+
    œ°³j1!  ã                   ó¾   € R t ^ RIHt ^RIHt ^RIHt ^RIHt ^RI	H
t
 ]! RR7      ] ! R R	]4      4       4       t]! RR7      ] ! R
 R]4      4       4       tRR	.tR# )zMoshi model configuration)Ústrict©ÚPreTrainedConfig)ÚRopeParameters)Úauto_docstring)Ú
AutoConfigzkmhf/hf-moshiko)Ú
checkpointc                   ó¦   a a€ ] tR t^t oRtRtR.tRtRtRt	^t
^tRtRt^	tR	tRtR
tRt^tRtRtRt^tRtRtRtRtV 3R ltR tV3R ltRtVt V ;t!# )ÚMoshiDepthConfigaž  
input_size (`int`, *optional*, defaults to 4096):
    Dimensionality of the input hidden states. Used to connect the main decoder to the depth decoder.
audio_vocab_size (`int`, *optional*, defaults to 2048):
    Vocabulary size of the audio part of model. Defines the number of different tokens that can be
    represented by the `audio_codes` passed when calling the Moshi models.
ffn_dim (`int`, *optional*, defaults to 5632):
    Dimensionality of the "intermediate" (often named feed-forward) layer in the depth decoder block. Must be even.

Example:

```python
>>> from transformers import (
...     MoshiDepthConfig,
...     MoshiDepthDecoder,
... )

>>> configuration = MoshiDepthConfig()

>>> # Initializing a MoshiDepthDecoder (with random weights) from the kmhf/hf-moshiko style configuration
>>> model = MoshiDepthDecoder(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config
```Úmoshi_depthÚpast_key_valuesé }  i   é   Ni   Úsiluç{®Gáz”?Tç        i   ç:Œ0âŽyE>Fc                óê   <€ V P                   e   V P                   MV P                  V n         V P                  ;'       g    V P                  V P                  ,          V n        \        SV `  ! R/ VB  R # )N© )Únum_key_value_headsÚnum_attention_headsÚhead_dimÚhidden_sizeÚsuperÚ__post_init__)ÚselfÚkwargsÚ	__class__s   &,€Ún/home/luca/screen-checking/.venv/lib/python3.14/site-packages/transformers/models/moshi/configuration_moshi.pyr   ÚMoshiDepthConfig.__post_init__N   s^   ø€ à(,×(@Ñ(@Ò(LˆD×$Ò$ÐRV×RjÑRjð 	Ô ð Ÿ™×UÐU¨×)9Ñ)9¸T×=UÑ=UÕ)UˆŒÜ‰ÒÑ' Ô'ó    c                óh   € V P                   ^,          ^8X  d   \        RV P                    R24      hR# )úOPart of `@strict`-powered validation. Validates the architecture of the config.ú	`ffn_dim=ú` must be even.N)Úffn_dimÚ
ValueError©r   s   &r   Úvalidate_architectureÚ&MoshiDepthConfig.validate_architectureU   s0   € à�<‰<˜!Õ˜qÔ Ü˜y¨¯©¨°oÐFÓGÐGñ !r    c                óŒ  <€ V ^8„  d   Qh/ S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[ R,          ;R&   S[ ;R&   S[ ;R	&   S[;R
&   S[ R,          ;R&   S[;R&   S[;R&   S[ ;R&   S[S[ ,          ;R&   S[ ;R&   S[;R&   S[ ;R&   S[;R&   S[ R,          ;R&   S[ R,          ;R&   S[ S[S[ ,          ,          R,          ;R&   # )é   Ú
vocab_sizer   Ú
input_sizeÚnum_hidden_layersr   Nr   Úaudio_vocab_sizeÚmax_position_embeddingsÚ
hidden_actr   Úinitializer_rangeÚ	use_cacheÚsliding_windowÚattention_dropoutr%   Úrms_norm_epsÚnum_codebooksÚtie_word_embeddingsÚpad_token_idÚbos_token_idÚeos_token_id)ÚintÚstrÚfloatÚboolÚlist)ÚformatÚ__classdict__s   "€r   Ú__annotate__ÚMoshiDepthConfig.__annotate__   sa  ø‡ ‚ ñ@ ÑñA ñB ÑñC ñD ÑñE ñF ÑñG ñH Ñ!ñI ñJ ˜t�Ñ*ñK ñL Ñ ñM ñN !Ñ$ñO ñP ÑñQ ñR �D�jÑñS ñT Ñ#ñU ñV ÑñW ñX ÑñY ñZ ™s•{Ñ(ñ[ ñ\ Ññ] ñ^ Ññ_ ñ` Ñña ñb Ñ%ñc ñd ˜•*Ñ#ñe ñf ˜•*Ñ#ñg ñh ™™S�	•/ DÕ(Ñ/òi r    )r   r   )"Ú__name__Ú
__module__Ú__qualname__Ú__firstlineno__Ú__doc__Ú
model_typeÚkeys_to_ignore_at_inferencer,   r   r-   r.   r   r   r/   r0   r1   r   r2   r3   r4   r5   r%   r6   r7   r8   r9   r:   r;   r   r(   Ú__annotate_func__Ú__static_attributes__Ú__classdictcell__Ú__classcell__©r   rB   s   @@r   r
   r
      sž   ù‡ € ñð4 €JØ#4Ð"5Ðà€JØ€KØ€JØÐØ!ÐØ&*ÐØ ÐØ#$ÐØ€JØ€HØ#ÐØ€IØ€NØ%(ÐØ€GØ€LØ€MØ %ÐØ#€LØ#€LØ+/€Lõ(òH÷{ … r    r
   c                   óæ   a a€ ] tR t^[t oRtRtR.tR]R]/t	Rt
Rt^ t^ tRtRtR	tRtR
tRtRtRtR	tRtRtRt^tRtRtRtRtRtRt V 3R lt!R t"]#R 4       t$]%V3R lR l4       t&V3R lt'Rt(Vt)V ;t*# )ÚMoshiConfiga  
audio_vocab_size (`int`, *optional*):
    Vocabulary size of the audio part of model. Defines the number of different tokens that can be
    represented by the `audio_codes` passed when calling the Moshi models.
ffn_dim (`int`, *optional*, defaults to 22528):
    Dimensionality of the "intermediate" (often named feed-forward) layer in the main decoder block. Must be even.
audio_encoder_config (`PreTrainedConfig | dict`, *optional*):
    Configuration for the audio encoder.
depth_decoder_config (`PreTrainedConfig | dict`, *optional*):
    Configuration for the depth decoder.

Example:

```python
>>> from transformers import (
...     MoshiConfig,
...     MoshiForConditionalGeneration,
... )

>>> configuration = MoshiConfig()

>>> # Initializing a MoshiForConditionalGeneration (with random weights) from the kmhf/hf-moshiko style configuration
>>> model = MoshiForConditionalGeneration(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config

>>> # Saving the model, including its configuration
>>> model.save_pretrained("kmhf/hf-moshiko")

>>> # loading model and config from pretrained folder
>>> moshi_config = MoshiConfig.from_pretrained("kmhf/hf-moshiko")
>>> model = MoshiForConditionalGeneration.from_pretrained("kmhf/hf-moshiko", config=moshi_config)
```Úmoshir   Úaudio_encoder_configÚdepth_decoder_configr   r   Ni¸  r   r   Tr   i X  r   Fc           
     ó²  <€ V P                   e   V P                   MV P                  V n         V P                  ;'       g    V P                  V P                  ,          V n        \	        V P
                  \        4      '       dD   V P
                  P                  RR4      p\        P                  ! V3/ V P
                  B V n        M)V P
                  f   \        P                  ! R4      V n        V P                  f   V P
                  P                  MV P                  V n
        \	        V P                  \        4      '       dg   V P                  P                  RV P                  RV P                  RV P                  RV P                  /4       \!        R/ V P                  B V n        MV P                  f   \!        4       V n        \"        SV `H  ! R/ VB  R # )NrJ   Úmimir/   r-   r,   r7   r   )r   r   r   r   Ú
isinstancerT   ÚdictÚpopr   Ú	for_modelr/   Úcodebook_sizerU   Úupdater,   r7   r
   r   r   )r   r   Úaudio_encoder_model_typer   s   &, €r   r   ÚMoshiConfig.__post_init__�   s~  ø€ à(,×(@Ñ(@Ò(LˆD×$Ò$ÐRV×RjÑRjð 	Ô ð Ÿ™×UÐU¨×)9Ñ)9¸T×=UÑ=UÕ)UˆŒä�d×/Ñ/´×6Ò6Ø'+×'@Ñ'@×'DÑ'DÀ\ÐSYÓ'ZÐ$Ü(2×(<Ò(<Ð=UÑ(sÐY]×YrÑYrÑ(sˆDÕ%Ø×&Ñ&Ò.Ü(2×(<Ò(<¸VÓ(DˆDÔ%ð 8<×7LÑ7LÒ7TˆD×%Ñ%×3Ò3ÐZ^×ZoÑZoð 	Ôô �d×/Ñ/´×6Ò6Ø×%Ñ%×,Ñ,à&¨×(=Ñ(=Ø  $×"2Ñ"2Ø  $§/¡/Ø# T×%7Ñ%7ð	ôô )9Ñ(U¸4×;TÑ;TÑ(UˆDÕ%Ø×&Ñ&Ò.Ü(8Ó(:ˆDÔ%Ü‰ÒÑ' Ô'r    c                ó  € V P                   ^,          ^8X  d   \        RV P                    R24      hV P                  V P                  P                  8”  d1   \        RV P                   RV P                  P                   R24      hR# )r"   r#   r$   z`num_codebooks=zX` is greater than the maximum number of codebooks that the audio encoder can deal with (z). Please lower it.N)r%   r&   r7   rT   r'   s   &r   r(   Ú!MoshiConfig.validate_architecture»   s£   € à�<‰<˜!Õ˜qÔ Ü˜y¨¯©¨°oÐFÓGÐGà×Ñ × 9Ñ 9× GÑ GÔGÜØ! $×"4Ñ"4Ð!5ð  6Nð  OS÷  Ohñ  Oh÷  Ovñ  Ovð  Nwð  wJð  Kóð ñ Hr    c                ó.   € V P                   P                  # )N)rT   Úsampling_rater'   s   &r   rc   ÚMoshiConfig.sampling_rateÅ   s   € à×(Ñ(×6Ñ6Ð6r    c                ó    <€ V ^8„  d   QhRS[ /# )r+   rT   r   )rA   rB   s   "€r   rC   ÚMoshiConfig.__annotate__Ê   s   ø€ ÷ 
ñ 
á.ñ
r    c                ó2   € V ! RRVP                  4       /VB # )zž
Instantiate a [`MoshiConfig`] (or a derived class) from an audio encoder configuration.

Returns:
    [`MoshiConfig`]: An instance of a configuration object
rT   r   )Úto_dict)ÚclsrT   r   s   &&,r   Úfrom_audio_encoder_configÚ%MoshiConfig.from_audio_encoder_configÉ   s*   € ñ ñ 
Ø!5×!=Ñ!=Ó!?ð
àñ
ð 	
r    c                ó  <€ V ^8„  d   Qh/ S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[ R,          ;R&   S[ R,          ;R&   S[ ;R&   S[S[,          R,          ;R	&   S[;R
&   S[ R,          ;R&   S[;R&   S[;R&   S[ ;R&   S[S[ ,          ;R&   S[ ;R&   S[;R&   S[ ;R&   S[;R&   S[ R,          ;R&   S[ R,          ;R&   S[ S[S[ ,          ,          R,          ;R&   S[S[,          R,          ;R&   S[S[,          R,          ;R&   # )r+   r,   r   r.   r   Nr   r/   r0   Úrope_parametersr1   r   r2   r3   r4   r5   r%   r6   r7   r8   r9   r:   r;   rT   rU   )r<   r   rY   r=   r>   r?   r@   r   )rA   rB   s   "€r   rC   rf   [   s£  ø‡ ‚ ñT ÑñU ñV ÑñW ñX ÑñY ñZ Ñ!ñ[ ñ\ ˜t�Ñ*ñ] ñ^ ˜D•jÑ'ñ_ ñ` !Ñ'ña ñb $¡dÕ*¨TÕ1Ñ8ñc ñd Ññe ñf �D�jÑñg ñh Ñ#ñi ñj Ññk ñl Ññm ñn ™s•{Ñ(ño ñp Ññq ñr Ññs ñt Ññu ñv Ñ%ñw ñx ˜•*Ñ#ñy ñz ˜•*Ñ#ñ{ ñ| ™™S�	•/ DÕ(Ñ/ñ} ñ~ Ñ!1Õ1°DÕ8Ñ?ñ ñ@ Ñ!1Õ1°DÕ8Ñ?òA r    )rT   r/   rU   r   r   )+rE   rF   rG   rH   rI   rJ   rK   r   r
   Úsub_configsr,   r   r.   r   r   r/   r0   rm   r1   r   r2   r3   r4   r5   r%   r6   r7   r8   r9   r:   r;   rT   rU   r   r(   Úpropertyrc   Úclassmethodrj   rL   rM   rN   rO   rP   s   @@r   rR   rR   [   sâ   ù‡ € ñ!ðF €JØ#4Ð"5ÐØ)¨:Ð7MÐO_Ð`€Kà€JØ€KØÐØ!ÐØ&*ÐØ#'ÐØ#'ÐØ48€OØ€JØ€HØ#ÐØ€IØ€NØ%(ÐØ€GØ€LØ€MØ %ÐØ#€LØ#€LØ+/€LØ;?ÐØ;?Ðõ(ò<ð ñ7ó ð7ð ÷
ó ð
÷_ … r    rR   N)rI   Úhuggingface_hub.dataclassesr   Úconfiguration_utilsr   Úmodeling_rope_utilsr   Úutilsr   Úauto.configuration_autor   r
   rR   Ú__all__r   r    r   Ú<module>rw      s‚   ðñ  å .å 3Ý 1Ý #Ý 0ñ Ð,Ô-Øô>HÐ'ó >Hó ó .ð>HñB Ð,Ô-Øô|
Ð"ó |
ó ó .ð|
ð~ Ð,Ð
-‚r    