relayrl_framework 0.5.0-beta.2

A system-oriented multi-agent reinforcement learning framework.
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

{
    "inference_server_config": {
        "config_update_polling_seconds": 10.0,
        "transport_config": {
            "nats_addresses": {
                "inference_server": {
                    "host": "127.0.0.1",
                    "port": "50050"
                },
                "training_server": {
                    "host": "127.0.0.1",
                    "port": "50051"
                }
            },
            "zmq_addresses": {
                "inference_addresses": {
                    "inference_server": {
                        "host": "127.0.0.1",
                        "port": "7800"
                    },
                    "inference_scaling_server": {
                        "host": "127.0.0.1",
                        "port": "7801"
                    }
                },
                "training_addresses": {
                    "model_server": {
                        "host": "127.0.0.1",
                        "port": "50051"
                    },
                    "trajectory_server": {
                        "host": "127.0.0.1",
                        "port": "7776"
                    },
                    "agent_listener": {
                        "host": "127.0.0.1",
                        "port": "7777"
                    },
                    "training_scaling_server": {
                        "host": "127.0.0.1",
                        "port": "7778"
                    }
                }
            },
            "local_model_module": {
                "directory": "model_module",
                "model_name": "client_model",
                "format": "pt"
            },
            "max_traj_length": 1000
        }
    }
}