train5.py

<<<<<<< HEAD:train5.py
#!/usr/bin/env python3
# -*- coding:utf-8 -*-
# Copyright (c) Megvii, Inc. and its affiliates.
import os
import moxing as mox
os.system("pip install loguru")
os.system("pip install thop")
os.system("pip install pycocotools")
os.system("pip install tensorboard")
os.system("pip install opencv_python")
os.system("pip install tqdm")
os.system("pip install ninja")
os.system("pip install tabulate")
os.system("pip install scikit-image")
os.system("pip install Pillow")
os.system("pip uninstall mmcv -y")
os.system("pip uninstall mmcv-full -y")
mox.file.copy('obs://chuanhaimangoking939/mmdetection/mmcv_full-1.4.8-cp37-cp37m-manylinux1_x86_64 .whl','/cache/mmcv_full-1.4.8-cp37-cp37m-manylinux1_x86_64.whl')
os.system('pip install /cache/mmcv_full-1.4.8-cp37-cp37m-manylinux1_x86_64.whl')
os.system("pip install mmcls")
import argparse
import random
import warnings
from loguru import logger
import torch
import torch.backends.cudnn as cudnn
from yolox.core import Trainer, launch
from yolox.utils import configure_nccl, configure_omp, get_num_devices
import os
from yolox.exp import Exp as MyExp
from yolox.models import EfficientNet
import moxing as mox
class Exp(MyExp):
    def __init__(self,output_dir):
        super(Exp, self).__init__()
        # self.data_dir="datasets/COCO/"
        self.data_dir = "/home/ma-user/modelarts/user-job-dir/model/datasets/COCO_playground/"
        self.output_dir = output_dir
        # 大模型，640，做一个大epoch的实验
        self.depth = 1
        self.width = 1
        size = 544
        lrd = 10
        self.act = "gelu"
        self.multiscale_range = 0
        self.warmup_lr = 1e-7
        self.max_epoch = 65
        self.warmup_epochs = 10
        self.no_aug_epochs = 15
        self.num_classes = 10
        self.min_lr_ratio = 0.01
        self.input_size = (size, size)
        self.test_size = (size, size)
        self.basic_lr_per_img = 0.01 / (64.0 * lrd)
        # 让最小学习率再小一点，可能能学到东西
        self.exp_name = "yolox_l_s{0}_lrd{1}_mp{2}w{3}n{4}_544-65-gelu".format(size, lrd, self.max_epoch,self.warmup_epochs, self.no_aug_epochs)

    def get_model(self):
        from yolox.utils import freeze_module
        model = super().get_model()
        return model

def make_parser():
    resume = False
    resum_name = "yolox_l_s640_lrd10_mp200w10n10_640-200epoch"
    parser = argparse.ArgumentParser("YOLOX train parser")
    parser.add_argument("-expn", "--experiment-name", type=str, default=None)
    parser.add_argument("-n", "--name", type=str, default=None, help="model name")

    # distributed
    parser.add_argument(
        "--dist-backend", default="nccl", type=str, help="distributed backend"
    )
    parser.add_argument(
        "--dist-url",
        default=None,
        type=str,
        help="url used to set up distributed training",
    )
    parser.add_argument("-b", "--batch-size", type=int, default=32, help="batch size")
    parser.add_argument(
        "-d", "--devices", type=int, default=1, help="device for training"
    )
    parser.add_argument(
        "-f",
        "--exp_file",
        default="exps/example/yolo_mango.py ",
        type=str,
        help="plz input your experiment description file",
    )

    if not resume:
        parser.add_argument("-c", "--ckpt", default="/home/ma-user/modelarts/user-job-dir/model/ckpt/yolox_l.ckpt", type=str, help="checkpoint file")
        parser.add_argument("--resume", default=False, action="store_true", help="resume training")
    else:
        model_best_path='obs://chuanhaimangoking939/yolox/ckpt5/'+resum_name+'/last_epoch_ckpt.pth'
        mox.file.copy(model_best_path,
                      '/home/ma-user/modelarts/user-job-dir/model/ckpt/last_epoch_ckpt.ckpt')
        parser.add_argument("-c", "--ckpt", default="/home/ma-user/modelarts/user-job-dir/model/ckpt/last_epoch_ckpt.ckpt",type=str, help="checkpoint file")
        parser.add_argument("--resume", default=True, action="store_true", help="resume training")
    parser.add_argument(
        "-e",
        "--start_epoch",
        default=None,
        type=int,
        help="resume training start epoch",
    )
    parser.add_argument(
        "--num_machines", default=1, type=int, help="num of node for training"
    )
    parser.add_argument(
        "--machine_rank", default=0, type=int, help="node rank for multi-node training"
    )
    parser.add_argument(
        "--fp16",
        dest="fp16",
        default=True,
        action="store_true",
        help="Adopting mix precision training.",
    )
    parser.add_argument(
        "--cache",
        dest="cache",
        default=True,
        action="store_true",
        help="Caching imgs to RAM for fast training.",
    )
    parser.add_argument(
        "-o",
        "--occupy",
        dest="occupy",
        default=True,
        action="store_true",
        help="occupy GPU memory first for training.",
    )
    parser.add_argument(
        "-l",
        "--logger",
        type=str,
        help="Logger to be used for metrics",
        default="tensorboard"
    )
    parser.add_argument(
        "opts",
        help="Modify config options using the command-line",
        default=None,
        nargs=argparse.REMAINDER,
    )
    parser.add_argument("--model",type=str,default="",help='the path model saved')

    return parser

@logger.catch
def main(exp, args):
    if exp.seed is not None:
        random.seed(exp.seed)
        torch.manual_seed(exp.seed)
        cudnn.deterministic = True
        warnings.warn(
            "You have chosen to seed training. This will turn on the CUDNN deterministic setting, "
            "which can slow down your training considerably! You may see unexpected behavior "
            "when restarting from checkpoints."
        )
    # set environment variables for distributed training
    configure_nccl()
    configure_omp()
    cudnn.benchmark = True

    trainer = Trainer(exp, args)
    trainer.train()
    torch.nn.ReLU

if __name__ == "__main__":
    args = make_parser().parse_args()
    print("save model is "+str(args.model))
    exp=Exp(output_dir=args.model)
    exp.merge(args.opts)

    if not args.experiment_name:
        args.experiment_name = exp.exp_name

    num_gpu = get_num_devices() if args.devices is None else args.devices
    assert num_gpu <= get_num_devices()

    dist_url = "auto" if args.dist_url is None else args.dist_url
    launch(
        main,
        num_gpu,
        args.num_machines,
        args.machine_rank,
        backend=args.dist_backend,
        dist_url=dist_url,
        args=(exp, args),
    )