tanml 0.1.6__tar.gz

tanml-0.1.6/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 Tanmay Sah and Dolly Sah
+
+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.

tanml-0.1.6/MANIFEST.in

@@ -0,0 +1,6 @@
+include README.md
+include LICENSE
+include pyproject.toml
+
+include tanml/report/templates/*.docx
+include tanml/config_templates/*.yaml

tanml-0.1.6/PKG-INFO

@@ -0,0 +1,317 @@
+Metadata-Version: 2.4
+Name: tanml
+Version: 0.1.6
+Summary: Automated validation toolkit for tabular ML models in finance and regulated domains.
+Author: Dolly Sah
+Author-email: Tanmay Sah <tradertanmay@gmail.com>
+License: MIT
+Project-URL: Homepage, https://github.com/tdlabs-ai/tanml
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: scikit-learn
+Requires-Dist: pandas
+Requires-Dist: matplotlib
+Requires-Dist: seaborn
+Requires-Dist: shap
+Requires-Dist: docxtpl
+Requires-Dist: python-docx
+Requires-Dist: docxcompose
+Requires-Dist: PyYAML
+Requires-Dist: scipy
+Requires-Dist: statsmodels
+Requires-Dist: joblib
+Requires-Dist: tzlocal
+Requires-Dist: tqdm
+Requires-Dist: imgkit
+Requires-Dist: xgboost
+Provides-Extra: dev
+Requires-Dist: pytest; extra == "dev"
+Requires-Dist: black; extra == "dev"
+Requires-Dist: isort; extra == "dev"
+Dynamic: license-file
+
+# TanML: Automated Model Validation Toolkit for Tabular Machine Learning  
+[![Cite this repo](https://img.shields.io/badge/Cite-this_repo-blue)](https://github.com/tdlabs-ai/tanml)
+
+TanML is a modular, automated model validation toolkit for tabular machine learning workflows. It supports end-to-end validation with just a YAML configuration and a single command, performing checks across data quality, robustness, explainability, and model performance.
+
+TanML generates structured Word (DOCX) reports suitable for internal model reviews, audit documentation, and stakeholder presentations. It is designed for general-purpose ML validation and works well in domains where interpretability, fairness, and reliability are critical.
+
+While TanML currently operates as a command-line toolkit, its architecture is designed to evolve into an intelligent validation agent—capable of integrating with AutoML pipelines, CI/CD workflows, and human-in-the-loop validation systems.
+
+## Key Features
+
+* One-command validation using CLI and YAML
+* Supports models developed in Python (scikit-learn, XGBoost), R, or SAS
+* Scenario-based flexibility:
+
+  * Scenario A: Single model and dataset
+  * Scenario B: One model per segment with separate datasets
+  * Scenario C: One model and dataset with an internal segmentation column
+* Comprehensive validation checks:
+
+  * Model performance (e.g., accuracy, AUC, KS)
+  * Data quality diagnostics
+  * Stress testing for input robustness
+  * SHAP-based explainability
+  * Segment-wise validation
+  * Logistic regression coefficient summaries (if applicable)
+  * VIF, correlation, and multicollinearity checks
+  * EDA summaries
+  * Rule-based threshold validation using YAML configuration
+* Professional report generation in Word (DOCX) format
+* Easily extensible architecture to add custom checks or outputs
+
+## Installation
+
+TanML can be installed directly from PyPI using pip:
+
+```bash
+pip install tanml
+```
+
+To upgrade to the latest version:
+
+```bash
+pip install --upgrade tanml
+```
+
+After installation, you can verify the CLI is working by running:
+
+```bash
+tanml --help
+```
+
+This will display the list of available commands and options.
+
+TanML supports Python 3.8 and above. It is recommended to use a virtual environment for clean dependency management:
+
+```bash
+python3 -m venv .venv
+source .venv/bin/activate  # On Windows: .venv\Scripts\activate
+```
+
+## Folder Structure
+
+A typical TanML project is organized as follows:
+
+```
+TanML/
+├── models/                # Trained model files (e.g., .pkl, .sas7bdat)
+├── data/
+│   ├── raw/               # Raw data files (optional)
+│   └── cleaned/           # Cleaned datasets for validation
+├── examples/
+│   ├── scenario_a/        # Example for single model and dataset
+│   ├── scenario_b/        # One model per segment
+│   └── scenario_c/        # Single model with segmentation column
+├── reports/
+│   ├── images/            # SHAP plots, cluster visualizations
+│   ├── clusters/          # Cluster summary CSVs
+│   └── *.docx             # Final validation reports
+├── tanml/                 # Core source code
+│   ├── cli/               # CLI logic
+│   ├── engine/            # Orchestration and segmentation logic
+│   ├── checks/            # Validation check classes
+│   ├── check_runners/     # Check runners linked to the engine
+│   ├── report/            # Report generation using docxtpl
+│   └── utils/             # Data/model loaders and helpers
+├── tests/                 # Unit tests (WIP or planned)
+├── setup.py               # Package installer for pip
+├── pyproject.toml         # Modern Python build metadata (PEP 518)
+├── MANIFEST.in            # Files to include when packaging
+└── README.md              # This documentation
+```
+
+## How to Use TanML
+
+To use TanML effectively, follow these three main steps. This process is designed to be simple and intuitive, even for users without a programming background.
+
+1. **Initialize the Configuration File:**
+   Start by generating a YAML configuration file that tells TanML how to run the validation. You do this using a command-line instruction where you select one of the three supported validation scenarios:
+
+   * Scenario A: For validating one model using a single dataset
+   * Scenario B: For validating multiple models, each with its own segment-specific dataset
+   * Scenario C: For validating a single model that needs to be run separately on different segments within one dataset
+
+   You can also choose where the generated YAML file should be saved. If you don’t provide a location, it will be saved in your current directory with a default name.
+
+2. **Fill in the Configuration File:**
+   Once the YAML file is created, you’ll open it and fill in the necessary details:
+
+   * Where your model file is located (e.g., a `.pkl` file for Python, or equivalent for SAS or R)
+   * Where your cleaned dataset is saved
+   * (Optional) Where your raw dataset is saved, if you want data quality comparisons
+   * Which input features your model expects, and what the target column is
+   * Where you want the final report to be saved (Word `.docx` format)
+
+   This YAML file acts like a blueprint—TanML reads it and follows the instructions you provide to run all relevant validation checks.
+
+3. **Run the Validation Process:**
+   After the YAML file is completed, you will run the validation process. This will trigger TanML to:
+
+   * Load your model and data
+   * Perform all configured validation checks (like performance, SHAP explainability, stress testing, etc.)
+   * Automatically generate a professional report with summaries, tables, and visuals
+
+You can use TanML either through simple command-line instructions or directly in Python by calling its functions. Both methods achieve the same results. The command-line approach is ideal for repeatable, scriptable workflows, while the Python interface is useful for advanced users who want to integrate TanML into larger systems or notebooks.
+
+TanML is controlled entirely through a YAML configuration file and a single CLI command. The configuration specifies the model, data, validation rules, and output paths.
+
+### Scenario A: Single Model, Single Dataset
+
+This is the simplest usage mode. You have one model file and one cleaned dataset.
+
+**rules.yaml**
+
+```yaml
+model:
+  features:
+    - age
+    - income
+    - debt_to_income
+    - credit_score
+    - employment_length
+  target: default_flag
+
+paths:
+  model: models/model.pkl
+  cleaned_data: data/cleaned/cleaned.csv
+  raw_data: data/raw/raw.csv
+
+output:
+  report_path: reports/validation_report.docx
+```
+
+**Run the validation:**
+
+```bash
+tanml validate --rules rules.yaml
+```
+
+This generates a `.docx` report along with SHAP plots and other artifacts.
+
+### Scenario B: One Model per Segment
+
+Use this if you have multiple segments, each with its own cleaned dataset and model.
+
+**rules.yaml**
+
+```yaml
+segment:
+  runs:
+    segment_A:
+      model: models/model_segment_A.pkl
+      cleaned: data/cleaned/segment_A.csv
+      raw: data/raw/segment_A.csv
+      output_report: reports/report_segment_A.docx
+
+    segment_B:
+      model: models/model_segment_B.pkl
+      cleaned: data/cleaned/segment_B.csv
+      raw: data/raw/segment_B.csv
+      output_report: reports/report_segment_B.docx
+
+model:
+  features:
+    - age
+    - income
+    - debt_to_income
+    - credit_score
+    - employment_length
+  target: default_flag
+```
+
+**Run the validation:**
+
+```bash
+tanml validate --rules rules.yaml
+```
+
+Each segment will be validated independently with its own output report.
+
+### Scenario C: One Model with Segmentation Column
+
+Use this if you have one dataset with a segmentation column (e.g., region, product type).
+
+**rules.yaml**
+
+```yaml
+segment:
+  column: segment_id  # This column will be used to split the data automatically
+
+model:
+  features:
+    - age
+    - income
+    - debt_to_income
+    - credit_score
+    - employment_length
+  target: default_flag
+
+paths:
+  model: models/credit_model.pkl
+  cleaned_data: data/cleaned/combined.csv
+  raw_data: data/raw/combined.csv
+
+output:
+  report_path: reports/report_{segment}.docx
+```
+
+**Run the validation:**
+
+```bash
+tanml validate --rules rules.yaml
+```
+
+This will automatically split the dataset by the `segment_id` column, apply the same model to each subset, and produce one report per segment.
+
+## Output Artifacts
+
+After a successful validation run, TanML generates a set of output files based on your configuration:
+
+- **Validation Report (.docx):**  
+  A professionally formatted Word document containing:
+  - A summary of all validation checks
+  - Tables for performance metrics, data quality, logistic regression coefficients, VIF, and more
+  - SHAP-based feature importance visualizations (if enabled)
+  - Segment-wise validation summaries (for Scenario B and C)
+
+- **SHAP Visualizations:**  
+  Summary bar plots and other SHAP outputs are saved in the `reports/images/` folder.
+
+- **Input Cluster Coverage Charts and CSVs:**  
+  If cluster coverage analysis is enabled, visual and tabular summaries are stored in:
+  - `reports/images/` (cluster bar plots)
+  - `reports/clusters/` (CSV summaries)
+
+- **Logs and Intermediate Results:**  
+  Optional debug or intermediate outputs (e.g., cleaned data snapshots or rule validation results) can be generated depending on configuration or verbosity level.
+
+By default, all outputs are saved to the paths you define in the YAML configuration. Each segment (in Scenario B or C) will generate its own report file if multiple runs are triggered.
+
+Extending TanML
+
+TanML is designed with modularity in mind. Advanced users and developers can easily extend its capabilities by adding new validation checks or modifying report components. Here's how extension works:
+
+Add a Custom Check:Create a new check class in the tanml/checks/ directory and implement the validation logic based on the BaseCheck interface.
+
+Create a Check Runner:Add a corresponding runner in tanml/check_runners/. This file controls how the check is executed and connected to the engine.
+
+Register the Check:Link your runner in the central registry inside the validation engine. The YAML config will trigger it automatically based on user-defined rules.
+
+This modular structure ensures that domain-specific validations (e.g., industry regulations, fairness audits) can be added without modifying core logic.
+
+## License and Citation
+
+TanML is open-source under the MIT License.  
+Copyright © 2025 Tanmay Sah and Dolly Sah.
+
+You are free to use, modify, and distribute this software with appropriate attribution.
+
+If you use TanML in your work or publication, please cite it as:
+
+> Sah, T., & Sah, D. (2025). *TanML: Automated Model Validation Toolkit for Tabular Machine Learning*. GitHub. https://github.com/tdlabs-ai/tanml
+
+📄 A machine-readable citation file (`CITATION.cff`) is included for use with citation tools and GitHub's “Cite this repository” button.

tanml-0.1.6/README.md

@@ -0,0 +1,284 @@
+# TanML: Automated Model Validation Toolkit for Tabular Machine Learning  
+[![Cite this repo](https://img.shields.io/badge/Cite-this_repo-blue)](https://github.com/tdlabs-ai/tanml)
+
+TanML is a modular, automated model validation toolkit for tabular machine learning workflows. It supports end-to-end validation with just a YAML configuration and a single command, performing checks across data quality, robustness, explainability, and model performance.
+
+TanML generates structured Word (DOCX) reports suitable for internal model reviews, audit documentation, and stakeholder presentations. It is designed for general-purpose ML validation and works well in domains where interpretability, fairness, and reliability are critical.
+
+While TanML currently operates as a command-line toolkit, its architecture is designed to evolve into an intelligent validation agent—capable of integrating with AutoML pipelines, CI/CD workflows, and human-in-the-loop validation systems.
+
+## Key Features
+
+* One-command validation using CLI and YAML
+* Supports models developed in Python (scikit-learn, XGBoost), R, or SAS
+* Scenario-based flexibility:
+
+  * Scenario A: Single model and dataset
+  * Scenario B: One model per segment with separate datasets
+  * Scenario C: One model and dataset with an internal segmentation column
+* Comprehensive validation checks:
+
+  * Model performance (e.g., accuracy, AUC, KS)
+  * Data quality diagnostics
+  * Stress testing for input robustness
+  * SHAP-based explainability
+  * Segment-wise validation
+  * Logistic regression coefficient summaries (if applicable)
+  * VIF, correlation, and multicollinearity checks
+  * EDA summaries
+  * Rule-based threshold validation using YAML configuration
+* Professional report generation in Word (DOCX) format
+* Easily extensible architecture to add custom checks or outputs
+
+## Installation
+
+TanML can be installed directly from PyPI using pip:
+
+```bash
+pip install tanml
+```
+
+To upgrade to the latest version:
+
+```bash
+pip install --upgrade tanml
+```
+
+After installation, you can verify the CLI is working by running:
+
+```bash
+tanml --help
+```
+
+This will display the list of available commands and options.
+
+TanML supports Python 3.8 and above. It is recommended to use a virtual environment for clean dependency management:
+
+```bash
+python3 -m venv .venv
+source .venv/bin/activate  # On Windows: .venv\Scripts\activate
+```
+
+## Folder Structure
+
+A typical TanML project is organized as follows:
+
+```
+TanML/
+├── models/                # Trained model files (e.g., .pkl, .sas7bdat)
+├── data/
+│   ├── raw/               # Raw data files (optional)
+│   └── cleaned/           # Cleaned datasets for validation
+├── examples/
+│   ├── scenario_a/        # Example for single model and dataset
+│   ├── scenario_b/        # One model per segment
+│   └── scenario_c/        # Single model with segmentation column
+├── reports/
+│   ├── images/            # SHAP plots, cluster visualizations
+│   ├── clusters/          # Cluster summary CSVs
+│   └── *.docx             # Final validation reports
+├── tanml/                 # Core source code
+│   ├── cli/               # CLI logic
+│   ├── engine/            # Orchestration and segmentation logic
+│   ├── checks/            # Validation check classes
+│   ├── check_runners/     # Check runners linked to the engine
+│   ├── report/            # Report generation using docxtpl
+│   └── utils/             # Data/model loaders and helpers
+├── tests/                 # Unit tests (WIP or planned)
+├── setup.py               # Package installer for pip
+├── pyproject.toml         # Modern Python build metadata (PEP 518)
+├── MANIFEST.in            # Files to include when packaging
+└── README.md              # This documentation
+```
+
+## How to Use TanML
+
+To use TanML effectively, follow these three main steps. This process is designed to be simple and intuitive, even for users without a programming background.
+
+1. **Initialize the Configuration File:**
+   Start by generating a YAML configuration file that tells TanML how to run the validation. You do this using a command-line instruction where you select one of the three supported validation scenarios:
+
+   * Scenario A: For validating one model using a single dataset
+   * Scenario B: For validating multiple models, each with its own segment-specific dataset
+   * Scenario C: For validating a single model that needs to be run separately on different segments within one dataset
+
+   You can also choose where the generated YAML file should be saved. If you don’t provide a location, it will be saved in your current directory with a default name.
+
+2. **Fill in the Configuration File:**
+   Once the YAML file is created, you’ll open it and fill in the necessary details:
+
+   * Where your model file is located (e.g., a `.pkl` file for Python, or equivalent for SAS or R)
+   * Where your cleaned dataset is saved
+   * (Optional) Where your raw dataset is saved, if you want data quality comparisons
+   * Which input features your model expects, and what the target column is
+   * Where you want the final report to be saved (Word `.docx` format)
+
+   This YAML file acts like a blueprint—TanML reads it and follows the instructions you provide to run all relevant validation checks.
+
+3. **Run the Validation Process:**
+   After the YAML file is completed, you will run the validation process. This will trigger TanML to:
+
+   * Load your model and data
+   * Perform all configured validation checks (like performance, SHAP explainability, stress testing, etc.)
+   * Automatically generate a professional report with summaries, tables, and visuals
+
+You can use TanML either through simple command-line instructions or directly in Python by calling its functions. Both methods achieve the same results. The command-line approach is ideal for repeatable, scriptable workflows, while the Python interface is useful for advanced users who want to integrate TanML into larger systems or notebooks.
+
+TanML is controlled entirely through a YAML configuration file and a single CLI command. The configuration specifies the model, data, validation rules, and output paths.
+
+### Scenario A: Single Model, Single Dataset
+
+This is the simplest usage mode. You have one model file and one cleaned dataset.
+
+**rules.yaml**
+
+```yaml
+model:
+  features:
+    - age
+    - income
+    - debt_to_income
+    - credit_score
+    - employment_length
+  target: default_flag
+
+paths:
+  model: models/model.pkl
+  cleaned_data: data/cleaned/cleaned.csv
+  raw_data: data/raw/raw.csv
+
+output:
+  report_path: reports/validation_report.docx
+```
+
+**Run the validation:**
+
+```bash
+tanml validate --rules rules.yaml
+```
+
+This generates a `.docx` report along with SHAP plots and other artifacts.
+
+### Scenario B: One Model per Segment
+
+Use this if you have multiple segments, each with its own cleaned dataset and model.
+
+**rules.yaml**
+
+```yaml
+segment:
+  runs:
+    segment_A:
+      model: models/model_segment_A.pkl
+      cleaned: data/cleaned/segment_A.csv
+      raw: data/raw/segment_A.csv
+      output_report: reports/report_segment_A.docx
+
+    segment_B:
+      model: models/model_segment_B.pkl
+      cleaned: data/cleaned/segment_B.csv
+      raw: data/raw/segment_B.csv
+      output_report: reports/report_segment_B.docx
+
+model:
+  features:
+    - age
+    - income
+    - debt_to_income
+    - credit_score
+    - employment_length
+  target: default_flag
+```
+
+**Run the validation:**
+
+```bash
+tanml validate --rules rules.yaml
+```
+
+Each segment will be validated independently with its own output report.
+
+### Scenario C: One Model with Segmentation Column
+
+Use this if you have one dataset with a segmentation column (e.g., region, product type).
+
+**rules.yaml**
+
+```yaml
+segment:
+  column: segment_id  # This column will be used to split the data automatically
+
+model:
+  features:
+    - age
+    - income
+    - debt_to_income
+    - credit_score
+    - employment_length
+  target: default_flag
+
+paths:
+  model: models/credit_model.pkl
+  cleaned_data: data/cleaned/combined.csv
+  raw_data: data/raw/combined.csv
+
+output:
+  report_path: reports/report_{segment}.docx
+```
+
+**Run the validation:**
+
+```bash
+tanml validate --rules rules.yaml
+```
+
+This will automatically split the dataset by the `segment_id` column, apply the same model to each subset, and produce one report per segment.
+
+## Output Artifacts
+
+After a successful validation run, TanML generates a set of output files based on your configuration:
+
+- **Validation Report (.docx):**  
+  A professionally formatted Word document containing:
+  - A summary of all validation checks
+  - Tables for performance metrics, data quality, logistic regression coefficients, VIF, and more
+  - SHAP-based feature importance visualizations (if enabled)
+  - Segment-wise validation summaries (for Scenario B and C)
+
+- **SHAP Visualizations:**  
+  Summary bar plots and other SHAP outputs are saved in the `reports/images/` folder.
+
+- **Input Cluster Coverage Charts and CSVs:**  
+  If cluster coverage analysis is enabled, visual and tabular summaries are stored in:
+  - `reports/images/` (cluster bar plots)
+  - `reports/clusters/` (CSV summaries)
+
+- **Logs and Intermediate Results:**  
+  Optional debug or intermediate outputs (e.g., cleaned data snapshots or rule validation results) can be generated depending on configuration or verbosity level.
+
+By default, all outputs are saved to the paths you define in the YAML configuration. Each segment (in Scenario B or C) will generate its own report file if multiple runs are triggered.
+
+Extending TanML
+
+TanML is designed with modularity in mind. Advanced users and developers can easily extend its capabilities by adding new validation checks or modifying report components. Here's how extension works:
+
+Add a Custom Check:Create a new check class in the tanml/checks/ directory and implement the validation logic based on the BaseCheck interface.
+
+Create a Check Runner:Add a corresponding runner in tanml/check_runners/. This file controls how the check is executed and connected to the engine.
+
+Register the Check:Link your runner in the central registry inside the validation engine. The YAML config will trigger it automatically based on user-defined rules.
+
+This modular structure ensures that domain-specific validations (e.g., industry regulations, fairness audits) can be added without modifying core logic.
+
+## License and Citation
+
+TanML is open-source under the MIT License.  
+Copyright © 2025 Tanmay Sah and Dolly Sah.
+
+You are free to use, modify, and distribute this software with appropriate attribution.
+
+If you use TanML in your work or publication, please cite it as:
+
+> Sah, T., & Sah, D. (2025). *TanML: Automated Model Validation Toolkit for Tabular Machine Learning*. GitHub. https://github.com/tdlabs-ai/tanml
+
+📄 A machine-readable citation file (`CITATION.cff`) is included for use with citation tools and GitHub's “Cite this repository” button.

tanml-0.1.6/pyproject.toml

@@ -0,0 +1,54 @@
+[build-system]
+requires = ["setuptools>=61.0", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "tanml"
+version = "0.1.6"
+description = "Automated validation toolkit for tabular ML models in finance and regulated domains."
+readme = "README.md"
+requires-python = ">=3.8"
+license = {text = "MIT"}
+
+authors = [
+  { name = "Tanmay Sah", email = "tradertanmay@gmail.com" },
+  { name = "Dolly Sah" }
+]
+
+dependencies = [
+  "scikit-learn",
+  "pandas",
+  "matplotlib",
+  "seaborn",
+  "shap",
+  "docxtpl",
+  "python-docx",
+  "docxcompose",
+  "PyYAML",
+  "scipy",
+  "statsmodels",
+  "joblib",
+  "tzlocal",
+  "tqdm",
+  "imgkit",
+  "xgboost"
+]
+
+[project.optional-dependencies]
+dev = ["pytest", "black", "isort"]
+
+[project.scripts]
+tanml = "tanml.cli.main:main"
+
+[project.urls]
+"Homepage" = "https://github.com/tdlabs-ai/tanml"
+
+[tool.setuptools]
+include-package-data = true
+
+[tool.setuptools.packages.find]
+include = ["tanml*"]
+
+[tool.setuptools.package-data]
+"tanml.report.templates" = ["report_template.docx"]
+"tanml.config_templates" = ["*.yaml"]

tanml-0.1.6/setup.cfg

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

tanml-0.1.6/tanml/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.1.1"