nvidia.dali.fn.video_reader_resize#
- nvidia.dali.fn.video_reader_resize(*, additional_decode_surfaces=2, antialias=True, bytes_per_sample_hint=[0], channels=3, dont_use_mmap=False, dtype=DALIDataType.UINT8, enable_frame_num=False, enable_timestamps=False, file_list='', file_list_frame_num=False, file_list_include_preceding_frame=False, file_root='', filenames=[], image_type=DALIImageType.RGB, initial_fill=1024, interp_type=DALIInterpType.INTERP_LINEAR, labels=None, lazy_init=False, mag_filter=DALIInterpType.INTERP_LINEAR, max_size=None, min_filter=DALIInterpType.INTERP_LINEAR, minibatch_size=32, mode='default', normalized=False, num_shards=1, pad_last_batch=False, pad_sequences=False, prefetch_queue_depth=1, preserve=False, random_shuffle=False, read_ahead=False, resize_longer=0.0, resize_shorter=0.0, resize_x=0.0, resize_y=0.0, resize_z=0.0, roi_end=None, roi_relative=False, roi_start=None, seed=-1, sequence_length, shard_id=0, size=None, skip_cached_images=False, skip_vfr_check=False, step=-1, stick_to_shard=False, stride=1, subpixel_scale=True, temp_buffer_hint=0, tensor_init_bytes=1048576, device=None, name=None)#
警告
此运算符现已弃用。请改用
readers.video_resize()
。在 DALI 1.0 中,所有读取器都被移动到专用的
readers
子模块中,并被重命名以遵循通用模式。这是一个占位符运算符,功能相同,以实现向后兼容性。用于
readers.video_resize()
的旧别名。