dist: pandas>=1.0.1
+Requires-Dist: numpy<2.1,>=1.22
+Dynamic: author
+Dynamic: author-email
+Dynamic: classifier
+Dynamic: description
+Dynamic: description-content-type
+Dynamic: license
+Dynamic: license-file
+Dynamic: requires-dist
+Dynamic: requires-python
+Dynamic: summary
+
+# DATFID SDK
+
+A Python SDK to access the DATFID API running on Hugging Face Spaces.
+
+## Installation
+
+```bash
+pip install datfid
+```
+
+## Usage
+
+```python
+from datfid import DATFIDClient
+
+# Initialize the client with your Hugging Face token
+client = DATFIDClient(token="your_huggingface_token")
+
+# Fit a model
+fit_result = client.fit_model(
+    file_path="path/to/your/data.xlsx",
+    id_col="Individual",
+    time_col="Time",
+    y="Loan Probability",
+    lagged_features={"Repayment Amount": 1, "Missed Payments": 2},
+    current_features=["Credit Score", "Unemployment Rate"],
+    filter_by_significance=True,
+    meanvar_test=False
+)
+
+# Generate forecasts
+forecast_df = client.generate_forecast(
+    file_path="path/to/your/forecast_data.xlsx",
+    id_col="Individual",
+    time_col="Time",
+    y="Loan Probability",
+    lagged_features={"Repayment Amount": 1, "Missed Payments": 2},
+    current_features=["Credit Score", "Unemployment Rate"],
+    filter_by_significance=True,
+    meanvar_test=False
+)
+
+# The forecast DataFrame includes the original data plus forecast columns:
+# - forecast: The predicted values
+# - forecast_lower: Lower bound of the prediction interval
+# - forecast_upper: Upper bound of the prediction interval
+# - forecast_error: Standard error of the forecast
+```
+
+## API Reference
+
+### DATFIDClient
+
+#### `__init__(token: str)`
+Initialize the client with your Hugging Face token.
+
+#### `fit_model(file_path: str, id_col: str, time_col: str, y: str, lagged_features: Optional[Dict[str, int]] = None, current_features: Optional[list] = None, filter_by_significance: bool = False, meanvar_test: bool = False) -> Dict[str, Any]`
+Fit a model using the provided data.
+
+#### `generate_forecast(file_path: str, id_col: str, time_col: str, y: str, lagged_features: Optional[Dict[str, int]] = None, current_features: Optional[list] = None, filter_by_significance: bool = False, meanvar_test: bool = False) -> pd.DataFrame`
+Generate forecasts using the fitted model.

datfid-0.1.12/README.md

@@ -0,0 +1,61 @@
+# DATFID SDK
+
+A Python SDK to access the DATFID API running on Hugging Face Spaces.
+
+## Installation
+
+```bash
+pip install datfid
+```
+
+## Usage
+
+```python
+from datfid import DATFIDClient
+
+# Initialize the client with your Hugging Face token
+client = DATFIDClient(token="your_huggingface_token")
+
+# Fit a model
+fit_result = client.fit_model(
+    file_path="path/to/your/data.xlsx",
+    id_col="Individual",
+    time_col="Time",
+    y="Loan Probability",
+    lagged_features={"Repayment Amount": 1, "Missed Payments": 2},
+    current_features=["Credit Score", "Unemployment Rate"],
+    filter_by_significance=True,
+    meanvar_test=False
+)
+
+# Generate forecasts
+forecast_df = client.generate_forecast(
+    file_path="path/to/your/forecast_data.xlsx",
+    id_col="Individual",
+    time_col="Time",
+    y="Loan Probability",
+    lagged_features={"Repayment Amount": 1, "Missed Payments": 2},
+    current_features=["Credit Score", "Unemployment Rate"],
+    filter_by_significance=True,
+    meanvar_test=False
+)
+
+# The forecast DataFrame includes the original data plus forecast columns:
+# - forecast: The predicted values
+# - forecast_lower: Lower bound of the prediction interval
+# - forecast_upper: Upper bound of the prediction interval
+# - forecast_error: Standard error of the forecast
+```
+
+## API Reference
+
+### DATFIDClient
+
+#### `__init__(token: str)`
+Initialize the client with your Hugging Face token.
+
+#### `fit_model(file_path: str, id_col: str, time_col: str, y: str, lagged_features: Optional[Dict[str, int]] = None, current_features: Optional[list] = None, filter_by_significance: bool = False, meanvar_test: bool = False) -> Dict[str, Any]`
+Fit a model using the provided data.
+
+#### `generate_forecast(file_path: str, id_col: str, time_col: str, y: str, lagged_features: Optional[Dict[str, int]] = None, current_features: Optional[list] = None, filter_by_significance: bool = False, meanvar_test: bool = False) -> pd.DataFrame`
+Generate forecasts using the fitted model.

datfid-0.1.12/datfid/__init__.py

@@ -0,0 +1 @@
+from .client import DATFIDClient

datfid-0.1.12/datfid/client.py

@@ -0,0 +1,146 @@
+import requests
+import pandas as pd
+from typing import Dict, Any, Optional, Union
+import json
+from types import SimpleNamespace
+import tempfile
+import os
+import gc
+import psutil
+import logging
+
+class DATFIDClient:
+    def __init__(self, token: str):
+        self.api_url = "https://datfid-org-datfid-sdk.hf.space/"
+        self.headers = {"Authorization": f"Bearer {token}"}
+        self.logger = logging.getLogger(__name__)
+
+    def _cleanup_memory(self):
+        """Clean up memory after operations"""
+        gc.collect()
+        if hasattr(psutil, 'Process'):
+            process = psutil.Process()
+            try:
+                process.memory_info().rss  # Force memory info update
+            except:
+                pass
+
+    def ping(self):
+        try:
+            response = requests.get(self.api_url, headers=self.headers).json()
+            self._cleanup_memory()
+            return response
+        except Exception as e:
+            self.logger.error(f"Ping failed: {str(e)}")
+            raise
+
+    def secure_ping(self):
+        try:
+            response = requests.get(f"{self.api_url}secure-ping/", headers=self.headers).json()
+            self._cleanup_memory()
+            return response
+        except Exception as e:
+            self.logger.error(f"Secure ping failed: {str(e)}")
+            raise
+
+    def fit_model(
+        self,
+        df: pd.DataFrame,
+        id_col: str,
+        time_col: str,
+        y: str,
+        lag_y: Optional[Union[int, str, list[int]]] = None,
+        lagged_features: Optional[Dict[str, int]] = None,
+        current_features: Optional[list] = None,
+        filter_by_significance: bool = False,
+        meanvar_test: bool = False
+    ) -> SimpleNamespace:
+        """
+        Fit a model using the DATFID API.
+        
+        Args:
+            df: DataFrame containing the data
+            id_col: Name of the ID column
+            time_col: Name of the time column
+            y: Name of the target variable
+            lagged_features: Dictionary of features and their lag values
+            current_features: List of current features to use
+            filter_by_significance: Whether to filter features by significance
+            meanvar_test: Whether to perform mean-variance test
+            
+        Returns:
+            SimpleNamespace containing the model fit results
+        """
+
+        df = df.copy()
+        for col in df.columns:
+            if pd.api.types.is_datetime64_any_dtype(df[col]):
+                df[col] = df[col].astype(str)
+
+        data = {
+            "df": df.to_dict(orient="records"),
+            "id_col": id_col,
+            "time_col": time_col,
+            "y": y,
+            "lag_y": lag_y,
+            "lagged_features": lagged_features or {},
+            "current_features": current_features or [],
+            "filter_by_significance": filter_by_significance,
+            "meanvar_test": meanvar_test
+        }
+        
+        response = requests.post(
+            f"{self.api_url}modelfit/",
+            json=data,
+            headers=self.headers
+        )
+        
+        if response.status_code != 200:
+            raise Exception(f"Model fit failed: {response.text}")
+            
+        result_dict = response.json()
+        return SimpleNamespace(**result_dict)
+
+    def forecast_model(
+        self,
+        df_forecast: pd.DataFrame
+    ) -> pd.DataFrame:
+        """
+        Generate forecasts using the fitted model.
+        
+        Args:
+            df_forecast: DataFrame containing the forecast data
+            
+        Returns:
+            DataFrame containing the forecast results
+        """
+
+        try:
+            df_forecast = df_forecast.copy()
+            for col in df_forecast.columns:
+                if pd.api.types.is_datetime64_any_dtype(df_forecast[col]):
+                    df_forecast[col] = df_forecast[col].astype(str)
+
+            # Convert DataFrame to list of records
+            data = df_forecast.to_dict(orient="records")
+            
+            response = requests.post(
+                f"{self.api_url}modelforecast/",
+                json=data,
+                headers=self.headers
+            )
+            
+            if response.status_code != 200:
+                raise Exception(f"Forecast generation failed: {response.text}")
+            
+            result = pd.DataFrame(response.json())
+            
+            # Clean up memory after operation
+            del df_forecast
+            del data
+            self._cleanup_memory()
+            
+            return result
+        except Exception as e:
+            self.logger.error(f"Forecast generation failed: {str(e)}")
+            raise

datfid-0.1.12/datfid.egg-info/PKG-INFO

@@ -0,0 +1,87 @@
+Metadata-Version: 2.4
+Name: datfid
+Version: 0.1.12
+Summary: SDK to access the DATFID API hosted on Hugging Face Spaces
+Author: DATFID
+Author-email: igor.schapiro@datfid.com
+License: MIT
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3
+Requires-Python: >=3.7
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: requests>=2.31.0
+Requires-Dist: pandas>=1.0.1
+Requires-Dist: numpy<2.1,>=1.22
+Dynamic: author
+Dynamic: author-email
+Dynamic: classifier
+Dynamic: description
+Dynamic: description-content-type
+Dynamic: license
+Dynamic: license-file
+Dynamic: requires-dist
+Dynamic: requires-python
+Dynamic: summary
+
+# DATFID SDK
+
+A Python SDK to access the DATFID API running on Hugging Face Spaces.
+
+## Installation
+
+```bash
+pip install datfid
+```
+
+## Usage
+
+```python
+from datfid import DATFIDClient
+
+# Initialize the client with your Hugging Face token
+client = DATFIDClient(token="your_huggingface_token")
+
+# Fit a model
+fit_result = client.fit_model(
+    file_path="path/to/your/data.xlsx",
+    id_col="Individual",
+    time_col="Time",
+    y="Loan Probability",
+    lagged_features={"Repayment Amount": 1, "Missed Payments": 2},
+    current_features=["Credit Score", "Unemployment Rate"],
+    filter_by_significance=True,
+    meanvar_test=False
+)
+
+# Generate forecasts
+forecast_df = client.generate_forecast(
+    file_path="path/to/your/forecast_data.xlsx",
+    id_col="Individual",
+    time_col="Time",
+    y="Loan Probability",
+    lagged_features={"Repayment Amount": 1, "Missed Payments": 2},
+    current_features=["Credit Score", "Unemployment Rate"],
+    filter_by_significance=True,
+    meanvar_test=False
+)
+
+# The forecast DataFrame includes the original data plus forecast columns:
+# - forecast: The predicted values
+# - forecast_lower: Lower bound of the prediction interval
+# - forecast_upper: Upper bound of the prediction interval
+# - forecast_error: Standard error of the forecast
+```
+
+## API Reference
+
+### DATFIDClient
+
+#### `__init__(token: str)`
+Initialize the client with your Hugging Face token.
+
+#### `fit_model(file_path: str, id_col: str, time_col: str, y: str, lagged_features: Optional[Dict[str, int]] = None, current_features: Optional[list] = None, filter_by_significance: bool = False, meanvar_test: bool = False) -> Dict[str, Any]`
+Fit a model using the provided data.
+
+#### `generate_forecast(file_path: str, id_col: str, time_col: str, y: str, lagged_features: Optional[Dict[str, int]] = None, current_features: Optional[list] = None, filter_by_significance: bool = False, meanvar_test: bool = False) -> pd.DataFrame`
+Generate forecasts using the fitted model.

datfid-0.1.12/datfid.egg-info/SOURCES.txt

@@ -0,0 +1,10 @@
+LICENSE
+README.md
+setup.py
+datfid/__init__.py
+datfid/client.py
+datfid.egg-info/PKG-INFO
+datfid.egg-info/SOURCES.txt
+datfid.egg-info/dependency_links.txt
+datfid.egg-info/requires.txt
+datfid.egg-info/top_level.txt

datfid-0.1.12/datfid.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

datfid-0.1.12/datfid.egg-info/requires.txt

@@ -0,0 +1,3 @@
+requests>=2.31.0
+pandas>=1.0.1
+numpy<2.1,>=1.22

datfid-0.1.12/datfid.egg-info/top_level.txt

@@ -0,0 +1 @@
+datfid

datfid-0.1.12/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

datfid-0.1.12/setup.py

@@ -0,0 +1,38 @@
+from setuptools import setup, find_packages
+
+# to build:
+# 1) open this file at level of datfid-sdk folder
+# 2) change version in this file
+# 3) delete folder datfid.egg-info
+# 4) delete older files from dist folder
+# 5) in terminal: python setup.py sdist bdist_wheel
+# 6) in terminal: twine upload --repository testpypi dist/*
+# 7) in hugging face delete older files from dist folder
+# 8) in hugging face upload updated files
+# 9) in terminal uninstall older version of datfid: pip uninstall datfid
+# 10) in terminal install new version of datfid: pip install --index-url https://test.pypi.org/simple/ datfid
+
+with open("README.md", "r", encoding="utf-8") as fh:
+    long_description = fh.read()
+
+setup(
+    name="datfid",
+    version="0.1.12",
+    description="SDK to access the DATFID API hosted on Hugging Face Spaces",
+    long_description=long_description,
+    long_description_content_type="text/markdown",  # Important!
+    author="DATFID",
+    author_email="igor.schapiro@datfid.com",
+    license="MIT",
+    packages=find_packages(),
+    install_requires=[
+        "requests>=2.31.0",
+        "pandas>=1.0.1",
+        "numpy>=1.22, <2.1"
+    ],
+    python_requires=">=3.7",
+    classifiers=[
+        "License :: OSI Approved :: MIT License",
+        "Programming Language :: Python :: 3",
+    ],
+)