modeling_qwen2
==============

.. py:module:: modeling_qwen2


Classes
-------

.. autoapisummary::

   modeling_qwen2.Qwen2RMSNorm
   modeling_qwen2.Qwen2PreTrainedModel
   modeling_qwen2.Qwen2Model
   modeling_qwen2.Qwen2ForCausalLM
   modeling_qwen2.Qwen2ForSequenceClassification
   modeling_qwen2.Qwen2ForTokenClassification
   modeling_qwen2.Qwen2ForQuestionAnswering


Module Contents
---------------

.. py:class:: Qwen2RMSNorm(hidden_size, eps = 1e-06, quant_bits=16)

   Bases: :py:obj:`pymllm.backends.qualcomm.transformers.core.rms_norm.QRMSNorm`


.. py:class:: Qwen2PreTrainedModel

   Bases: :py:obj:`transformers.modeling_utils.PreTrainedModel`


   .. py:attribute:: config
      :type:  transformers.models.qwen2.configuration_qwen2.Qwen2Config


   .. py:attribute:: base_model_prefix
      :value: 'model'



   .. py:attribute:: supports_gradient_checkpointing
      :value: True



.. py:class:: Qwen2Model(config)

   Bases: :py:obj:`Qwen2PreTrainedModel`


   .. py:attribute:: padding_idx


   .. py:attribute:: vocab_size


   .. py:attribute:: embed_tokens


   .. py:attribute:: layers


   .. py:attribute:: norm


   .. py:attribute:: rotary_emb


   .. py:attribute:: gradient_checkpointing
      :value: False



   .. py:attribute:: has_sliding_layers


   .. py:attribute:: sin_embedding_input_qdq


   .. py:attribute:: cos_embedding_input_qdq


   .. py:attribute:: norm_input_qdq


   .. py:method:: convert_rope_for_deploy()


   .. py:method:: forward(input_ids = None, attention_mask = None, position_ids = None, past_key_values = None, inputs_embeds = None, use_cache = None, cache_position = None, **kwargs)


.. py:class:: Qwen2ForCausalLM(config)

   Bases: :py:obj:`Qwen2PreTrainedModel`, :py:obj:`transformers.generation.GenerationMixin`


   .. py:attribute:: config


   .. py:attribute:: model


   .. py:attribute:: vocab_size


   .. py:attribute:: lm_head


   .. py:attribute:: mllm_qualcomm_max_length
      :value: None



   .. py:attribute:: lm_head_input_qdq


   .. py:attribute:: lm_head_output_qdq


   .. py:method:: copy_lm_head_weight_from_embed_tokens()


   .. py:method:: forward(input_ids = None, attention_mask = None, position_ids = None, past_key_values = None, inputs_embeds = None, labels = None, use_cache = None, cache_position = None, logits_to_keep = 0, **kwargs)

      Example:

      ```python
      >>> from transformers import AutoTokenizer, Qwen2ForCausalLM

      >>> model = Qwen2ForCausalLM.from_pretrained("meta-qwen2/Qwen2-2-7b-hf")
      >>> tokenizer = AutoTokenizer.from_pretrained("meta-qwen2/Qwen2-2-7b-hf")

      >>> prompt = "Hey, are you conscious? Can you talk to me?"
      >>> inputs = tokenizer(prompt, return_tensors="pt")

      >>> # Generate
      >>> generate_ids = model.generate(inputs.input_ids, max_length=30)
      >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
      "Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you."
      ```



.. py:class:: Qwen2ForSequenceClassification

   Bases: :py:obj:`transformers.modeling_layers.GenericForSequenceClassification`, :py:obj:`Qwen2PreTrainedModel`


.. py:class:: Qwen2ForTokenClassification

   Bases: :py:obj:`transformers.modeling_layers.GenericForTokenClassification`, :py:obj:`Qwen2PreTrainedModel`


.. py:class:: Qwen2ForQuestionAnswering

   Bases: :py:obj:`transformers.modeling_layers.GenericForQuestionAnswering`, :py:obj:`Qwen2PreTrainedModel`


   .. py:attribute:: base_model_prefix
      :value: 'transformer'