机器学习大作业-用于训练监督学习（模仿学习）或强化学习代理的2048游戏api+源代码+文档说明

# 2048-api A 2048 game api for training supervised learning (imitation learning) or reinforcement learning agents # Code structure * [`game2048/`](game2048/): the main package. * [`game.py`](game2048/game.py): the core 2048 `Game` class. * [`agents.py`](game2048/agents.py): the `Agent` class with instances. * [`displays.py`](game2048/displays.py): the `Display` class with instances, to show the `Game` state. * [`expectimax/`](game2048/expectimax): a powerful ExpectiMax agent by [here](https://github.com/nneonneo/2048-ai). * [`explore.ipynb`](explore.ipynb): introduce how to use the `Agent`, `Display` and `Game`. * [`static/`](static/): frontend assets (based on Vue.js) for web app. * [`webapp.py`](webapp.py): run the web app (backend) demo. * [`evaluate.py`](evaluate.py): evaluate your self-defined agent. * main.py: 主函数，包括从训练集中载入数据，模型的搭建以及模型训练。 * model文件夹中存储训练时产生的多个版本的model的pkl文件。 * 由于traindata.csv大小为237MB，无法上传。 * 代码运行： * 想要训练模型，运行python main.py。 * 想要查看已训练模型结果，运行python evaluate.py。 * 想要查看其他版本的模型结果，可以在game2048/agents.py中修改加载模型的路径，再运行python evaluate.py即可。 # Requirements * code only tested on linux system (ubuntu 16.04) * Python 3 (Anaconda 3.6.3 specifically) with numpy and flask # To define your own agents ```python from game2048.agents import Agent class YourOwnAgent(Agent): def step(self): '''To define the agent's 1-step behavior given the `game`. You can find more instance in [`agents.py`](game2048/agents.py). :return direction: 0: left, 1: down, 2: right, 3: up ''' direction = some_function(self.game) return direction ``` # To compile the pre-defined ExpectiMax agent ```bash cd game2048/expectimax bash configure make ``` # To run the web app ```bash python webapp.py ```  # LICENSE The code is under Apache-2.0 License. # For EE369 students from SJTU only Please read course project [requirements](EE369.md) and [description](Project2048.pdf).