# AgentGPT: Remote Env Integrated Cloud RL Training
**W&B Humanoid-v5 Benchmark (via Internet):** [Weights & Biases Dashboard](https://wandb.ai/junhopark/agentgpt-beta)
---
## Overview
AgentGPT is a one-click, cloud-based platform for distributed reinforcement learning. It lets you easily host your environment simulators—either locally or in the cloud—and connect them to a central training job on AWS SageMaker. This enables efficient data collection and scalable multi-agent training using a GPT-based RL policy.
## Installation
```markdown
pip install agent-gpt-aws --upgrade
```
### Configuration
- **Config hyperparams & SageMaker:**
```bash
agent-gpt config --batch_size 256
agent-gpt config --role_arn arn:aws:iam::123456789012:role/AgentGPTSageMakerRole
```
- **List & Clear current configuration:**
```bash
agent-gpt list
agent-gpt clear
```
### Simulation
- **Run your environment (gym/unity/unreal, etc.) before training starts:**
```bash
agent-gpt simulate local
agent-gpt simulate cloud
```
### Training & Inference
- **Train a gpt model on AWS:**
```bash
agent-gpt train
```
- **Run agent gpt on AWS:**
```bash
agent-gpt infer
```
## Key Features
- **Cloud & Local Hosting:** Quickly deploy environments (Gym/Unity) with a single command.
- **Parallel Training:** Connect multiple simulators to one AWS SageMaker trainer.
- **Real-Time Inference:** Serve a GPT-based RL policy for instant decision-making.
- **Cost-Optimized:** Minimize expenses by centralizing training while keeping simulations local if needed.
- **Scalable GPT Support:** Train Actor (policy) and Critic (value) GPT models together using reverse transitions.
Raw data
{
"_id": null,
"home_page": "https://github.com/ccnets-team/agent-gpt",
"name": "agent-gpt-aws",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.7",
"maintainer_email": null,
"keywords": "agent gpt reinforcement-learning sagemaker",
"author": "JunHo Park",
"author_email": "junho@ccnets.org",
"download_url": "https://files.pythonhosted.org/packages/b5/b5/7ecbf71682b2d1d6634127211cff3d525b41c72dab6f38826c2e13fd4930/agent_gpt_aws-0.4.0.tar.gz",
"platform": null,
"description": "# AgentGPT: Remote Env Integrated Cloud RL Training\r\n\r\n**W&B Humanoid-v5 Benchmark (via Internet):** [Weights & Biases Dashboard](https://wandb.ai/junhopark/agentgpt-beta)\r\n\r\n\r\n---\r\n\r\n## Overview\r\n\r\nAgentGPT is a one-click, cloud-based platform for distributed reinforcement learning. It lets you easily host your environment simulators\u2014either locally or in the cloud\u2014and connect them to a central training job on AWS SageMaker. This enables efficient data collection and scalable multi-agent training using a GPT-based RL policy.\r\n\r\n## Installation\r\n\r\n```markdown\r\npip install agent-gpt-aws --upgrade\r\n```\r\n\r\n### Configuration\r\n\r\n- **Config hyperparams & SageMaker:**\r\n ```bash\r\n agent-gpt config --batch_size 256\r\n agent-gpt config --role_arn arn:aws:iam::123456789012:role/AgentGPTSageMakerRole\r\n ```\r\n- **List & Clear current configuration:**\r\n ```bash\r\n agent-gpt list\r\n agent-gpt clear\r\n ```\r\n\r\n### Simulation\r\n\r\n- **Run your environment (gym/unity/unreal, etc.) before training starts:** \r\n ```bash\r\n agent-gpt simulate local\r\n agent-gpt simulate cloud\r\n ```\r\n\r\n### Training & Inference\r\n\r\n- **Train a gpt model on AWS:**\r\n ```bash\r\n agent-gpt train\r\n ```\r\n\r\n- **Run agent gpt on AWS:**\r\n ```bash\r\n agent-gpt infer\r\n ```\r\n\r\n## Key Features\r\n\r\n- **Cloud & Local Hosting:** Quickly deploy environments (Gym/Unity) with a single command.\r\n- **Parallel Training:** Connect multiple simulators to one AWS SageMaker trainer.\r\n- **Real-Time Inference:** Serve a GPT-based RL policy for instant decision-making.\r\n- **Cost-Optimized:** Minimize expenses by centralizing training while keeping simulations local if needed.\r\n- **Scalable GPT Support:** Train Actor (policy) and Critic (value) GPT models together using reverse transitions.\r\n",
"bugtrack_url": null,
"license": "Dual Licensed (AGENT GPT COMMERCIAL LICENSE or GNU GPLv3)",
"summary": "AgentGPT CLI for training and inference on AWS SageMaker",
"version": "0.4.0",
"project_urls": {
"Homepage": "https://github.com/ccnets-team/agent-gpt"
},
"split_keywords": [
"agent",
"gpt",
"reinforcement-learning",
"sagemaker"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "d278205ca4b41e5b1f79611394a16c4f4344e7f546d506bf99c592f62b74244d",
"md5": "0f109664900c06230866a162e0d27b00",
"sha256": "7340f297e1b39622142f5b4033cb5bdb1e3000d56abfc5fcf6afa57686f4316c"
},
"downloads": -1,
"filename": "agent_gpt_aws-0.4.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "0f109664900c06230866a162e0d27b00",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.7",
"size": 51142,
"upload_time": "2025-03-03T20:19:34",
"upload_time_iso_8601": "2025-03-03T20:19:34.093968Z",
"url": "https://files.pythonhosted.org/packages/d2/78/205ca4b41e5b1f79611394a16c4f4344e7f546d506bf99c592f62b74244d/agent_gpt_aws-0.4.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "b5b57ecbf71682b2d1d6634127211cff3d525b41c72dab6f38826c2e13fd4930",
"md5": "a4e9d73e2ebcda6e5b835cc826f7d2c2",
"sha256": "b766f70eac3a0342e92e3552000c1ee3c6dea73fe8ce90b49185395382b41034"
},
"downloads": -1,
"filename": "agent_gpt_aws-0.4.0.tar.gz",
"has_sig": false,
"md5_digest": "a4e9d73e2ebcda6e5b835cc826f7d2c2",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.7",
"size": 44839,
"upload_time": "2025-03-03T20:19:35",
"upload_time_iso_8601": "2025-03-03T20:19:35.259036Z",
"url": "https://files.pythonhosted.org/packages/b5/b5/7ecbf71682b2d1d6634127211cff3d525b41c72dab6f38826c2e13fd4930/agent_gpt_aws-0.4.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-03-03 20:19:35",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "ccnets-team",
"github_project": "agent-gpt",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"requirements": [
{
"name": "msgpack",
"specs": [
[
">=",
"1.1.0"
]
]
},
{
"name": "uvicorn",
"specs": [
[
">=",
"0.34.0"
]
]
},
{
"name": "fastapi",
"specs": [
[
">=",
"0.115.7"
]
]
},
{
"name": "gymnasium",
"specs": [
[
">=",
"1.0.0"
]
]
}
],
"lcname": "agent-gpt-aws"
}
JSON
