train/inference/run_classifier_multi_label_infer.py

"""
  This script provides an example to wrap TencentPretrain for multi-label classification inference.
"""
import sys
import os
import torch
import argparse
import collections
import torch.nn as nn

tencentpretrain_dir = os.path.abspath(os.path.join(os.path.dirname(__file__), ".."))
sys.path.append(tencentpretrain_dir)

from tencentpretrain.utils.constants import *
from tencentpretrain.utils import *
from tencentpretrain.utils.config import load_hyperparam
from tencentpretrain.utils.seed import set_seed
from tencentpretrain.model_loader import load_model
from tencentpretrain.opts import infer_opts, tokenizer_opts
from finetune.run_classifier_multi_label import MultilabelClassifier
from inference.run_classifier_infer import read_dataset
from inference.run_classifier_infer import batch_loader


def main():
    parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFormatter)

    infer_opts(parser)

    parser.add_argument("--labels_num", type=int, required=True,
                        help="Number of prediction labels.")

    tokenizer_opts(parser)

    parser.add_argument("--output_logits", action="store_true", help="Write logits to output file.")
    parser.add_argument("--output_prob", action="store_true", help="Write probabilities to output file.")

    args = parser.parse_args()

    # Load the hyperparameters from the config file.
    args = load_hyperparam(args)

    # Build tokenizer.
    args.tokenizer = str2tokenizer[args.tokenizer](args)

    # Build classification model and load parameters.
    args.soft_targets, args.soft_alpha = False, False
    model = MultilabelClassifier(args)
    model = load_model(model, args.load_model_path)

    # For simplicity, we use DataParallel wrapper to use multiple GPUs.
    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
    model = model.to(device)
    if torch.cuda.device_count() > 1:
        print("{} GPUs are available. Let's use them.".format(torch.cuda.device_count()))
        model = torch.nn.DataParallel(model)

    dataset = read_dataset(args, args.test_path)

    src = torch.LongTensor([sample[0] for sample in dataset])
    seg = torch.LongTensor([sample[1] for sample in dataset])

    batch_size = args.batch_size
    instances_num = src.size()[0]

    print("The number of prediction instances: ", instances_num)

    model.eval()

    with open(args.prediction_path, mode="w", encoding="utf-8") as f:
        f.write("label")
        if args.output_logits:
            f.write("\t" + "logits")
        if args.output_prob:
            f.write("\t" + "prob")
        f.write("\n")
        for i, (src_batch, seg_batch) in enumerate(batch_loader(batch_size, src, seg)):
            src_batch = src_batch.to(device)
            seg_batch = seg_batch.to(device)
            with torch.no_grad():
                _, logits = model(src_batch, None, seg_batch)

            prob = nn.Sigmoid()(logits)
            prob = prob.cpu().numpy().tolist()
            logits = logits.cpu().numpy().tolist()

            for i, p in enumerate(prob):
                label = list()
                for j in range(len(p)):
                    if p[j] > 0.5:
                        label.append(str(j))
                f.write(",".join(label))
                if args.output_logits:
                    f.write("\t" + " ".join([str(v) for v in logits[i]]))
                if args.output_prob:
                    f.write("\t" + " ".join([str(v) for v in p]))
                f.write("\n")


if __name__ == "__main__":
    main()