Vji"dddlmZddlmZmZmZmZddlZddlmZddl m Z GddZ dS))Path)AnyDictOptionalUnionN)Tensor)_corec eZdZdZdedefdZdddddddeee fde ed e ed e ee efd e eee fd e e ee fd dfdZdddddddede ed e ed e ee efd e eee fd e e ee fd efdZdddddddede ed e ed e ee efd e eee fd e e ee fd dfdZdS) VideoEncoderaA video encoder. Args: frames (``torch.Tensor``): The frames to encode. This must be a 4D tensor of shape ``(N, C, H, W)`` where N is the number of frames, C is 3 channels (RGB), H is height, and W is width. Values must be uint8 in the range ``[0, 255]``. frame_rate (float): The frame rate of the **input** ``frames``. Also defines the encoded **output** frame rate. frames frame_ratectjdt|ts t dt |d|jdkrt d|jd|j tj krt d|j d|dkrt d|d ||_ ||_ dS) Nz torchcodec.encoders.VideoEncoderz3Expected frames to be a Tensor, got type(frames) = .z'Expected 4D frames, got frames.shape = z*Expected uint8 frames, got frames.dtype = rz frame_rate = z must be > 0.) torch_C_log_api_usage_once isinstancer ValueErrortypendimshapedtypeuint8_frames _frame_rate)selfr r s \/root/voice-cloning/.venv/lib/python3.11/site-packages/torchcodec/encoders/_video_encoder.py__init__zVideoEncoder.__init__s $$%GHHH&&)) XVDLLVVVWW W ;!  J JJJKK K <5; & &M6<MMMNN N ??< <<<== = %N)codec pixel_formatcrfpreset extra_optionsdestr!r"r#r$r%returnc t|trt|n|}tj|j|jt|||||d|piDdS)a>Encode frames into a file. Args: dest (str or ``pathlib.Path``): The path to the output file, e.g. ``video.mp4``. The extension of the file determines the video container format. codec (str, optional): The codec to use for encoding (e.g., "libx264", "h264"). If not specified, the default codec for the container format will be used. See :ref:`codec_selection` for details. pixel_format (str, optional): The pixel format for encoding (e.g., "yuv420p", "yuv444p"). If not specified, uses codec's default format. See :ref:`pixel_format` for details. crf (int or float, optional): Constant Rate Factor for encoding quality. Lower values mean better quality. Valid range depends on the encoder (e.g. 0-51 for libx264). Defaults to None (which will use encoder's default). See :ref:`crf` for details. preset (str or int, optional): Encoder option that controls the tradeoff between encoding encoding speed and compression (output size). Valid on the encoder (commonly a string: "fast", "medium", "slow"). Defaults to None (which will use encoder's default). See :ref:`preset` for details. extra_options (dict[str, Any], optional): A dictionary of additional encoder options to pass, e.g. ``{"qp": 5, "tune": "film"}``. See :ref:`extra_options` for details. c@g|]\}}||fD]}t|Sstr.0kvxs r z(VideoEncoder.to_file..PG1a1a&EFAr )r r filenamer!r"r#r$r%N)rintr,r encode_video_to_filerritems)rr&r!r"r#r$r%s rto_filezVideoEncoder.to_file#sH!+63 7 7CVV "<'YY%$1$7R#>#>#@#@ r formatc t|trt|n|}tj|j|j|||||d|piDS)aEncode frames into raw bytes, as a 1D uint8 Tensor. Args: format (str): The container format of the encoded frames, e.g. "mp4", "mov", "mkv", "avi", "webm", "flv", etc. codec (str, optional): The codec to use for encoding (e.g., "libx264", "h264"). If not specified, the default codec for the container format will be used. See :ref:`codec_selection` for details. pixel_format (str, optional): The pixel format to encode frames into (e.g., "yuv420p", "yuv444p"). If not specified, uses codec's default format. See :ref:`pixel_format` for details. crf (int or float, optional): Constant Rate Factor for encoding quality. Lower values mean better quality. Valid range depends on the encoder (e.g. 0-51 for libx264). Defaults to None (which will use encoder's default). See :ref:`crf` for details. preset (str or int, optional): Encoder option that controls the tradeoff between encoding encoding speed and compression (output size). Valid on the encoder (commonly a string: "fast", "medium", "slow"). Defaults to None (which will use encoder's default). See :ref:`preset` for details. extra_options (dict[str, Any], optional): A dictionary of additional encoder options to pass, e.g. ``{"qp": 5, "tune": "film"}``. See :ref:`extra_options` for details. Returns: Tensor: The raw encoded bytes as 1D uint8 Tensor. c@g|]\}}||fD]}t|Sr*r+r-s rr2z*VideoEncoder.to_tensor..r3r )r r r9r!r"r#r$r%)rr5r,r encode_video_to_tensorrrr7)rr9r!r"r#r$r% preset_values r to_tensorzVideoEncoder.to_tensorUsL'1&=&=Is6{{{6 +<'%$1$7R#>#>#@#@    r c t|trt|n|}tj|j|j||||||d|piD dS)amEncode frames into a file-like object. Args: file_like: A file-like object that supports ``write()`` and ``seek()`` methods, such as io.BytesIO(), an open file in binary write mode, etc. Methods must have the following signature: ``write(data: bytes) -> int`` and ``seek(offset: int, whence: int = 0) -> int``. format (str): The container format of the encoded frames, e.g. "mp4", "mov", "mkv", "avi", "webm", "flv", etc. codec (str, optional): The codec to use for encoding (e.g., "libx264", "h264"). If not specified, the default codec for the container format will be used. See :ref:`codec_selection` for details. pixel_format (str, optional): The pixel format for encoding (e.g., "yuv420p", "yuv444p"). If not specified, uses codec's default format. See :ref:`pixel_format` for details. crf (int or float, optional): Constant Rate Factor for encoding quality. Lower values mean better quality. Valid range depends on the encoder (e.g. 0-51 for libx264). Defaults to None (which will use encoder's default). See :ref:`crf` for details. preset (str or int, optional): Encoder option that controls the tradeoff between encoding encoding speed and compression (output size). Valid on the encoder (commonly a string: "fast", "medium", "slow"). Defaults to None (which will use encoder's default). See :ref:`preset` for details. extra_options (dict[str, Any], optional): A dictionary of additional encoder options to pass, e.g. ``{"qp": 5, "tune": "film"}``. See :ref:`extra_options` for details. c@g|]\}}||fD]}t|Sr*r+r-s rr2z-VideoEncoder.to_file_like..r3r ) r r r9 file_liker!r"r#r$r%N)rr5r,r encode_video_to_file_likerrr7)rrAr9r!r"r#r$r%s r to_file_likezVideoEncoder.to_file_likesR!+63 7 7CVV '<'%$1$7R#>#>#@#@ r )__name__ __module__ __qualname____doc__rfloatrrr,rrr5rrr8r>rCr*r rr r s6 &v &e & & & &$ $&*+/,0260 0 0 CI0 } 0 sm 0 eCJ' ( 0 sCx)0  S#X/0  0 0 0 0 l $&*+/,0262 2 2 2 } 2 sm 2 eCJ' ( 2 sCx)2  S#X/2  2 2 2 2 r $&*+/,0266 6 6 6 } 6 sm 6 eCJ' (6 sCx)6  S#X/6  6 6 6 6 6 6 r r ) pathlibrtypingrrrrrr torchcodecr r r*r rrLs------------ u u u u u u u u u u r