o 0i=@spddlZddlZddlZddlZddlZddlmZddlmZm Z m Z m Z m Z m Z ddlZddlZddlmZddlmZzddlZejejjeejjjdsWed ZWn eyeed ZYnwdde#d1e e%efd2e e e%efd!e#d3e#d e ej$ej$e e#efff d4d5Z*ddd efd6d7Z+ddd e e'fd8d9Z,d?de#d!e#d e e e'e e%ffd:d;Z-dS)@N)Fraction)AnyDictListOptionalTupleUnion)_log_api_usage_once) _video_opt pict_typea+Your version of PyAV is too old for the necessary video operations in torchvision. If you are on Python 3.5, you will have to build from source (the conda-forge packages are not up-to-date). See https://github.com/mikeboers/PyAV#installation for instructions on how to install PyAV on your system. zPyAV is not installed, and is necessary for the video operations in torchvision. See https://github.com/mikeboers/PyAV#installation for instructions on how to install PyAV on your system. returncCstttrtdSN isinstanceav ExceptionrrR/home/ubuntu/SoloSpeech/.venv/lib/python3.10/site-packages/torchvision/io/video.py_check_av_available's rcCs ttt Srrrrrr _av_available,s r libx264filename video_arrayfps video_codecoptions audio_array audio_fps audio_codec audio_optionsc  Cstjstjsttttj|tjd }t |t r%t |}tj|dd} | j||d} |jd| _|jd| _|dkrFdnd | _|pLi| _|d urd d d d d d ddddd } | j||d} |pii| _|jd} | dkrvdnd}| jjdjj}t | |}t| |}tjj|||d}||_| |D]}| |q| D]}| |q|D]}tj!j|d d}d|_"| |D]}| |qq| D]}| |qWd d S1swYd S)a| Writes a 4d tensor in [T, H, W, C] format in a video file Args: filename (str): path where the video will be saved video_array (Tensor[T, H, W, C]): tensor containing the individual frames, as a uint8 tensor in [T, H, W, C] format fps (Number): video frames per second video_codec (str): the name of the video codec, i.e. "libx264", "h264", etc. options (Dict): dictionary containing options to be passed into the PyAV video stream audio_array (Tensor[C, N]): tensor containing the audio, where C is the number of channels and N is the number of samples audio_fps (Number): audio sample rate, typically 44100 or 48000 audio_codec (str): the name of the audio codec, i.e. "mp3", "aac", etc. audio_options (Dict): dictionary containing options to be passed into the PyAV audio stream dtypew)mode)rater r libx264rgbyuv420prgb24Nzs2z%_read_from_stream..csg|]}|kr|qSrrro)rarrrtr) _CALLED_TIMES_GC_COLLECTION_INTERVALgccollectintmathfloor time_baserCceilwarningswarntype codec_context extradatafindresearchgroupmaxseekrAVError enumeratedecodernsortedleninsert)rVrarbrcrWrd should_buffermax_buffer_sizerposdo seek_offset buffer_count_idxr^resultpreceding_framesfirst_frame_ptsrrrr_read_from_streamsh            "raframes audio_frames ref_startref_endc Cs||dj|dj}}|jd}||d|}d}|} ||kr(t|||}||kr4t|||} |dd|| fS)Nrr )rnrHrz) rrrrstartend total_aframesstep_per_aframes_idxe_idxrrr_align_audio_framess rrnTHWC start_ptsend_pts output_formatc Cstjstjstt|}|dvrtd|dddlm }|dkrBt j |s6t d|t||||\}}}nt|durMtd }||kr[td |d |i}g} g} tj} zdtj|d d S} | jjrx| jjdj} | jjrt| |||| jjdddi} | jjdj} | durt| |d<| jjrt| |||| jjdddi} | jjdj|d<Wdn1swYWn tjyYnwdd| D}dd| D}|rtt|}ntj dtj!d}|r,t"|d}t|}|dkr$t#t$%|d| }|td kr$t#t$&|d| }t'|| ||}ntj dtj(d}|dkrA|)dddd}|||fS)aF Reads a video from a file, returning both the video frames and the audio frames Args: filename (str): path to the video file. If using the pyav backend, this can be whatever ``av.open`` accepts. start_pts (int if pts_unit = 'pts', float / Fraction if pts_unit = 'sec', optional): The start presentation time of the video end_pts (int if pts_unit = 'pts', float / Fraction if pts_unit = 'sec', optional): The end presentation time pts_unit (str, optional): unit in which start_pts and end_pts values will be interpreted, either 'pts' or 'sec'. Defaults to 'pts'. output_format (str, optional): The format of the output video tensors. Can be either "THWC" (default) or "TCHW". Returns: vframes (Tensor[T, H, W, C] or Tensor[T, C, H, W]): the `T` video frames aframes (Tensor[K, L]): the audio frames, where `K` is the number of channels and `L` is the number of points info (Dict): metadata for the video and audio. Can contain the fields video_fps (float) and audio_fps (int) )rTCHWz5output_format should be either 'THWC' or 'TCHW', got .rget_video_backendpyavzFile not found: Nrfz7end_pts should be larger than start_pts, got start_pts=z and end_pts=ignoremetadata_errorsrh video_fpsrMr cSsg|]}|qSr)to_rgb to_ndarrayrpr^rrrrtCruzread_video..cSsg|]}|qSr)rrrrrrtD)rr r rir#r re)r rrrir )*r;r<r=r>r read_videoupper ValueError torchvisionrospathexists RuntimeErrorr _read_videorrCdefault_timebaserrFrLrMr}rhr average_rater'rr@rDstackemptyrA concatenaterzr{r|r~rfloat32permute)rrrrcrrvframesrinfo video_framesraudio_timebaserVr vframes_list aframes_listrrrrs           rcCs*|jdjj}|dur dSd|vrdSdS)NrFsLavcT)rLrr)rVrrrr!_can_read_timestamps_from_packets]s rcCs4t|rdd|jddDSdd|jddDS)NcSg|] }|jdur|jqSrrmrpxrrrrtiz,_decode_video_timestamps..r)rhcSrrrmrrrrrtkr)rdemuxr)rVrrr_decode_video_timestampsfsrc sBtjstjsttddlm}|dkrt ||St d}g}zDt j |dd3}|j jrZ|j jd}|jzt|}Wnt jyTtd|Ynwt|j}Wdn1sdwYWn!t jy}zd|d |}t|tWYd}~nd}~ww||d krfd d |D}||fS) a  List the video frames timestamps. Note that the function decodes the whole video frame-by-frame. Args: filename (str): path to the video file pts_unit (str, optional): unit in which timestamp values will be returned either 'pts' or 'sec'. Defaults to 'pts'. Returns: pts (List[int] if pts_unit = 'pts', List[Fraction] if pts_unit = 'sec'): presentation timestamps for each one of the frames in the video. video_fps (float, optional): the frame rate for the video rrrNrrz Failed decoding frames for file zFailed to open container for z; Caught error: recsg|]}|qSrrrvideo_time_baserrrtrz)read_video_timestamps..)r;r<r=r>r read_video_timestampsrrr _read_video_timestampsrrrFrLrhr}rrrrrCrRuntimeWarningsort) rrcrrrnrV video_streamemsgrrrrns>       r)rN)rNNNNN)rNrnrrm).rxr{rrr fractionsrtypingrrrrrrrBrDr;utilsr r rlogging set_levelERRORhasattrrhr^rU ImportErrorrboolrrvrwstrTensorrCr?rzrrrrrrrrrrs        X$ P   q ,