download.py 10.2 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
#   Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

from __future__ import absolute_import
from __future__ import division
from __future__ import print_function

import os
import sys
import os.path as osp
import shutil
import requests
import hashlib
L
LielinJiang 已提交
25 26
import tarfile
import zipfile
27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
import time
from collections import OrderedDict

try:
    from tqdm import tqdm
except:

    class tqdm(object):
        def __init__(self, total=None):
            self.total = total
            self.n = 0

        def update(self, n):
            self.n += n
            if self.total is None:
                sys.stderr.write("\r{0:.1f} bytes".format(self.n))
            else:
                sys.stderr.write("\r{0:.1f}%".format(100 * self.n / float(
                    self.total)))
            sys.stderr.flush()

        def __enter__(self):
            return self

        def __exit__(self, exc_type, exc_val, exc_tb):
            sys.stderr.write('\n')


import logging
logger = logging.getLogger(__name__)

Z
zhiboniu 已提交
58
__all__ = ['get_weights_path_from_url']
59

60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87
WEIGHTS_HOME = osp.expanduser("~/.cache/paddle/hapi/weights")

DOWNLOAD_RETRY_LIMIT = 3


def is_url(path):
    """
    Whether path is URL.
    Args:
        path (string): URL string or not.
    """
    return path.startswith('http://') or path.startswith('https://')


def get_weights_path_from_url(url, md5sum=None):
    """Get weights path from WEIGHT_HOME, if not exists,
    download it from url.

    Args:
        url (str): download url
        md5sum (str): md5 sum of download package
    
    Returns:
        str: a local path to save downloaded weights.

    Examples:
        .. code-block:: python

L
LielinJiang 已提交
88
            from paddle.utils.download import get_weights_path_from_url
89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104

            resnet18_pretrained_weight_url = 'https://paddle-hapi.bj.bcebos.com/models/resnet18.pdparams'
            local_weight_path = get_weights_path_from_url(resnet18_pretrained_weight_url)

    """
    path = get_path_from_url(url, WEIGHTS_HOME, md5sum)
    return path


def _map_path(url, root_dir):
    # parse path after download under root_dir
    fname = osp.split(url)[-1]
    fpath = fname
    return osp.join(root_dir, fpath)


105 106 107 108 109 110 111 112 113 114 115 116 117 118 119
def _get_unique_endpoints(trainer_endpoints):
    # Sorting is to avoid different environmental variables for each card
    trainer_endpoints.sort()
    ips = set()
    unique_endpoints = set()
    for endpoint in trainer_endpoints:
        ip = endpoint.split(":")[0]
        if ip in ips:
            continue
        ips.add(ip)
        unique_endpoints.add(endpoint)
    logger.info("unique_endpoints {}".format(unique_endpoints))
    return unique_endpoints


120 121 122 123 124
def get_path_from_url(url,
                      root_dir,
                      md5sum=None,
                      check_exist=True,
                      decompress=True):
125 126 127 128 129 130 131 132 133 134 135 136 137 138
    """ Download from given url to root_dir.
    if file or directory specified by url is exists under
    root_dir, return the path directly, otherwise download
    from url and decompress it, return the path.

    Args:
        url (str): download url
        root_dir (str): root dir for downloading, it should be
                        WEIGHTS_HOME or DATASET_HOME
        md5sum (str): md5 sum of download package
    
    Returns:
        str: a local path to save downloaded models & weights & datasets.
    """
139 140 141

    from paddle.fluid.dygraph.parallel import ParallelEnv

142 143 144
    assert is_url(url), "downloading from {} not a url".format(url)
    # parse path after download to decompress under root_dir
    fullpath = _map_path(url, root_dir)
145 146 147 148
    # Mainly used to solve the problem of downloading data from different 
    # machines in the case of multiple machines. Different ips will download 
    # data, and the same ip will only download data once.
    unique_endpoints = _get_unique_endpoints(ParallelEnv().trainer_endpoints[:])
149 150 151
    if osp.exists(fullpath) and check_exist and _md5check(fullpath, md5sum):
        logger.info("Found {}".format(fullpath))
    else:
152
        if ParallelEnv().current_endpoint in unique_endpoints:
153 154 155 156
            fullpath = _download(url, root_dir, md5sum)
        else:
            while not os.path.exists(fullpath):
                time.sleep(1)
L
LielinJiang 已提交
157

158
    if ParallelEnv().current_endpoint in unique_endpoints:
159 160
        if decompress and (tarfile.is_tarfile(fullpath) or
                           zipfile.is_zipfile(fullpath)):
L
LielinJiang 已提交
161 162
            fullpath = _decompress(fullpath)

163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229
    return fullpath


def _download(url, path, md5sum=None):
    """
    Download from url, save to path.

    url (str): download url
    path (str): download to given path
    """
    if not osp.exists(path):
        os.makedirs(path)

    fname = osp.split(url)[-1]
    fullname = osp.join(path, fname)
    retry_cnt = 0

    while not (osp.exists(fullname) and _md5check(fullname, md5sum)):
        if retry_cnt < DOWNLOAD_RETRY_LIMIT:
            retry_cnt += 1
        else:
            raise RuntimeError("Download from {} failed. "
                               "Retry limit reached".format(url))

        logger.info("Downloading {} from {}".format(fname, url))

        req = requests.get(url, stream=True)
        if req.status_code != 200:
            raise RuntimeError("Downloading from {} failed with code "
                               "{}!".format(url, req.status_code))

        # For protecting download interupted, download to
        # tmp_fullname firstly, move tmp_fullname to fullname
        # after download finished
        tmp_fullname = fullname + "_tmp"
        total_size = req.headers.get('content-length')
        with open(tmp_fullname, 'wb') as f:
            if total_size:
                with tqdm(total=(int(total_size) + 1023) // 1024) as pbar:
                    for chunk in req.iter_content(chunk_size=1024):
                        f.write(chunk)
                        pbar.update(1)
            else:
                for chunk in req.iter_content(chunk_size=1024):
                    if chunk:
                        f.write(chunk)
        shutil.move(tmp_fullname, fullname)

    return fullname


def _md5check(fullname, md5sum=None):
    if md5sum is None:
        return True

    logger.info("File {} md5 checking...".format(fullname))
    md5 = hashlib.md5()
    with open(fullname, 'rb') as f:
        for chunk in iter(lambda: f.read(4096), b""):
            md5.update(chunk)
    calc_md5sum = md5.hexdigest()

    if calc_md5sum != md5sum:
        logger.info("File {} md5 check failed, {}(calc) != "
                    "{}(base)".format(fullname, calc_md5sum, md5sum))
        return False
    return True
L
LielinJiang 已提交
230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322


def _decompress(fname):
    """
    Decompress for zip and tar file
    """
    logger.info("Decompressing {}...".format(fname))

    # For protecting decompressing interupted,
    # decompress to fpath_tmp directory firstly, if decompress
    # successed, move decompress files to fpath and delete
    # fpath_tmp and remove download compress file.

    if tarfile.is_tarfile(fname):
        uncompressed_path = _uncompress_file_tar(fname)
    elif zipfile.is_zipfile(fname):
        uncompressed_path = _uncompress_file_zip(fname)
    else:
        raise TypeError("Unsupport compress file type {}".format(fname))

    return uncompressed_path


def _uncompress_file_zip(filepath):
    files = zipfile.ZipFile(filepath, 'r')
    file_list = files.namelist()

    file_dir = os.path.dirname(filepath)

    if _is_a_single_file(file_list):
        rootpath = file_list[0]
        uncompressed_path = os.path.join(file_dir, rootpath)

        for item in file_list:
            files.extract(item, file_dir)

    elif _is_a_single_dir(file_list):
        rootpath = os.path.splitext(file_list[0])[0].split(os.sep)[-1]
        uncompressed_path = os.path.join(file_dir, rootpath)

        for item in file_list:
            files.extract(item, file_dir)

    else:
        rootpath = os.path.splitext(filepath)[0].split(os.sep)[-1]
        uncompressed_path = os.path.join(file_dir, rootpath)
        if not os.path.exists(uncompressed_path):
            os.makedirs(uncompressed_path)
        for item in file_list:
            files.extract(item, os.path.join(file_dir, rootpath))

    files.close()

    return uncompressed_path


def _uncompress_file_tar(filepath, mode="r:*"):
    files = tarfile.open(filepath, mode)
    file_list = files.getnames()

    file_dir = os.path.dirname(filepath)

    if _is_a_single_file(file_list):
        rootpath = file_list[0]
        uncompressed_path = os.path.join(file_dir, rootpath)
        for item in file_list:
            files.extract(item, file_dir)
    elif _is_a_single_dir(file_list):
        rootpath = os.path.splitext(file_list[0])[0].split(os.sep)[-1]
        uncompressed_path = os.path.join(file_dir, rootpath)
        for item in file_list:
            files.extract(item, file_dir)
    else:
        rootpath = os.path.splitext(filepath)[0].split(os.sep)[-1]
        uncompressed_path = os.path.join(file_dir, rootpath)
        if not os.path.exists(uncompressed_path):
            os.makedirs(uncompressed_path)

        for item in file_list:
            files.extract(item, os.path.join(file_dir, rootpath))

    files.close()

    return uncompressed_path


def _is_a_single_file(file_list):
    if len(file_list) == 1 and file_list[0].find(os.sep) < -1:
        return True
    return False


def _is_a_single_dir(file_list):
S
Steffy-zxf 已提交
323 324 325 326 327 328 329 330 331 332 333
    new_file_list = []
    for file_path in file_list:
        if '/' in file_path:
            file_path = file_path.replace('/', os.sep)
        elif '\\' in file_path:
            file_path = file_path.replace('\\', os.sep)
        new_file_list.append(file_path)

    file_name = new_file_list[0].split(os.sep)[0]
    for i in range(1, len(new_file_list)):
        if file_name != new_file_list[i].split(os.sep)[0]:
L
LielinJiang 已提交
334 335
            return False
    return True