- Isolated torch-related code

- Removed componentDisabled

src/colvar.cpp

@@ -894,9 +894,7 @@ void colvar::define_component_types()
 
   add_component_type<neuralNetwork>("neural network CV for other CVs", "neuralNetwork");
 
-#ifdef COLVARS_TORCH
   add_component_type<torchANN>("CV defined by PyTorch artifical neural network models", "torchANN");
-#endif
 
   if (proxy->check_volmaps_available() == COLVARS_OK) {
     add_component_type<map_total>("total value of atomic map", "mapTotal");

src/colvar.h

@@ -586,9 +586,6 @@ class colvar : public colvarparse, public colvardeps {
   // collective variable component base class
   class cvc;
 
-  // placeholder/stub for unavailable functionality
-  class componentDisabled;
-
   // list of available collective variable components
 
   // scalar colvar components

src/colvarcomp.cpp

@@ -98,7 +98,8 @@ int colvar::cvc::init(std::string const &conf)
   if (period != 0.0) {
     if (!is_available(f_cvc_periodic)) {
       error_code |=
-          cvm::error("Error: invalid use of period and/or wrapAround in a \"" +
+          cvm::error("Error: invalid use of period and/or "
+                     "wrapAround in a \"" +
                          function_type() + "\" component.\n" + "Period: " + cvm::to_str(period) +
                          " wrapAround: " + cvm::to_str(wrap_center),
                      COLVARS_INPUT_ERROR);
@@ -706,20 +707,6 @@ void colvar::cvc::wrap(colvarvalue &x_unwrapped) const
 }
 
 
-
-colvar::componentDisabled::componentDisabled() {}
-
-colvar::componentDisabled::~componentDisabled() {}
-
-int colvar::componentDisabled::init(std::string const & /* conf */)
-{
-  return cvm::error("Error: components of type " + function_type() +
-                        " are not enabled in the current build",
-                    COLVARS_NOT_IMPLEMENTED);
-}
-
-
-
 // Static members
 
 std::vector<colvardeps::feature *> colvar::cvc::cvc_features;

src/colvarcomp.h

@@ -26,11 +26,6 @@
 #include "colvar.h"
 #include "colvar_geometricpath.h"
 
-#ifdef COLVARS_TORCH
-#include <torch/torch.h>
-#include <torch/script.h>
-#endif
-
 
 /// \brief Colvar component (base class for collective variables)
 ///
@@ -320,16 +315,6 @@ inline colvarvalue const & colvar::cvc::Jacobian_derivative() const
 }
 
 
-/// \brief Colvar component class for a feature not currently available
-class colvar::componentDisabled
-  : public colvar::cvc
-{
-public:
-  componentDisabled();
-  virtual ~componentDisabled();
-  int init(std::string const & /* conf */);
-};
-
 
 /// \brief Colvar component: distance between the centers of mass of
 /// two groups (colvarvalue::type_scalar type, range [0:*))
@@ -1547,40 +1532,6 @@ class colvar::neuralNetwork
   virtual void wrap(colvarvalue &x_unwrapped) const;
 };
 
-#ifdef COLVARS_TORCH
-// only when LibTorch is available
-class colvar::torchANN
-  : public colvar::linearCombination
-{
-protected:
-    torch::jit::script::Module nn;
-    /// the index of nn output component
-    size_t m_output_index;
-    bool use_double_input;
-    //bool use_gpu;
-    // 1d tensor, concatenation of values of sub-cvcs
-    torch::Tensor input_tensor;
-    torch::Tensor nn_outputs;
-    torch::Tensor input_grad;
-    // record the initial index of of sub-cvcs in input_tensor
-    std::vector<int> cvc_indices;
-public:
-    torchANN();
-    virtual ~torchANN();
-    virtual int init(std::string const &conf);
-    virtual void calc_value();
-    virtual void calc_gradients();
-    virtual void apply_force(colvarvalue const &force);
-};
-#else
-class colvar::torchANN
-  : public colvar::componentDisabled
-{
-public:
-    torchANN();
-    virtual ~torchANN();
-};
-#endif // COLVARS_TORCH checking
 
 // \brief Colvar component: total value of a scalar map
 // (usually implemented as a grid by the simulation engine)

src/colvarcomp_torchann.cpp

@@ -13,6 +13,7 @@
 #include "colvarparse.h"
 #include "colvarvalue.h"
 
+#include "colvarcomp_torchann.h"
 
 #ifdef COLVARS_TORCH
 
@@ -202,14 +203,21 @@ void colvar::torchANN::apply_force(colvarvalue const &force) {
 
 #else
 
-
 colvar::torchANN::torchANN()
 {
   set_function_type("torchANN");
 }
 
-
 colvar::torchANN::~torchANN() {}
 
+int colvar::torchANN::init(std::string const &conf) {
+
+  return cvm::error(
+          "torchANN requires the libtorch library, but it is not enabled during compilation.\n"
+          "Please refer to the Compilation Notes section of the Colvars manual for more "
+          "information.\n",
+          COLVARS_NOT_IMPLEMENTED);
+
+}
 
 #endif

src/colvarcomp_torchann.h

@@ -0,0 +1,63 @@
+// -*- c++ -*-
+
+// This file is part of the Collective Variables module (Colvars).
+// The original version of Colvars and its updates are located at:
+// https://github.com/Colvars/colvars
+// Please update all Colvars source files before making any changes.
+// If you wish to distribute your changes, please submit them to the
+// Colvars repository at GitHub.
+//
+#ifndef COLVARCOMP_TORCH_H
+#define COLVARCOMP_TORCH_H
+
+// Declaration of torchann 
+
+#include <memory>
+
+#include "colvar.h"
+#include "colvarcomp.h"
+#include "colvarmodule.h"
+
+#ifdef COLVARS_TORCH 
+
+#include <torch/torch.h>
+#include <torch/script.h>
+
+class colvar::torchANN
+  : public colvar::linearCombination
+{
+protected:
+    torch::jit::script::Module nn;
+    /// the index of nn output component
+    size_t m_output_index;
+    bool use_double_input;
+    //bool use_gpu;
+    // 1d tensor, concatenation of values of sub-cvcs
+    torch::Tensor input_tensor;
+    torch::Tensor nn_outputs;
+    torch::Tensor input_grad;
+    // record the initial index of of sub-cvcs in input_tensor
+    std::vector<int> cvc_indices;
+public:
+    torchANN();
+    virtual ~torchANN();
+    virtual int init(std::string const &conf);
+    virtual void calc_value();
+    virtual void calc_gradients();
+    virtual void apply_force(colvarvalue const &force);
+};
+
+#else
+
+class colvar::torchANN
+  : public colvar::cvc
+{
+public:
+    torchANN();
+    virtual ~torchANN();
+    virtual int init(std::string const &conf);
+};
+#endif // COLVARS_TORCH checking
+
+#endif 
+