Explain calibration and calibration cache

aidge_export_tensorrt/static/tensorrt_8.6/python_binding/pybind_export.cpp

@@ -59,8 +59,10 @@ void init_Graph(py::module& m)
 
         .def("calibrate", &Graph::calibrate, py::arg("calibration_folder_path") = "./calibration_folder/", py::arg("cache_file_path") = "./calibration_cache", py::arg("batch_size") = 1, 
             R"mydelimiter(
-        Calibrate the graph using the calibration data found inside the `calibration` folder.
+        Calibrate the graph to determine the appropriate scaling factors for converting floating-point values to lower-precision representations, using the calibration data found inside the specified `calibration_folder`.
 
+        Calibration can be expensive, so it is beneficial to generate the calibration data once and then reuse it for subsequent builds of the network. The cache includes the regression cutoff and quantile values used to generate it, and will not be used if these do not match the settings of the current calibrator. However, the network should be recalibrated if its structure changes or if the input data set changes, and it is the responsibility of the application to ensure this.
+        
         :param calibration_folder_path: Path to the calibration folder.
         :type calibration_folder_path: str
         :param cache_file_path: Path to the cache file.