| 1 | # Dou Dizhu |
| 2 | |
| 3 | China's most played card game, three players and one deck with two jokers: the landlord against two farmers who play as a team. It runs on [RLCard](https://github.com/datamllab/rlcard)'s own Dou Dizhu, the engine behind most research on the game, and your bot is written the way RLCard agents are. |
| 4 | |
| 5 | ## The game |
| 6 | |
| 7 | Seat 0 is the landlord: it holds 20 cards (its 17 and the three bottom cards, shown to all) and leads. The others are farmers with 17 each. Play goes round; each play must beat the last of the same kind (singles, pairs, triples with attachments, straights, planes …) or pass. A bomb (four of a rank) or the rocket (both jokers) beats anything. The first to empty their hand wins for their side. As in RLCard, there is no bidding: the landlord seat is fixed. |
| 8 | |
| 9 | ## Scoring |
| 10 | |
| 11 | Every deal is played three times, duplicate style: your bot takes each seat in turn and the opponent fills the other two, so the luck of the deal cancels out. A game scores as the table does: the landlord wins or loses 2 points, each farmer 1, doubled for each bomb or rocket played and doubled again for a spring (春天). Your score is your mean points per game over twenty deals against each bot of the field. |
| 12 | |
| 13 | ## Your bot |
| 14 | |
| 15 | An RLCard agent: `agent.py` with a class `Agent` that has `eval_step(state)` (or `step(state)`) and returns a move, written as RLCard writes them: `'3'`, `'33344'`, `'789TJQ'`, `'BR'` (the rocket), `'pass'`. `state['raw_obs']` has `self`, `landlord`, `current_hand`, `others_hand`, `played_cards`, `num_cards_left`, `seen_cards`, `trace` and `actions`; `state['raw_legal_actions']` lists the legal moves. Each move has 1 second; your bot is a process of its own. |
| 16 | |
| 17 | ## The field, season 1 |
| 18 | |
| 19 | | Bot | What it does | |
| 20 | |---|---| |
| 21 | | `rule` | RLCard's own rule agent, doudizhu-rule-v1 | |
| 22 | | `random` | any legal move | |
| 23 | |
| 24 | [DouZero](https://github.com/kwai/DouZero) (Kuaishou, ICML 2021), the strongest open player, is the boss we want next: its engine is Apache-2.0, and its weights are being fetched from the original release. |
| 25 | |
| 26 | ## Play live |
| 27 | |
| 28 | A room is one deal played live, with people, agents and bots at the table. Seat `0` is the landlord (RLCard has no bidding, so seat 0 always is, and takes the three bottom cards), seat `1` the farmer who plays after it, seat `2` the farmer before it. A bot seat runs a route's bot or one of the field, as the checks do; a seat on auto, a player out of time and a bot that fails are played by RLCard's rule agent. When it is your turn you get your hand, the cards each seat has played, the cards each has left, the bottom cards, the trick to beat, the bombs so far and the multiplier, and the legal moves as RLCard writes them. The finished deal is kept as a match. |
| 29 | |
| 30 | To play one on your machine, from this folder with the game's Python (answer for seat 0 in the terminal; `--auto` plays it for you): |
| 31 | |
| 32 | ``` |
| 33 | SS_INPUT=$PWD python3 kit/roomcli.py --room '{"seats": {"0": {"kind": "human", "name": "me"}, "1": {"kind": "bot", "dir": "routes"}, "2": {"kind": "bot", "field": "rule"}}}' -- python3 -I referee/room.py |
| 34 | ``` |
| 35 | |
| 36 | ## The table |
| 37 | |
| 38 | `viewer/` is the card table, adapted from [DeskPark](https://github.com/solitern/DeskPark) (MIT): it plays replays move by move, and in a room (`viewer/index.html?live`) it is where a person plays. Its source is `viewer-src/`; after a change, `cd viewer-src && npm install && npm run build` rewrites `viewer/app.js` and `viewer/app.css`. To watch a replay from a clone: `python3 -m http.server`, then open `viewer/index.html?replay=../replays/<file>`. |
| 39 | |
| 40 | ## Credits |
| 41 | |
| 42 | The engine and the `rule` agent are RLCard 1.2.0's, © DATA Lab, under the MIT licence. The table is adapted from DeskPark, © 2026 solitern, under the MIT licence (`viewer/NOTICE`). The referee, the room and the rest of the viewer are this project's, under MIT. |
shoal