6cd0cd3bd99ec2df47bb571a18e0b70819134504,rllib/agents/impala/impala.py,OverrideDefaultResourceRequest,default_resource_request,#Any#Any#,101

Before Change

    def default_resource_request(cls, config):
        cf = dict(cls._default_config, **config)
        Trainer._validate_config(cf)
        return Resources(
            cpu=cf["num_cpus_for_driver"],
            gpu=cf["num_gpus"],
            memory=cf["memory"],
            object_store_memory=cf["object_store_memory"],
            extra_cpu=cf["num_cpus_per_worker"] * cf["num_workers"] +
            cf["num_aggregation_workers"],
            extra_gpu=cf["num_gpus_per_worker"] * cf["num_workers"],
            extra_memory=cf["memory_per_worker"] * cf["num_workers"],
            extra_object_store_memory=cf["object_store_memory_per_worker"] *
            cf["num_workers"])


def make_learner_thread(local_worker, config):
    if config["num_gpus"] > 1 or config["num_data_loader_buffers"] > 1:

After Change

                    "CPU": cf["num_cpus_per_worker"],
                    "GPU": cf["num_gpus_per_worker"],
                } for _ in range(cf["num_workers"])
            ] + ([
                {
                    // Evaluation workers (+1 b/c of the additional local
                    // worker)
                    "CPU": eval_config.get("num_cpus_per_worker",
                                           cf["num_cpus_per_worker"]),
                    "GPU": eval_config.get("num_gpus_per_worker",
                                           cf["num_gpus_per_worker"]),
                } for _ in range(cf["evaluation_num_workers"] + 1)
            ] if cf["evaluation_interval"] else []),
            strategy=config.get("placement_strategy", "PACK"))

Instances