#include <SavedPolicy.hh>
Public Member Functions | |
| virtual int | first_action (const std::vector< float > &s) |
| virtual void | last_action (float r) |
| void | loadPolicy (const char *filename) |
| virtual int | next_action (float r, const std::vector< float > &s) |
| SavedPolicy (int numactions, const char *filename) | |
| virtual void | seedExp (std::vector< experience >) |
| virtual void | setDebug (bool d) |
| virtual | ~SavedPolicy () |
Protected Types | |
| typedef const std::vector < float > * | state_t |
Protected Member Functions | |
| state_t | canonicalize (const std::vector< float > &s) |
| void | printState (const std::vector< float > &s) |
Private Attributes | |
| bool | ACTDEBUG |
| bool | LOADDEBUG |
| bool | loaded |
| const int | numactions |
| std::map< state_t, std::vector < float > > | Q |
| std::set< std::vector< float > > | statespace |
Detailed Description
Agent that uses a saved policy from a file.
Definition at line 16 of file SavedPolicy.hh.
Member Typedef Documentation
typedef const std::vector<float>* SavedPolicy::state_t [protected] |
The implementation maps all sensations to a set of canonical pointers, which serve as the internal representation of environment state.
Definition at line 37 of file SavedPolicy.hh.
Constructor & Destructor Documentation
| SavedPolicy::SavedPolicy | ( | int | numactions, |
| const char * | filename | ||
| ) |
Standard constructor
- Parameters:
-
numactions The number of possible actions
Definition at line 4 of file SavedPolicy.cc.
| SavedPolicy::~SavedPolicy | ( | ) | [virtual] |
Definition at line 17 of file SavedPolicy.cc.
Member Function Documentation
| SavedPolicy::state_t SavedPolicy::canonicalize | ( | const std::vector< float > & | s | ) | [protected] |
Produces a canonical representation of the given sensation.
- Parameters:
-
s The current sensation from the environment.
- Returns:
- A pointer to an equivalent state in statespace.
Definition at line 86 of file SavedPolicy.cc.
| int SavedPolicy::first_action | ( | const std::vector< float > & | s | ) | [virtual] |
Implements Agent.
Definition at line 19 of file SavedPolicy.cc.
| void SavedPolicy::last_action | ( | float | r | ) | [virtual] |
Implements Agent.
Definition at line 78 of file SavedPolicy.cc.
| void SavedPolicy::loadPolicy | ( | const char * | filename | ) |
Definition at line 119 of file SavedPolicy.cc.
| int SavedPolicy::next_action | ( | float | r, |
| const std::vector< float > & | s | ||
| ) | [virtual] |
Implements Agent.
Definition at line 48 of file SavedPolicy.cc.
| void SavedPolicy::printState | ( | const std::vector< float > & | s | ) | [protected] |
Definition at line 106 of file SavedPolicy.cc.
| void SavedPolicy::seedExp | ( | std::vector< experience > | seeds | ) | [virtual] |
Reimplemented from Agent.
Definition at line 114 of file SavedPolicy.cc.
| virtual void SavedPolicy::setDebug | ( | bool | d | ) | [inline, virtual] |
Implements Agent.
Definition at line 28 of file SavedPolicy.hh.
Member Data Documentation
bool SavedPolicy::ACTDEBUG [private] |
Definition at line 60 of file SavedPolicy.hh.
bool SavedPolicy::LOADDEBUG [private] |
Definition at line 61 of file SavedPolicy.hh.
bool SavedPolicy::loaded [private] |
Definition at line 62 of file SavedPolicy.hh.
const int SavedPolicy::numactions [private] |
Definition at line 58 of file SavedPolicy.hh.
std::map<state_t, std::vector<float> > SavedPolicy::Q [private] |
The primary data structure of the learning algorithm, the value function Q. For state_t s and int a, Q[s][a] gives the learned maximum expected future discounted reward conditional on executing action a in state s.
Definition at line 56 of file SavedPolicy.hh.
std::set<std::vector<float> > SavedPolicy::statespace [private] |
Set of all distinct sensations seen. Pointers to elements of this set serve as the internal representation of the environment state.
Definition at line 50 of file SavedPolicy.hh.
The documentation for this class was generated from the following files: