gcastle_corl

CORL

Package	gCastle
Version	1.0.3
Language	Python
Docs
Paper	Wang et al.[1]
Graph type	DAG
Docker	bpimages/gcastle:1.0.3
Module folder	gcastle_corl

Description

A RL- and order-based algorithm that improves the efficiency and scalability of previous RL-based approach.

Example JSON

[
  {
    "id": "gcastle_corl",
    "batch_size": 64,
    "input_dim": 100,
    "embed_dim": 256,
    "normalize": false,
    "encoder_name": "transformer",
    "encoder_heads": 8,
    "encoder_blocks": 3,
    "encoder_dropout_rate": 0.1,
    "decoder_name": "lstm",
    "reward_mode": "episodic",
    "reward_score_type": "BIC",
    "reward_regression_type": "LR",
    "reward_gpr_alpha": 1.0,
    "iteration": 10,
    "actor_lr": "1e-4",
    "critic_lr": "1e-3",
    "alpha": 0.99,
    "init_baseline": -1.0,
    "random_seed": 0,
    "device_type": "cpu",
    "device_ids": 0,
    "timeout": null
  }
]