fix: function stubs for compilation without AI surrogate

README.md

@@ -99,8 +99,9 @@ following available options:
 - **POET_PREPROCESS_BENCHS**=*boolean* - enables the preprocessing of
   predefined models/benchmarks. Defaults to *ON*.
 - **USE_AI_SURROGATE**=*boolean* - includes the functions of the AI
-  surrogate model. This relies on the presence of a Python environment
-  where Keras is installed.
+  surrogate model. When active, CMake relies on `find_package()` to find
+  an a implementation of `Threads` and a Python environment where Numpy
+  and Keras need to be installed. Defaults to _OFF_.
 
 
 ### Example: Build from scratch
@@ -235,7 +236,10 @@ mpirun -n 4 ./poet --dht --dht-snaps=2 barite_het_rt.R barite_het.rds output
 
 ### Example: Preparing Environment and Running with AI surrogate
 
-To run the AI surrogate, you need to install the R package `keras3`. The
+To run the AI surrogate, you need to have a Keras installed in your Python environment.
+The implementation in POET is agnostic to the exact Keras version, but If you use a 
+pretrained model, the model file must match your Keras version. Using Keras 3 with 
+`.keras` model files is recommended. recomenden
 compilation process of POET remains the same as shown above.
 
 In the following code block, the installation process on the Turing Cluster is

src/Chemistry/SurrogateModels/AI_functions.hpp

@@ -14,13 +14,15 @@
 #ifndef AI_FUNCTIONS_H
 #define AI_FUNCTIONS_H
 
+#include <string>
+#include <vector>
+
 // PhreeqC definition of pi clashes with Eigen macros so we have to temporarily undef it 
 #pragma push_macro("pi")
 #undef pi
 #include <Eigen/Dense> 
 #pragma pop_macro("pi")
 
-#include <string>
 
 namespace poet {
 // Define an aligned allocator for std::vector
@@ -48,8 +50,6 @@ int Python_Keras_load_model(std::string model_file_path);
 
 std::vector<double> Python_Keras_predict(std::vector<std::vector<double>> x, int batch_size);  
 
-void Python_Keras_train(std::vector<std::vector<double>> x, std::vector<std::vector<double>> y, int batch_size);
-
 int Python_Keras_training_thread(EigenModel* Eigen_model,
                                  std::mutex* Eigen_model_mutex,
                                  TrainingData* training_data_buffer,
@@ -68,21 +68,16 @@ std::vector<double> Eigen_predict(const EigenModel& model, std::vector<std::vect
 // Otherwise, define the necessary stubs
 #else
 inline void Python_Keras_setup(std::string functions_file_path){}
-inline void Python_finalize(){};
+inline void Python_finalize(){}
 inline void Python_Keras_load_model(std::string model_file_path){}
 inline std::vector<double> Python_Keras_predict(std::vector<std::vector<double>>, int){return {};}
-inline void Python_Keras_train(vector<std::vector<double>>, vector<std::vector<double>>, int){}
 inline int Python_Keras_training_thread(EigenModel*, std::mutex*, 
                                         TrainingData*, std::mutex*,
                                         std::condition_variable*, bool*, 
                                         int, int, int, bool){return {};}
 inline void Python_Keras_set_weights_as_Eigen(EigenModel&){}
-inline EigenModel transform_weights(const std::vector<std::vector<std::vector<double>>>){return {};}
-inline std::vector<std::vector<std::vector<double>>> Python_Keras_get_weights(){return {};}
 inline std::vector<double> Eigen_predict(const EigenModel&, std::vector<std::vector<double>>, int){return {};}
-
-
 #endif
-
 } // namespace poet
+
 #endif // AI_FUNCTIONS_HPP