File size: 13,294 Bytes
fc70cd1 47be66b fc70cd1 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 |
---
language: en
library_name: pytorch
license: apache-2.0
pipeline_tag: reinforcement-learning
tags:
- reinforcement-learning
- Generative Model
- GenerativeRL
- LunarLanderContinuous-v2
benchmark_name: Box2d
task_name: LunarLanderContinuous-v2
model-index:
- name: QGPO
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: LunarLanderContinuous-v2
type: LunarLanderContinuous-v2
metrics:
- type: mean_reward
value: '200.0'
name: mean_reward
verified: false
---
# Play **LunarLanderContinuous-v2** with **QGPO** Policy
## Model Description
<!-- Provide a longer summary of what this model is. -->
This implementation applies **QGPO** to the Box2d **LunarLanderContinuous-v2** environment using [GenerativeRL](https://github.com/opendilab/di-engine).
## Model Usage
### Install the Dependencies
<details close>
<summary>(Click for Details)</summary>
```shell
# install GenerativeRL with huggingface support
pip3 install GenerativeRL[huggingface]
# install environment dependencies if needed
pip3 install gym[box2d]==0.23.1
```
</details>
### Download Model from Huggingface and Run the Model
<details close>
<summary>(Click for Details)</summary>
```shell
# running with trained model
python3 -u run.py
```
**run.py**
```python
import gym
from grl.algorithms.qgpo import QGPOAlgorithm
from grl.datasets import QGPOCustomizedTensorDictDataset
from grl.utils.huggingface import pull_model_from_hub
def qgpo_pipeline():
policy_state_dict, config = pull_model_from_hub(
repo_id="OpenDILabCommunity/LunarLanderContinuous-v2-QGPO",
)
qgpo = QGPOAlgorithm(
config,
dataset=QGPOCustomizedTensorDictDataset(
numpy_data_path="./data.npz",
action_augment_num=config.train.parameter.action_augment_num,
),
)
qgpo.model.load_state_dict(policy_state_dict)
# ---------------------------------------
# Customized train code ↓
# ---------------------------------------
# qgpo.train()
# ---------------------------------------
# Customized train code ↑
# ---------------------------------------
# ---------------------------------------
# Customized deploy code ↓
# ---------------------------------------
agent = qgpo.deploy()
env = gym.make(config.deploy.env.env_id)
observation = env.reset()
images = [env.render(mode="rgb_array")]
for _ in range(config.deploy.num_deploy_steps):
observation, reward, done, _ = env.step(agent.act(observation))
image = env.render(mode="rgb_array")
images.append(image)
# save images into mp4 files
import imageio.v3 as imageio
import numpy as np
images = np.array(images)
imageio.imwrite("replay.mp4", images, fps=30, quality=8)
# ---------------------------------------
# Customized deploy code ↑
# ---------------------------------------
if __name__ == "__main__":
qgpo_pipeline()
```
</details>
## Model Training
### Train the Model and Push to Huggingface_hub
<details close>
<summary>(Click for Details)</summary>
```shell
#Training Your Own Agent
python3 -u train.py
```
**train.py**
```python
import gym
from grl.algorithms.qgpo import QGPOAlgorithm
from grl.datasets import QGPOCustomizedTensorDictDataset
from grl.utils.log import log
from grl_pipelines.diffusion_model.configurations.lunarlander_continuous_qgpo import (
config,
)
def qgpo_pipeline(config):
qgpo = QGPOAlgorithm(
config,
dataset=QGPOCustomizedTensorDictDataset(
numpy_data_path="./data.npz",
action_augment_num=config.train.parameter.action_augment_num,
),
)
# ---------------------------------------
# Customized train code ↓
# ---------------------------------------
qgpo.train()
# ---------------------------------------
# Customized train code ↑
# ---------------------------------------
# ---------------------------------------
# Customized deploy code ↓
# ---------------------------------------
agent = qgpo.deploy()
env = gym.make(config.deploy.env.env_id)
observation = env.reset()
for _ in range(config.deploy.num_deploy_steps):
env.render()
observation, reward, done, _ = env.step(agent.act(observation))
# ---------------------------------------
# Customized deploy code ↑
# ---------------------------------------
if __name__ == "__main__":
log.info("config: \n{}".format(config))
qgpo_pipeline(config)
```
</details>
**Configuration**
<details close>
<summary>(Click for Details)</summary>
```python
{'train': {'project': 'LunarLanderContinuous-v2-QGPO-VPSDE', 'device': 'cuda', 'wandb': {'project': 'IQL-LunarLanderContinuous-v2-QGPO-VPSDE'}, 'simulator': {'type': 'GymEnvSimulator', 'args': {'env_id': 'LunarLanderContinuous-v2'}}, 'model': {'QGPOPolicy': {'device': 'cuda', 'critic': {'device': 'cuda', 'q_alpha': 1.0, 'DoubleQNetwork': {'backbone': {'type': 'ConcatenateMLP', 'args': {'hidden_sizes': [10, 256, 256], 'output_size': 1, 'activation': 'relu'}}}}, 'diffusion_model': {'device': 'cuda', 'x_size': 2, 'alpha': 1.0, 'solver': {'type': 'DPMSolver', 'args': {'order': 2, 'device': 'cuda', 'steps': 17}}, 'path': {'type': 'linear_vp_sde', 'beta_0': 0.1, 'beta_1': 20.0}, 'reverse_path': {'type': 'linear_vp_sde', 'beta_0': 0.1, 'beta_1': 20.0}, 'model': {'type': 'noise_function', 'args': {'t_encoder': {'type': 'GaussianFourierProjectionTimeEncoder', 'args': {'embed_dim': 32, 'scale': 30.0}}, 'backbone': {'type': 'TemporalSpatialResidualNet', 'args': {'hidden_sizes': [512, 256, 128], 'output_dim': 2, 't_dim': 32, 'condition_dim': 8, 'condition_hidden_dim': 32, 't_condition_hidden_dim': 128}}}}, 'energy_guidance': {'t_encoder': {'type': 'GaussianFourierProjectionTimeEncoder', 'args': {'embed_dim': 32, 'scale': 30.0}}, 'backbone': {'type': 'ConcatenateMLP', 'args': {'hidden_sizes': [42, 256, 256], 'output_size': 1, 'activation': 'silu'}}}}}}, 'parameter': {'behaviour_policy': {'batch_size': 1024, 'learning_rate': 0.0001, 'epochs': 500}, 'action_augment_num': 16, 'fake_data_t_span': None, 'energy_guided_policy': {'batch_size': 256}, 'critic': {'stop_training_epochs': 500, 'learning_rate': 0.0001, 'discount_factor': 0.99, 'update_momentum': 0.005}, 'energy_guidance': {'epochs': 1000, 'learning_rate': 0.0001}, 'evaluation': {'evaluation_interval': 50, 'guidance_scale': [0.0, 1.0, 2.0]}, 'checkpoint_path': './LunarLanderContinuous-v2-QGPO'}}, 'deploy': {'device': 'cuda', 'env': {'env_id': 'LunarLanderContinuous-v2', 'seed': 0}, 'num_deploy_steps': 1000, 't_span': None}}
```
```json
{
"train": {
"project": "LunarLanderContinuous-v2-QGPO-VPSDE",
"device": "cuda",
"wandb": {
"project": "IQL-LunarLanderContinuous-v2-QGPO-VPSDE"
},
"simulator": {
"type": "GymEnvSimulator",
"args": {
"env_id": "LunarLanderContinuous-v2"
}
},
"model": {
"QGPOPolicy": {
"device": "cuda",
"critic": {
"device": "cuda",
"q_alpha": 1.0,
"DoubleQNetwork": {
"backbone": {
"type": "ConcatenateMLP",
"args": {
"hidden_sizes": [
10,
256,
256
],
"output_size": 1,
"activation": "relu"
}
}
}
},
"diffusion_model": {
"device": "cuda",
"x_size": 2,
"alpha": 1.0,
"solver": {
"type": "DPMSolver",
"args": {
"order": 2,
"device": "cuda",
"steps": 17
}
},
"path": {
"type": "linear_vp_sde",
"beta_0": 0.1,
"beta_1": 20.0
},
"reverse_path": {
"type": "linear_vp_sde",
"beta_0": 0.1,
"beta_1": 20.0
},
"model": {
"type": "noise_function",
"args": {
"t_encoder": {
"type": "GaussianFourierProjectionTimeEncoder",
"args": {
"embed_dim": 32,
"scale": 30.0
}
},
"backbone": {
"type": "TemporalSpatialResidualNet",
"args": {
"hidden_sizes": [
512,
256,
128
],
"output_dim": 2,
"t_dim": 32,
"condition_dim": 8,
"condition_hidden_dim": 32,
"t_condition_hidden_dim": 128
}
}
}
},
"energy_guidance": {
"t_encoder": {
"type": "GaussianFourierProjectionTimeEncoder",
"args": {
"embed_dim": 32,
"scale": 30.0
}
},
"backbone": {
"type": "ConcatenateMLP",
"args": {
"hidden_sizes": [
42,
256,
256
],
"output_size": 1,
"activation": "silu"
}
}
}
}
}
},
"parameter": {
"behaviour_policy": {
"batch_size": 1024,
"learning_rate": 0.0001,
"epochs": 500
},
"action_augment_num": 16,
"fake_data_t_span": null,
"energy_guided_policy": {
"batch_size": 256
},
"critic": {
"stop_training_epochs": 500,
"learning_rate": 0.0001,
"discount_factor": 0.99,
"update_momentum": 0.005
},
"energy_guidance": {
"epochs": 1000,
"learning_rate": 0.0001
},
"evaluation": {
"evaluation_interval": 50,
"guidance_scale": [
0.0,
1.0,
2.0
]
},
"checkpoint_path": "./LunarLanderContinuous-v2-QGPO"
}
},
"deploy": {
"device": "cuda",
"env": {
"env_id": "LunarLanderContinuous-v2",
"seed": 0
},
"num_deploy_steps": 1000,
"t_span": null
}
}
```
</details>
**Training Procedure**
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
- **Weights & Biases (wandb):** [monitor link](https://wandb.ai/zjowowen/IQL-LunarLanderContinuous-v2-QGPO-VPSDE)
## Model Information
<!-- Provide the basic links for the model. -->
- **Github Repository:** [repo link](https://github.com/opendilab/GenerativeRL/)
- **Doc**: [Algorithm link](https://opendilab.github.io/GenerativeRL/)
- **Configuration:** [config link](https://huggingface.co/OpenDILabCommunity/LunarLanderContinuous-v2-QGPO/blob/main/policy_config.json)
- **Demo:** [video](https://huggingface.co/OpenDILabCommunity/LunarLanderContinuous-v2-QGPO/blob/main/replay.mp4)
<!-- Provide the size information for the model. -->
- **Parameters total size:** 8799.79 KB
- **Last Update Date:** 2024-12-04
## Environments
<!-- Address questions around what environment the model is intended to be trained and deployed at, including the necessary information needed to be provided for future users. -->
- **Benchmark:** Box2d
- **Task:** LunarLanderContinuous-v2
- **Gym version:** 0.23.1
- **GenerativeRL version:** v0.0.1
- **PyTorch version:** 2.4.1+cu121
- **Doc**: [Environments link](https://www.gymlibrary.dev/environments/box2d/lunar_lander/)
|