package diplomacy_proto

Get desktop application:
View/edit binary Protocol Buffers messages

• repeated bytes value = 1

Used in: SavedGame

• repeated float value = 1

• repeated int64 value = 1

• repeated int32 value = 1

• map<string, StringList> value = 1

• string sender = 1

  Power sending the message or SYSTEM

• string recipient = 2

  Power the message or GLOBAL

• int64 time_sent = 3

  Timestamp where the message was sent

• string phase = 4

  Phase name (e.g. 'S1901M')

• string message = 5

  Actual message

• repeated uint32 tokens = 6

  Token representation of message with BOM and EOM

Used in: SavedGame

• string name = 1

• optional State state = 2

• map<string, StringList> orders = 3

• map<string, StringList> results = 4

• map<string, PhaseHistory.PolicyDetails> policy = 6

• repeated uint32 prev_orders_state = 8

  Flatten representation of prev_orders_state (M phases)

• map<string, float> state_value = 9

  Value of this state for each power

• map<string, StringList> possible_orders = 7

  List of possible orders for each power

Used in: PhaseHistory

• repeated string locs = 1

  List of locs where orders were issued (e.g. BRE, WAIVE_0)

• repeated int32 tokens = 2

  List of tokens outputted by the model

• repeated float log_probs = 3

  List of log probs for each token outputted

• bool draw_action = 4

  Whether player wanted to accept a draw or not

• float draw_prob = 5

  The model's probability of doing accepting a draw

• string id = 1

• string map = 2

• repeated string rules = 3

• repeated PhaseHistory phases = 4

• string done_reason = 6

  Reason why game was ended.

• repeated string assigned_powers = 5

  List of assigned powers (RL player should be at position 0)

• repeated string players = 11

  List of player names (same length as assigned_powers)

• map<string, SavedGame.KeywordArgs> kwargs = 10

  Settings used by each power.

• bool is_partial_game = 12

  Indicates that this is a partial game

• int32 start_phase_ix = 9

  For partial traj, the ix of the first phase to use for learning.

• string reward_fn = 7

  Name of the reward function used to compute rewards

• map<string, FloatList> rewards = 8

  List of rewards (no penalty) for each transition for each power

• map<string, FloatList> returns = 13

  List of returns for each transition for each power

Used in: SavedGame

• int32 player_seed = 1

  The player seed used by this power

• float noise = 2

  The noise level used by this power

• float temperature = 3

  The temperature used by this power

• float dropout_rate = 5

  The dropout rate used by this power

Used in: PhaseHistory

• string game_id = 1

  The game id

• string name = 2

  Phase name (e.g. 'S1901M')

• string map = 3

  Name of the map played (e.g. 'standard')

• string zobrist_hash = 13

  Unique hash of the game state

• repeated string rules = 4

  List of rules

• map<string, StringList> units = 5

  Dict of units for each power

• map<string, StringList> centers = 6

  Dict of centers for each power

• map<string, StringList> homes = 7

  Dict of homes for each power

• map<string, StringList> influence = 8

  Dict of influence for each power

• map<string, int32> civil_disorder = 9

  Dict of civil disorder for each power

• map<string, State.Builds> builds = 10

  Dict of build counts and build locs for each power

• string note = 11

  Note to be displayed on the rendered map

• repeated uint32 board_state = 12

  Flatten representation of board_state

Used in: State

• int32 count = 1

• repeated string homes = 2

Common proto classes

Used in: MapStringList, PhaseHistory, State

• repeated string value = 1