codes/rnn_training/train_nli_ray.py [233:296]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
    eval_acc = round(100 * correct / total_samples, 2)
    if final_eval:
        print("finalgrep : accuracy {0} : {1}".format(eval_type, eval_acc))
        # ex.log_metric('{}_accuracy'.format(eval_type), eval_acc, step=epoch)
    else:
        print(
            "togrep : results : epoch {0} ; mean accuracy {1} :\
              {2}".format(
                epoch, eval_type, eval_acc
            )
        )
        # ex.log_metric('{}_accuracy'.format(eval_type), eval_acc, step=epoch)

    if eval_type == "valid" and epoch <= params.n_epochs:
        if eval_acc > train_config["val_acc_best"]:
            print("saving model at epoch {0}".format(epoch))
            # if not os.path.exists(params.outputdir):
            #    os.makedirs(params.outputdir)
            torch.save(nli_net.state_dict(), params.outputmodelname)
            train_config["val_acc_best"] = eval_acc
        else:
            if "sgd" in params.optimizer:
                optimizer.param_groups[0]["lr"] = (
                    optimizer.param_groups[0]["lr"] / params.lrshrink
                )
                print(
                    "Shrinking lr by : {0}. New lr = {1}".format(
                        params.lrshrink, optimizer.param_groups[0]["lr"]
                    )
                )
                if optimizer.param_groups[0]["lr"] < params.minlr:
                    train_config["stop_training"] = True
            if "adam" in params.optimizer:
                # early stopping (at 2nd decrease in accuracy)
                train_config["stop_training"] = train_config["adam_stop"]
                # adam_stop = True

    return eval_acc, optimizer, train_config


@ray.remote(num_gpus=1)
def HyperEvaluate(config):
    print(config)
    parser = argparse.ArgumentParser()
    parser.add_argument("--node-ip-address=")  # ,192.168.2.19
    parser.add_argument("--node-manager-port=")
    parser.add_argument("--object-store-name=")
    parser.add_argument(
        "--raylet-name="
    )  # /tmp/ray/session_2020-07-15_12-00-45_292745_38156/sockets/raylet
    parser.add_argument("--redis-address=")  # 192.168.2.19:6379
    parser.add_argument("--config-list=", action="store_true")  #
    parser.add_argument("--temp-dir=")  # /tmp/ray
    parser.add_argument("--redis-password=")  # 5241590000000000
    # /////////NLI-Args//////////////
    parser = argparse.ArgumentParser(description="NLI training")
    # paths
    parser.add_argument(
        "--nlipath",
        type=str,
        default=config["dataset"],
        help="NLI data (SNLI or MultiNLI)",
    )
    parser.add_argument(
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -



codes/rnn_training/train_nli_w2v.py [231:294]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
    eval_acc = round(100 * correct / total_samples, 2)
    if final_eval:
        print("finalgrep : accuracy {0} : {1}".format(eval_type, eval_acc))
        # ex.log_metric('{}_accuracy'.format(eval_type), eval_acc, step=epoch)
    else:
        print(
            "togrep : results : epoch {0} ; mean accuracy {1} :\
              {2}".format(
                epoch, eval_type, eval_acc
            )
        )
        # ex.log_metric('{}_accuracy'.format(eval_type), eval_acc, step=epoch)

    if eval_type == "valid" and epoch <= params.n_epochs:
        if eval_acc > train_config["val_acc_best"]:
            print("saving model at epoch {0}".format(epoch))
            # if not os.path.exists(params.outputdir):
            #    os.makedirs(params.outputdir)
            torch.save(nli_net.state_dict(), params.outputmodelname)
            train_config["val_acc_best"] = eval_acc
        else:
            if "sgd" in params.optimizer:
                optimizer.param_groups[0]["lr"] = (
                    optimizer.param_groups[0]["lr"] / params.lrshrink
                )
                print(
                    "Shrinking lr by : {0}. New lr = {1}".format(
                        params.lrshrink, optimizer.param_groups[0]["lr"]
                    )
                )
                if optimizer.param_groups[0]["lr"] < params.minlr:
                    train_config["stop_training"] = True
            if "adam" in params.optimizer:
                # early stopping (at 2nd decrease in accuracy)
                train_config["stop_training"] = train_config["adam_stop"]
                # adam_stop = True

    return eval_acc, optimizer, train_config


@ray.remote(num_gpus=1)
def HyperEvaluate(config):
    print(config)
    parser = argparse.ArgumentParser()
    parser.add_argument("--node-ip-address=")  # ,192.168.2.19
    parser.add_argument("--node-manager-port=")
    parser.add_argument("--object-store-name=")
    parser.add_argument(
        "--raylet-name="
    )  # /tmp/ray/session_2020-07-15_12-00-45_292745_38156/sockets/raylet
    parser.add_argument("--redis-address=")  # 192.168.2.19:6379
    parser.add_argument("--config-list=", action="store_true")  #
    parser.add_argument("--temp-dir=")  # /tmp/ray
    parser.add_argument("--redis-password=")  # 5241590000000000
    # /////////NLI-Args//////////////
    parser = argparse.ArgumentParser(description="NLI training")
    # paths
    parser.add_argument(
        "--nlipath",
        type=str,
        default=config["dataset"],
        help="NLI data (SNLI or MultiNLI)",
    )
    parser.add_argument(
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -