§
    9²¸jlG  ã                   óD  — d dl mZ d dlZddlmZ ddlmZmZm	Z	 ddl
mZ ddlmZ ddlmZ dd	lmZmZ dd
lmZ ddlmZ ddlmZmZmZmZ ddlmZmZmZ ddl m!Z! ddl"m#Z# ddl$m%Z%m&Z&m'Z'm(Z( ddl)m*Z*m+Z+ ddl,m-Z- ddl.m/Z/m0Z0m1Z1 ddl2m3Z3m4Z4  e¦   «         r
d dl5Z5d dl5m6Z6  ej7        e8¦  «        Z9 G d„ de+¦  «        Z: e!d¬¦  «        e G d„ de*¦  «        ¦   «         ¦   «         Z; G d„ de-¦  «        Z<d4d „Z=e G d!„ d"e/¦  «        ¦   «         Z> G d#„ d$e6j?        ¦  «        Z@ G d%„ d&e¦  «        ZA G d'„ d(e(¦  «        ZB G d)„ d*eB¦  «        ZC G d+„ d,e'¦  «        ZD ed-¬.¦  «         G d/„ d0e&¦  «        ¦   «         ZE ed-¬.¦  «         G d1„ d2e%¦  «        ¦   «         ZFg d3¢ZGdS )5é    )ÚCallableNé   )ÚACT2FN)Ú
AudioInputÚ make_audio_chat_template_contentÚ make_list_of_audio_chat_template)ÚCache)ÚBatchFeature)ÚGradientCheckpointingLayer)ÚBaseModelOutputWithPoolingÚCausalLMOutputWithPast)ÚALL_ATTENTION_FUNCTIONS)ÚUnpack)ÚTransformersKwargsÚauto_docstringÚis_torch_availableÚlogging)Úcan_return_tupleÚmerge_with_config_defaultsÚno_inherit_decorator)Úrequires)Úcapture_outputsé   )Ú&AudioFlamingo3ForConditionalGenerationÚAudioFlamingo3ModelÚ!AudioFlamingo3MultiModalProjectorÚAudioFlamingo3PreTrainedModel)ÚAudioFlamingo3ProcessorÚAudioFlamingo3ProcessorKwargs)ÚGlmRotaryEmbedding)ÚLlamaAttentionÚeager_attention_forwardÚrotate_halfé   )ÚGlmAsrConfigÚGlmAsrEncoderConfig)Únnc                   ó   — e Zd ZdS )ÚGlmAsrProcessorKwargsN©Ú__name__Ú
__module__Ú__qualname__© ó    úy/root/private/project/pallet_detection_v2/.venv/lib/python3.11/site-packages/transformers/models/glmasr/modular_glmasr.pyr)   r)   3   ó   € € € € € € € r/   r)   )Útorch)Úbackendsc            	       óˆ   ‡ — e Zd Z	 	 	 	 dˆ fd„	Zdd	„Z	 dd
eee         z  ez  deee         z  dz  dee	         de
fd„Zˆ xZS )ÚGlmAsrProcessorNú<|pad|>ú&Please transcribe this audio into texté�  c                 óV   •— t          ¦   «                              ||||||¬¦  «         dS )a‡  
        audio_token (`Optional[str]`, *optional*, defaults to `"<|pad|>`"):
            Special token used to represent audio inputs in the chat template.
        default_transcription_prompt (`str`, *optional*, defaults to `"Please transcribe this audio into text"`):
            Default prompt to use for transcription tasks when applying transcription requests.
        max_audio_len (`int`, *optional*, defaults to 655):
            Maximum length of audio sequences in seconds. Audio longer than this will be truncated.
            655 gives approximately 8192 tokens, corresponding to the maximum sequence length of the text model.
        )Úchat_templateÚaudio_tokenÚdefault_transcription_promptÚmax_audio_lenN)ÚsuperÚ__init__)ÚselfÚfeature_extractorÚ	tokenizerr:   r;   r<   r=   Ú	__class__s          €r0   r?   zGlmAsrProcessor.__init__9   sB   ø€ õ$ 	‰Œ×ÒØØØ'Ø#Ø)EØ'ð 	ñ 	
ô 	
ð 	
ð 	
ð 	
r/   Úaudio_lengthsútorch.TensorÚreturnc                 ó`   — d}dD ]\  }}}|d|z  z   |dz
  z
  dz
  |z  dz   }Œ||z
  |z  dz   }|S )Né   ©)r$   r   r$   )r$   r   r   r   r$   r.   )r@   rD   Úmerge_factorÚpaddingÚkernel_sizeÚstrideÚ
num_tokenss          r0   Ú_get_audio_token_lengthz'GlmAsrProcessor._get_audio_token_lengthT   sc   € ØˆØ,Bð 	`ð 	`Ñ(ˆG�[ &Ø*¨Q°©[Ñ8¸KÈ!¹OÑLÈqÑPÐU[Ñ[Ð^_Ñ_ˆMˆMà# lÑ2°|ÑCÀaÑGˆ
ØÐr/   ÚaudioÚpromptÚkwargsc                 óÄ  — t          t          |¦  «        ¦  «        }d„ |D ¦   «         }t          |¦  «        }|dk    rt          d¦  «        ‚|€| j        g|z  }nÛt          |t          ¦  «        r|g|z  }n¿t          |t           t          f¦  «        r”t          |¦  «        |k    r#t          dt          |¦  «        › d|› d�¦  «        ‚g }|D ]X}|€|                     | j        ¦  «         Œt          |t          ¦  «        r|                     |¦  «         ŒJt          d¦  «        ‚nt          d	¦  «        ‚d
„ t          ||¦  «        D ¦   «         } | j        |fddddœ|¤ŽS )a	  
        Prepare inputs for automatic speech recognition without manually writing the default transcription prompt.

        Args:
            audio (`str`, `list[str]`, `np.ndarray`, `torch.Tensor`, `list[np.ndarray]`, `list[torch.Tensor]`):
                Audio to transcribe. Strings are interpreted as local paths or URLs and will be loaded automatically by
                the chat template loader; NumPy arrays and PyTorch tensors are forwarded directly.
            prompt (`str` or `list[str]`, *optional*):
                Custom prompt(s) to include in the user turn. A list must be the same length as the batch. When `None`,
                each sample uses `"Transcribe the input speech."`.
            **kwargs:
                Additional keyword arguments forwarded to [`~GlmAsrProcessor.apply_chat_template`] (for example
                `text_kwargs`, `audio_kwargs`, ...).

        Returns:
            [`BatchFeature`]: Processor outputs ready to be passed to [`GlmAsrForConditionalGeneration.generate`].

        c                 ó¶   — g | ]V}t          |t          j        ¦  «        r8|                     ¦   «                              ¦   «                              ¦   «         n|‘ŒWS r.   )Ú
isinstancer2   ÚTensorÚdetachÚcpuÚnumpy)Ú.0Úels     r0   ú
<listcomp>z?GlmAsrProcessor.apply_transcription_request.<locals>.<listcomp>v   sN   € ÐoÐoÐoÐ]_µJ¸rÅ5Ä<Ñ4PÔ4PÐX�r—y’y‘{”{—’Ñ(Ô(×.Ò.Ñ0Ô0Ð0ÐVXÐoÐoÐor/   r   z)`audio` must contain at least one sample.Nz	Received z prompt(s) for z$ audio sample(s); counts must match.z'Each prompt must be a string or `None`.z<`prompt` must be a string, a sequence of strings, or `None`.c                 óD   — g | ]\  }}d t          |¦  «        d|dœgdœg‘ŒS )ÚuserÚtext)Útyper_   )ÚroleÚcontent)r   )rZ   Úprompt_textÚ
audio_items      r0   r\   z?GlmAsrProcessor.apply_transcription_request.<locals>.<listcomp>�   sX   € ð 
ð 
ð 
ñ (�˜Zð #å8¸ÑDÔDØ!'°Ð=Ð=ð ðð ðð
ð 
ð 
r/   T)ÚtokenizeÚadd_generation_promptÚreturn_dict)Úlistr   ÚlenÚ
ValueErrorr<   rU   ÚstrÚtupleÚappendÚ	TypeErrorÚzipÚapply_chat_template)	r@   rP   rQ   rR   Úaudio_itemsÚ
batch_sizeÚpromptsÚitemÚconversationss	            r0   Úapply_transcription_requestz+GlmAsrProcessor.apply_transcription_request\   sÈ  € õ2 /3Õ3SÐTYÑ3ZÔ3ZÑ.[Ô.[ˆØoÐoÐcnÐoÑoÔoˆå˜Ñ%Ô%ˆ
Ø˜Š?ˆ?ÝÐHÑIÔIÐIàˆ>ØÔ8Ð9¸JÑFˆGˆGÝ˜¥Ñ$Ô$ð 	\Ø�h Ñ+ˆGˆGÝ˜¥¥u Ñ.Ô.ð 	\Ý�6‰{Œ{˜jÒ(Ð(Ý Øl¥ F¡¤ÐlÐl¸JÐlÐlÐlñô ð ð ˆGØð Oð O�Ø�<Ø—N’N 4Ô#DÑEÔEÐEÐEÝ ¥cÑ*Ô*ð OØ—N’N 4Ñ(Ô(Ð(Ð(å#Ð$MÑNÔNÐNðOõ ÐZÑ[Ô[Ð[ð
ð 
õ ,/¨w¸Ñ+DÔ+Dð
ñ 
ô 
ˆð (ˆtÔ'Øð
àØ"&Øð	
ð 
ð
 ð
ð 
ð 	
r/   )Nr6   r7   r8   )rD   rE   rF   rE   ©N)r+   r,   r-   r?   rO   rk   rh   r   r   r)   r
   rv   Ú__classcell__©rC   s   @r0   r5   r5   6   sÐ   ø€ € € € € ð ØØ%MØð
ð 
ð 
ð 
ð 
ð 
ð6ð ð ð ð *.ðG
ð G
à�T˜#”Y‰ Ñ+ðG
ð �d˜3”i‘ $Ñ&ðG
ð Ð.Ô/ð	G
ð
 
ðG
ð G
ð G
ð G
ð G
ð G
ð G
ð G
r/   r5   c                   ó   — e Zd ZdS )ÚGlmAsrRotaryEmbeddingNr*   r.   r/   r0   r{   r{   ¦   r1   r/   r{   c                 ó˜  — |                      |¦  «        }|                      |¦  «        }|j        d         }| dd |…f         | d|d …f         }}|dd |…f         |d|d …f         }
}	||z  t          |¦  «        |z  z   }|	|z  t          |	¦  «        |z  z   }t          j        ||gd¬¦  «        }t          j        ||
gd¬¦  «        }||fS )Néÿÿÿÿ.)Údim)Ú	unsqueezeÚshaper#   r2   Úcat)ÚqÚkÚcosÚsinÚposition_idsÚunsqueeze_dimÚ
rotary_dimÚq_rotÚq_passÚk_rotÚk_passÚq_embedÚk_embeds                r0   Úapply_rotary_pos_embr�   ©   sð   € Ø
�-Š-˜Ñ
&Ô
&€CØ
�-Š-˜Ñ
&Ô
&€Cà”˜2”€JØ�c˜;˜J˜;Ð&Ô'¨¨3°
°°Ð+;Ô)<ˆ6€EØ�c˜;˜J˜;Ð&Ô'¨¨3°
°°Ð+;Ô)<ˆ6€Eð �s‰{�{¨5Ñ1Ô1°CÑ7Ñ8€GØ�s‰{�{¨5Ñ1Ô1°CÑ7Ñ8€Gõ Œi˜ &Ð)¨rÐ2Ñ2Ô2€GÝŒi˜ &Ð)¨rÐ2Ñ2Ô2€GØ�GÐÐr/   c                   ó¨   ‡ — e Zd Zdedefˆ fd„Z	 d
dej        deej        ej        f         dz  de	e
         deej        ej        f         fd	„Zˆ xZS )ÚGlmAsrAttentionÚconfigÚ	layer_idxc                 óÊ  •— t          ¦   «                              ||¦  «         d| _        t          j        |j        |j        | j        z  d¬¦  «        | _        t          j        |j        |j	        | j        z  d¬¦  «        | _
        t          j        |j        |j	        | j        z  d¬¦  «        | _        t          j        |j        | j        z  |j        d¬¦  «        | _        d S )NFT)Úbias)r>   r?   Ú	is_causalr'   ÚLinearÚhidden_sizeÚnum_attention_headsÚhead_dimÚq_projÚnum_key_value_headsÚk_projÚv_projÚo_proj©r@   r’   r“   rC   s      €r0   r?   zGlmAsrAttention.__init__½   sÆ   ø€ Ý‰Œ×Ò˜ Ñ+Ô+Ð+ØˆŒÝ”i Ô 2°FÔ4NÐQUÔQ^Ñ4^ÐeiÐjÑjÔjˆŒÝ”i Ô 2°FÔ4NÐQUÔQ^Ñ4^ÐejÐkÑkÔkˆŒÝ”i Ô 2°FÔ4NÐQUÔQ^Ñ4^ÐeiÐjÑjÔjˆŒÝ”i Ô :¸T¼]Ñ JÈFÔL^ÐeiÐjÑjÔjˆŒˆˆr/   NÚhidden_statesÚposition_embeddingsrR   rF   c                 óà  — |j         d d…         }g |¢d‘| j        ‘R }|                      |¦  «                             |¦  «                             dd¦  «        }|                      |¦  «                             |¦  «                             dd¦  «        }|                      |¦  «                             |¦  «                             dd¦  «        }|\  }	}
t          |||	|
¦  «        \  }}t          j	        | j
        j        t          ¦  «        } || |||fd | j        sdn| j        | j        dœ|¤Ž\  }} |j        g |¢d‘R Ž                      ¦   «         }|                      |¦  «        }||fS )Nr}   r$   r   g        )Úattention_maskÚdropoutÚscaling)r€   rš   r›   ÚviewÚ	transposer�   rž   r�   r   Úget_interfacer’   Ú_attn_implementationr"   ÚtrainingÚattention_dropoutr¦   ÚreshapeÚ
contiguousrŸ   )r@   r¡   r¢   rR   Úinput_shapeÚhidden_shapeÚquery_statesÚ
key_statesÚvalue_statesr„   r…   Úattention_interfaceÚattn_outputÚattn_weightss                 r0   ÚforwardzGlmAsrAttention.forwardÅ   sš  € ð $Ô)¨#¨2¨#Ô.ˆØ8˜Ð8 bÐ8¨$¬-Ð8Ð8ˆà—{’{ =Ñ1Ô1×6Ò6°|ÑDÔD×NÒNÈqÐRSÑTÔTˆØ—[’[ Ñ/Ô/×4Ò4°\ÑBÔB×LÒLÈQÐPQÑRÔRˆ
Ø—{’{ =Ñ1Ô1×6Ò6°|ÑDÔD×NÒNÈqÐRSÑTÔTˆà&‰ˆˆSÝ#7¸ÀjÐRUÐWZÑ#[Ô#[Ñ ˆ�jå(?Ô(MØŒKÔ,Õ.Eñ)
ô )
Ðð %8Ð$7ØØØØð		%
ð
  Ø#œ}ÐH�C�C°$Ô2HØ”Lð	%
ð 	%
ð ð	%
ð 	%
Ñ!ˆ�\ð *�kÔ)Ð;¨;Ð;¸Ð;Ð;Ð;×FÒFÑHÔHˆØ—k’k +Ñ.Ô.ˆØ˜LÐ(Ð(r/   rw   ©r+   r,   r-   r%   Úintr?   r2   rV   rl   r   r   r·   rx   ry   s   @r0   r‘   r‘   »   sÂ   ø€ € € € € ðk˜|ð k¸ð kð kð kð kð kð kð IMð!)ð !)à”|ð!)ð # 5¤<°´Ð#=Ô>ÀÑEð!)ð Ð+Ô,ð	!)ð
 
ˆuŒ|˜Uœ\Ð)Ô	*ð!)ð !)ð !)ð !)ð !)ð !)ð !)ð !)r/   r‘   c                   ó4   ‡ — e Zd Zˆ fd„Zdej        fd„Zˆ xZS )Ú	GlmAsrMLPc                 ó  •— t          ¦   «                              ¦   «          t          j        |j        |j        ¦  «        | _        t          j        |j        |j        ¦  «        | _        t          |j	                 | _
        d S rw   )r>   r?   r'   r—   r˜   Úintermediate_sizeÚfc1Úfc2r   Ú
hidden_actÚact_fn©r@   r’   rC   s     €r0   r?   zGlmAsrMLP.__init__ê   s^   ø€ Ý‰Œ×ÒÑÔÐÝ”9˜VÔ/°Ô1IÑJÔJˆŒÝ”9˜VÔ5°vÔ7IÑJÔJˆŒÝ˜VÔ.Ô/ˆŒˆˆr/   r¡   c                 ó„   — |                       |¦  «        }|                      |¦  «        }|                      |¦  «        }|S rw   )r¾   rÁ   r¿   )r@   r¡   s     r0   r·   zGlmAsrMLP.forwardð   s;   € ØŸš Ñ/Ô/ˆØŸš MÑ2Ô2ˆØŸš Ñ/Ô/ˆØÐr/   )r+   r,   r-   r?   r2   rV   r·   rx   ry   s   @r0   r»   r»   é   sU   ø€ € € € € ð0ð 0ð 0ð 0ð 0ð U¤\ð ð ð ð ð ð ð ð r/   r»   c            	       óŽ   ‡ — e Zd Zdedefˆ fd„Z	 d
dej        deej        ej        f         dz  de	e
         dej        fd	„Zˆ xZS )ÚGlmAsrEncoderLayerr’   r“   c                 ó,  •— t          ¦   «                              ¦   «          |j        | _        t          ||¬¦  «        | _        t          |¦  «        | _        t          j        |j        ¦  «        | _	        t          j        |j        ¦  «        | _
        d S )N)r’   r“   )r>   r?   r˜   r‘   Ú	self_attnr»   Úmlpr'   Ú	LayerNormÚinput_layernormÚpost_attention_layernormr    s      €r0   r?   zGlmAsrEncoderLayer.__init__ø   sw   ø€ Ý‰Œ×ÒÑÔÐØ!Ô-ˆÔå(°À)ÐLÑLÔLˆŒå˜VÑ$Ô$ˆŒÝ!œ|¨FÔ,>Ñ?Ô?ˆÔÝ(*¬°VÔ5GÑ(HÔ(HˆÔ%Ð%Ð%r/   Nr¡   r¢   rR   rF   c                 óÆ   — |}|                       |¦  «        } | j        d||dœ|¤Ž\  }}||z   }|}|                      |¦  «        }|                      |¦  «        }||z   }|S )N)r¡   r¢   r.   )rÊ   rÇ   rË   rÈ   )r@   r¡   r¢   rR   ÚresidualÚ_s         r0   r·   zGlmAsrEncoderLayer.forward  s•   € ð !ˆØ×,Ò,¨]Ñ;Ô;ˆà)˜4œ>ð 
Ø'Ø 3ð
ð 
ð ð
ð 
Ñˆ�qð
 ! =Ñ0ˆð !ˆØ×5Ò5°mÑDÔDˆØŸš Ñ/Ô/ˆØ  =Ñ0ˆØÐr/   rw   r¸   ry   s   @r0   rÅ   rÅ   ÷   s¶   ø€ € € € € ðI˜|ð I¸ð Ið Ið Ið Ið Ið Ið IMðð à”|ðð # 5¤<°´Ð#=Ô>ÀÑEðð Ð+Ô,ð	ð
 
Œðð ð ð ð ð ð ð r/   rÅ   c                   ó   — e Zd ZdS )ÚGlmAsrPreTrainedModelNr*   r.   r/   r0   rÐ   rÐ     r1   r/   rÐ   c                   ó�   ‡ — e Zd ZU eed<   dZdZdgZee	dœZ
defˆ fd„Zeeedee         fd„¦   «         ¦   «         ¦   «         Zˆ xZS )	ÚGlmAsrEncoderr’   Úinput_featuresrP   rÅ   )r¡   Ú
attentionsc                 óô  •‡— t          ¦   «                              ‰¦  «         t          j        ‰j        ‰j        dd¬¦  «        | _        t          j        ‰j        ‰j        ddd¬¦  «        | _        t          j        ˆfd„t          ‰j
        ¦  «        D ¦   «         ¦  «        | _        t          j        ‰j        ¦  «        | _        t          ‰¬¦  «        | _        d| _        |                      ¦   «          d S )	Nr   r$   )rL   rK   r   )rL   rM   rK   c                 ó0   •— g | ]}t          ‰|¦  «        ‘ŒS r.   )rÅ   )rZ   r“   r’   s     €r0   r\   z*GlmAsrEncoder.__init__.<locals>.<listcomp>.  s$   ø€ ÐdÐdÐd°yÕ ¨	Ñ2Ô2ÐdÐdÐdr/   )r’   F)r>   r?   r'   ÚConv1dÚnum_mel_binsr˜   Úconv1Úconv2Ú
ModuleListÚrangeÚnum_hidden_layersÚlayersrÉ   Únormr{   Ú
rotary_embÚgradient_checkpointingÚ	post_initrÂ   s    `€r0   r?   zGlmAsrEncoder.__init__(  sÞ   øø€ Ý‰Œ×Ò˜Ñ Ô Ð Ý”Y˜vÔ2°FÔ4FÐTUÐ_`ÐaÑaÔaˆŒ
Ý”Y˜vÔ1°6Ô3EÐSTÐ]^ÐhiÐjÑjÔjˆŒ
å”mØdÐdÐdÐdÅEÈ&ÔJbÑDcÔDcÐdÑdÔdñ
ô 
ˆŒõ ”L Ô!3Ñ4Ô4ˆŒ	Ý/°vÐ>Ñ>Ô>ˆŒØ&+ˆÔ#Ø�ŠÑÔÐÐÐr/   rR   c                 óø  — t           j                             |                      |¦  «        ¦  «        }t           j                             |                      |¦  «        ¦  «        }|                     dd¦  «        }|}|                      |t          j        |j	        d         |j
        ¬¦  «        d d d …f         ¬¦  «        }| j        D ]} ||fd|i|¤Ž}Œ|                      |¦  «        }t          |¬¦  «        S )Nr$   r   ©Údevice)r†   r¢   )Úlast_hidden_state)r'   Ú
functionalÚgelurÙ   rÚ   r¨   rà   r2   Úaranger€   rå   rÞ   rß   r   )r@   rÓ   rR   Úinputs_embedsr¡   r¢   Úencoder_layers          r0   r·   zGlmAsrEncoder.forward5  s  € õ œ×*Ò*¨4¯:ª:°nÑ+EÔ+EÑFÔFˆÝœ×*Ò*¨4¯:ª:°mÑ+DÔ+DÑEÔEˆØ%×/Ò/°°1Ñ5Ô5ˆà%ˆØ"ŸošoØ­¬°]Ô5HÈÔ5KÐTaÔThÐ(iÑ(iÔ(iÐjnÐpqÐpqÐpqÐjqÔ(rð .ñ 
ô 
Ðð "œ[ð 	lð 	lˆMØ)˜M¨-ÐkÐkÐM`ÐkÐdjÐkÐkˆMˆMàŸ	š	 -Ñ0Ô0ˆÝ)¸MÐJÑJÔJÐJr/   )r+   r,   r-   r&   Ú__annotations__Úmain_input_nameÚinput_modalitiesÚ_no_split_modulesrÅ   r‘   Ú_can_record_outputsr?   r   r   r   r   r   r·   rx   ry   s   @r0   rÒ   rÒ     sÅ   ø€ € € € € € ØÐÐÑØ&€OØÐØ-Ð.Ðà+Ø%ðð Ðð
Ð2ð ð ð ð ð ð ð  ØØðK°Ð7IÔ0Jð Kð Kð Kñ „^ñ „_ñ  ÔðKð Kð Kð Kð Kr/   rÒ   c                   ó$   ‡ — e Zd Zdefˆ fd„Zˆ xZS )ÚGlmAsrMultiModalProjectorr’   c                 ó  •— t          ¦   «                              ¦   «          t          j        |j        j        |j        j        dz  ¦  «        | _        t          j        |j        j        dz  |j        j        ¦  «        | _	        d S )Nr   )
r>   r?   r'   r—   Úaudio_configr½   Útext_configr˜   Úlinear_1Úlinear_2rÂ   s     €r0   r?   z"GlmAsrMultiModalProjector.__init__J  sf   ø€ Ý‰Œ×ÒÑÔÐÝœ	 &Ô"5Ô"GÈÔI[ÔIgÐjkÑIkÑlÔlˆŒÝœ	 &Ô"4Ô"@À1Ñ"DÀfÔFXÔFdÑeÔeˆŒˆˆr/   )r+   r,   r-   r%   r?   rx   ry   s   @r0   rò   rò   I  sO   ø€ € € € € ðf˜|ð fð fð fð fð fð fð fð fð fð fr/   rò   z~
    The GlmAsr model which consists of a fine-tuned Whisper encoder, a multi-modal projector and a Llama language model.
    ©Úcustom_introc                   ó€   — e Zd Ze ed¬¦  «        dej        dej        dee	         de
ez  fd„¦   «         ¦   «         ZdS )	ÚGlmAsrModelzgCompute audio embeddings from log-mel input features using the audio encoder and multi-modal projector.rø   rÓ   Úinput_features_maskrR   rF   c                 ó
  —  | j         |fddi|¤Ž}|j        }|                     |j        d         d| j        j        j        ¦  «        }|                      |¦  «        }|                     d¦  «        }dD ]\  }}	}
|d|z  z   |	dz
  z
  dz
  |
z  dz   }Œd}||z
  |z  dz   }t          j
        |j        d         |j        ¬	¦  «        d d d …f         |d d …d f         k     }||                     |j        ¦  «                 |_        |S )
Nrg   Tr   r}   rI   r   r$   rH   rä   )Úaudio_towerræ   r­   r€   r’   rô   r½   Úmulti_modal_projectorÚsumr2   ré   rå   ÚtoÚpooler_output)r@   rÓ   rü   rR   Úaudio_outputsÚaudio_hidden_statesÚaudio_embedsrD   rK   rL   rM   rJ   Úpost_lengthsÚ
valid_masks                 r0   Úget_audio_featureszGlmAsrModel.get_audio_featuresV  sH  € ð )˜Ô(¨ÐTÐTÀTÐTÈVÐTÐTˆØ+Ô=ÐØ1×9Ò9ØÔ  Ô# R¨¬Ô)AÔ)Sñ
ô 
Ðð ×1Ò1Ð2EÑFÔFˆà+×/Ò/°Ñ3Ô3ˆØ,Bð 	`ð 	`Ñ(ˆG�[ &Ø*¨Q°©[Ñ8¸KÈ!¹OÑLÈqÑPÐU[Ñ[Ð^_Ñ_ˆMˆMØˆØ%¨Ñ4¸ÑEÈÑIˆå”\ ,Ô"4°QÔ"7ÀÔ@SÐTÑTÔTÐUYÐ[\Ð[\Ð[\ÐU\Ô]Ð`lÐmnÐmnÐmnÐptÐmtÔ`uÒuˆ
Ø&2°:·=²=ÀÔATÑ3UÔ3UÔ&VˆÔ#àÐr/   N)r+   r,   r-   r   r   r2   ÚFloatTensorrV   r   r   rl   r   r  r.   r/   r0   rû   rû   P  s�   € € € € € ð Ø€^Ø~ðñ ô ðàÔ)ðð #œ\ðð Ð+Ô,ð	ð
 
Ð+Ñ	+ðð ð ñô ñ Ôðð ð r/   rû   c                   ó  ‡ — e Zd ZddiZˆ fd„Z	 	 	 	 	 	 	 	 	 	 ddej        dz  dej        dz  dej        dz  d	ej        dz  d
ej        dz  de	dz  dej        dz  dej        dz  de
dz  deej        z  dee         defˆ fd„Zˆ xZS )ÚGlmAsrForConditionalGenerationzlm_head.weightz(model.language_model.embed_tokens.weightc                 óš   •— t          ¦   «                              |¦  «         t          |¦  «        | _        |                      ¦   «          d S rw   )r>   r?   rû   Úmodelrâ   rÂ   s     €r0   r?   z'GlmAsrForConditionalGeneration.__init__{  s@   ø€ Ý‰Œ×Ò˜Ñ Ô Ð Ý  Ñ(Ô(ˆŒ
Ø�ŠÑÔÐÐÐr/   Nr   Ú	input_idsrÓ   rü   r¤   r†   Úpast_key_valuesrê   ÚlabelsÚ	use_cacheÚlogits_to_keeprR   rF   c                 óH   •—  t          ¦   «         j        d|||||||	|
dœ|¤ŽS )a  
        input_features_mask (`torch.Tensor` of shape `(batch_size, feature_sequence_length)`):
            Mask to avoid performing attention on padding feature indices. Mask values selected in `[0, 1]`:

            - 1 for tokens that are **not masked**,
            - 0 for tokens that are **masked**.
        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
            Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
            config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
            (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.

        Example:

        ```python
        >>> from transformers import GlmAsrForConditionalGeneration, AutoProcessor

        >>> model_id = "zai-org/GLM-ASR-Nano-2512"
        >>> processor = AutoProcessor.from_pretrained(model_id)
        >>> model = GlmAsrForConditionalGeneration.from_pretrained(model_id, dtype="auto", device_map="auto")
        >>> inputs = processor.apply_transcription_request("https://huggingface.co/datasets/hf-internal-testing/dummy-audio-samples/resolve/main/bcn_weather.mp3")

        >>> inputs = inputs.to(model.device, dtype=model.dtype)

        >>> outputs = model.generate(**inputs, do_sample=False, max_new_tokens=500)

        >>> decoded_outputs = processor.batch_decode(outputs[:, inputs.input_ids.shape[1] :], skip_special_tokens=True)
        >>> print(decoded_outputs)
        ```)r  r¤   r†   r  rê   r  r  r  r.   )r>   r·   )r@   r  rÓ   rü   r¤   r†   r  rê   r  r  r  rR   rC   s               €r0   r·   z&GlmAsrForConditionalGeneration.forward€  sL   ø€ ðT �u‰wŒwŒð 

ØØ)Ø%Ø+Ø'ØØØ)ð

ð 

ð ð

ð 

ð 
	
r/   )
NNNNNNNNNr   )r+   r,   r-   Ú_tied_weights_keysr?   r2   Ú
LongTensorr	  rV   r	   Úboolr¹   r   r   r   r·   rx   ry   s   @r0   r  r  s  sJ  ø€ € € € € ð +Ð,VÐWÐðð ð ð ð ð .2Ø37Ø37Ø.2Ø04Ø(,Ø26Ø*.Ø!%Ø-.ð4
ð 4
àÔ# dÑ*ð4
ð Ô)¨DÑ0ð4
ð #œ\¨DÑ0ð	4
ð
 œ tÑ+ð4
ð Ô&¨Ñ-ð4
ð  ™ð4
ð Ô(¨4Ñ/ð4
ð Ô  4Ñ'ð4
ð ˜$‘;ð4
ð ˜eœlÑ*ð4
ð Ð+Ô,ð4
ð 
 ð4
ð 4
ð 4
ð 4
ð 4
ð 4
ð 4
ð 4
ð 4
ð 4
r/   r  )rÒ   r  rû   r5   rÐ   )Nr$   )HÚcollections.abcr   rY   ÚnpÚactivationsr   Úaudio_utilsr   r   r   Úcache_utilsr	   Úfeature_extraction_utilsr
   Úmodeling_layersr   Úmodeling_outputsr   r   Úmodeling_utilsr   Úprocessing_utilsr   Úutilsr   r   r   r   Úutils.genericr   r   r   Úutils.import_utilsr   Úutils.output_capturingr   Ú&audioflamingo3.modeling_audioflamingo3r   r   r   r   Ú(audioflamingo3.processing_audioflamingo3r   r   Úglm.modeling_glmr    Úllama.modeling_llamar!   r"   r#   Úconfiguration_glmasrr%   r&   r2   r'   Ú
get_loggerr+   Úloggerr)   r5   r{   r�   r‘   ÚModuler»   rÅ   rÐ   rÒ   rò   rû   r  Ú__all__r.   r/   r0   ú<module>r.     s€  ðð %Ð $Ð $Ð $Ð $Ð $à Ð Ð Ð à !Ð !Ð !Ð !Ð !Ð !Ø iÐ iÐ iÐ iÐ iÐ iÐ iÐ iÐ iÐ iØ  Ð  Ð  Ð  Ð  Ð  Ø 4Ð 4Ð 4Ð 4Ð 4Ð 4Ø 9Ð 9Ð 9Ð 9Ð 9Ð 9Ø RÐ RÐ RÐ RÐ RÐ RÐ RÐ RØ 5Ð 5Ð 5Ð 5Ð 5Ð 5Ø &Ð &Ð &Ð &Ð &Ð &Ø TÐ TÐ TÐ TÐ TÐ TÐ TÐ TÐ TÐ TÐ TÐ TØ _Ð _Ð _Ð _Ð _Ð _Ð _Ð _Ð _Ð _Ø *Ð *Ð *Ð *Ð *Ð *Ø 5Ð 5Ð 5Ð 5Ð 5Ð 5ðð ð ð ð ð ð ð ð ð ð ð ð nÐ mÐ mÐ mÐ mÐ mÐ mÐ mØ 1Ð 1Ð 1Ð 1Ð 1Ð 1Ø WÐ WÐ WÐ WÐ WÐ WÐ WÐ WÐ WÐ WØ CÐ CÐ CÐ CÐ CÐ CÐ CÐ Cð ÐÑÔð Ø€L€L€LØÐÐÐÐÐð 
ˆÔ	˜HÑ	%Ô	%€ð @Ð ?Ð ?Ð ?Ð ?Ð9Ñ ?Ô ?Ð ?ð 
€�:ÐÑÔØðk
ð k
ð k
ð k
ð k
Ð-ñ k
ô k
ñ „ñ Ôðk
ð\ 5Ð 4Ð 4Ð 4Ð 4Ð.Ñ 4Ô 4Ð 4ðð ð ð ð$ ð*)ð *)ð *)ð *)ð *)�nñ *)ô *)ñ Ôð*)ðZð ð ð ð �”	ñ ô ð ð ð  ð  ð  ð  Ð3ñ  ô  ð  ðF @Ð ?Ð ?Ð ?Ð ?Ð9Ñ ?Ô ?Ð ?ð(Kð (Kð (Kð (Kð (KÐ)ñ (Kô (Kð (KðVfð fð fð fð fÐ Añ fô fð fð €ððñ ô ð
ð ð ð ð Ð%ñ ô ñô ð
ð< €ððñ ô ð
<
ð <
ð <
ð <
ð <
Ð%Kñ <
ô <
ñô ð
<
ð~ð ð €€€r/   