+ j<^RIt^RIHt^RIHt^RIt^RIt^RIH t ^RI H uH t ^RIH t ^RIHt^RIHt^RIHt^RIHtHt^R IHt^R IHtHt^R IHtHt^R I H!t!^R I"H#t#H$t$H%t%H&t&^RI'H(t(^RI)H*t*^RI+H,t,^RI-H.t.]%]!RR]#444t/]%]!RR]#444t0]%]!RR]#444t1!RR] Pd4t3!RR] Pd4t4Rt5]!R4RKRl4t6RR lt7RLR!R"llt8]!]64!R#R$] Pd44t9]!R%4!R&R'] Pd44t:!R(R)]4t;!R*R+] Pd4t!R/R0] Pd4t?!R1R2] Pd4t@!R3R4] Pd4tA!R5R6] Pd4tB!R7R8] Pd4tC!R9R:] Pd4tD!R;R<] Pd4tE!R=R>] Pd4tF!R?R@] Pd4tG!RARB] Pd4tH!RCRD] Pd4tI]%!RERF]44tJ]%!RGRH7!RIRJ]J44tKRJRF.tLR#)MN)Callable) dataclass) Parameter)initialization)ACT2FN)Cache)use_kernel_forward_from_hubuse_kernelized_func)GradientCheckpointingLayer)ROPE_INIT_FUNCTIONSdynamic_rope_update)ALL_ATTENTION_FUNCTIONSPreTrainedModel)Unpack) ModelOutputTransformersKwargsauto_docstringcan_return_tuple)deprecate_kwarg)maybe_autocast) AutoModel Xcodec2Configc@a]tRt^-toRtRtRtRtRtV3Rlt Rt Vt R#) Xcodec2Outputa  audio_values (`torch.FloatTensor` of shape `(batch_size, 1, sequence_length)`, *optional*): Decoded audio waveform values in the time domain, obtained using the decoder part of Xcodec2. These represent the reconstructed audio signal. audio_codes (`torch.LongTensor` of shape `(batch_size, 1, codes_length)`, *optional*): Discrete code embeddings computed using `model.encode`. These are the quantized representations of the input audio used for further processing or generation. latents (`torch.Tensor` of shape `(batch_size, dimension, time_steps)`): Quantized continuous representation of input's embedding. audio_codes_mask (`torch.int32` of shape `(batch_size, 1, codes_length)`, *optional*): Downsampled `padding_mask` for indicating valid audio codes in `audio_codes`. Nc<V^8dQh/S[PR,;R&S[PR,;R&S[PR,;R&S[PR,;R&#)N audio_values audio_codeslatentsaudio_codes_mask)torch FloatTensor LongTensorTensor)format __classdict__s"m/home/luca/screen-checking/.venv/lib/python3.14/site-packages/transformers/models/xcodec2/modeling_xcodec2.py __annotate__Xcodec2Output.__annotate__-sa ##d*1!"!!D(/#$\\D '%&llT)0') __name__ __module__ __qualname____firstlineno____doc__rrr r!__annotate_func____static_attributes____classdictcell__r's@r(rr-s( .2L+/K#'G,0'r+rc<a]tRt^CtoRtRtRtRtV3RltRt Vt R#)Xcodec2EncoderOutputaP audio_codes (`torch.LongTensor` of shape `(batch_size, 1, codes_length)`, *optional*): Discrete code embeddings computed using `model.encode`. These represent the compressed, quantized form of the input audio signal that can be used for storage, transmission, or generation. latents (`torch.Tensor` of shape `(batch_size, dimension, time_steps)`): Quantized continuous representation of input's embedding. audio_codes_mask (`torch.int32` of shape `(batch_size, 1, codes_length)`, *optional*): Downsampled `padding_mask` for indicating valid audio codes in `audio_codes`. Nc<V^8dQh/S[PR,;R&S[PR,;R&S[PR,;R&#)rNrr r!)r"r$r%)r&r's"r(r)!Xcodec2EncoderOutput.__annotate__CsI!!D(/\\D ' llT)0!r+r,) r-r.r/r0r1rr r!r2r3r4r5s@r(r7r7Cs# ,0K#'G,0!r+r7c4a]tRt^VtoRtRtV3RltRtVtR#)Xcodec2DecoderOutputa) audio_values (`torch.FloatTensor` of shape `(batch_size, 1, segment_length)`, *optional*): Decoded audio waveform values in the time domain, obtained by converting the discrete codes back into continuous audio signals. This represents the reconstructed audio that can be played back. NcH<V^8dQh/S[PR,;R&#)rNr)r"r#)r&r's"r(r)!Xcodec2DecoderOutput.__annotate__Vs##d*1r+r,) r-r.r/r0r1rr2r3r4r5s@r(r;r;Vs.2Lr+r;caa]tRt^cto]!RRR7R V3RlV3Rlll4t]]!RRR7R V3RlRll44t]P!4] R44t R t Vt V;t#) Xcodec2RotaryEmbeddingdevicez5.18)versionc <V^8dQhRS[/#rconfigr)r&r's"r(r)#Xcodec2RotaryEmbedding.__annotate__esOO}Or+c<\SV`4VPVnVPVnWnVP P R,VnVPpVPR8wd\VP,pV!VP V4wq@n \P!VRR7Vn \P!VP4RR7VnR#) rope_typedefaultF persistentN)super__init__max_position_embeddingsmax_seq_len_cachedoriginal_max_seq_lenrDrope_parametersrGcompute_default_rope_parametersr attention_scalingnnBufferinv_freqcloneoriginal_inv_freq)selfrDr@ rope_init_fnrU __class__s&&& r(rLXcodec2RotaryEmbedding.__init__ds "("@"@$*$B$B! 44[A!%!E!E >>Y &.t~~>L+7 V+L(( (u= !#8>>+;!Nr+cP<V^8dQhRS[RS[S[PS[3,/#)rrDreturn)rtupler"r%float)r&r's"r(r)rEws.55 5Y^_d_k_kmr_rYs5r+c 0VPR,p\VRR4;'gVPVP,pRpRV\P !^V^\P R7V, ,, pVPV4V3#)a~ Computes the inverse frequencies according to the original RoPE implementation Args: config ([`~transformers.PreTrainedConfig`]): The model configuration. Returns: Tuple of (`torch.Tensor`, `float`), containing the inverse frequencies for the RoPE embeddings and the post-processing scaling factor applied to the computed cos/sin (unused in this type of RoPE). rope_thetahead_dimN?dtype)rPgetattr hidden_sizenum_attention_headsr"aranger_to)rDr@kwargsbasedimattention_factorrUs&&, r(rQ6Xcodec2RotaryEmbedding.compute_default_rope_parametersus~%%l3fj$/cc63E3EIcIc3c$5<<3#MPS#STU{{6"$444r+c^VPR,PVP^,R ^4P\P VP R7pVR ,P 4p\VP P\4'd2VP PR8wdVP PMRp\VRR7;_uu_4W4,P^^4p\P!Wf3R R7pVP4VP,pVP4VP,p RRR4XPVP R7X PVP R73# +'giLH;i) Nrer@mpscpuF device_typeenabledrmrdNNNNN)ryNry)rUexpandshaperjr"r_r@ isinstancetypestrr transposecatcosrRsinre) rXx position_idsinv_freq_expandedposition_ids_expandedrufreqsembrrs &&& r(forwardXcodec2RotaryEmbedding.forwards2 MM- ( / / 0B0B10Er1 M P PW\WbWbklksks P t !-Z 8 > > @'1!((--'E'E!((--[`J`ahhmmfk  U C C&>II!QOE))UN3C'')d444C'')d444C D vvAGGv$cff177f&;;; D Cs (A4F F, )rRrDrUrNrWrOrGN)r-r.r/r0rrL staticmethodrQr"no_gradr rr3r4 __classcell__rZr's@@r(r?r?csuXv.OO/O Xv.55/5$ ]]_<<