updated

Author: ChernovAndrey

README.md

@@ -1,7 +1,7 @@
 # FAIM SDK
 
 [![PyPI version](https://badge.fury.io/py/faim-sdk.svg)](https://badge.fury.io/py/faim-sdk)
-[![Python 3.10+](https://img.shields.io/badge/python-3.10+-blue.svg)](https://www.python.org/downloads/)
+[![Python 3.11+](https://img.shields.io/badge/python-3.11+-blue.svg)](https://www.python.org/downloads/)
 [![License: Apache 2.0](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
 
 Production-ready Python SDK for FAIM (Foundation AI Models) - a high-performance time-series forecasting platform powered by foundation models.
@@ -315,9 +315,9 @@ See the `examples/` directory for complete Jupyter notebook examples:
 
 ## Requirements
 
-- Python >= 3.10
-- numpy >= 1.20.0
-- pyarrow >= 10.0.0
+- Python >= 3.11
+- numpy >= 1.26.0
+- pyarrow >= 11.0.0
 - httpx >= 0.23.0
 - pydantic >= 2.0.0
 
@@ -359,4 +359,4 @@ If you use FAIM in your research, please cite:
   year = {2024},
   url = {https://github.com/your-org/faim-sdk}
 }
-```
+```

examples/flowstate_simple_example.ipynb

@@ -50,6 +50,15 @@
 ForecastResponse:
     Contains prediction outputs (point, quantiles, samples) and metadata.
 
+Evaluation Tools:
+    The faim_sdk.eval subpackage provides metrics and visualization:
+    - Metrics: mse, mase, crps_from_quantiles
+    - Visualization: plot_forecast (requires: pip install faim-sdk[viz])
+
+    >>> from faim_sdk.eval import mse, plot_forecast
+    >>> mse_score = mse(test_data, response.point)
+    >>> fig, ax = plot_forecast(train_data[0], response.point[0], test_data[0])
+
 Exceptions:
     - AuthenticationError: API key issues (401, 403)
     - ValidationError: Invalid request parameters (422)
@@ -133,4 +142,4 @@
     "ConfigurationError",
 ]
 
-__version__ = "0.2.0"
+__version__ = "0.2.1"