diff --git a/python/paddle/distributed/fleet/utils/fs.py b/python/paddle/distributed/fleet/utils/fs.py index 8895a529526f765a487326e318c2da04231cbe5d..378d7025a6818a70eb6d16b4386bb49693036dc4 100644 --- a/python/paddle/distributed/fleet/utils/fs.py +++ b/python/paddle/distributed/fleet/utils/fs.py @@ -842,8 +842,9 @@ class HDFSClient(FS): if self.is_file(fs_path): return self._try_download(fs_path, local_path) # download dir - _, all_filenames = self.ls_dir(fs_path) - all_files = [fs_path + i for i in all_filenames] + dirs, all_filenames = self.ls_dir(fs_path) + all_files = [fs_path + "/" + i for i in all_filenames] + all_files.extend([fs_path + "/" + i for i in dirs]) procs = [] for i in range(multi_processes): process_datas = self._split_files(all_files, i, multi_processes)