dmx.compressor.modeling.hf.DmxPreTrainedModel
=============================================

.. currentmodule:: dmx.compressor.modeling.hf

.. autoclass:: DmxPreTrainedModel

   
   .. automethod:: __init__

   
   .. rubric:: Methods

   .. autosummary::
   
      ~DmxPreTrainedModel.__init__
      ~DmxPreTrainedModel.active_adapter
      ~DmxPreTrainedModel.active_adapters
      ~DmxPreTrainedModel.add_adapter
      ~DmxPreTrainedModel.add_memory_hooks
      ~DmxPreTrainedModel.add_model_tags
      ~DmxPreTrainedModel.add_module
      ~DmxPreTrainedModel.apply
      ~DmxPreTrainedModel.bfloat16
      ~DmxPreTrainedModel.buffers
      ~DmxPreTrainedModel.can_generate
      ~DmxPreTrainedModel.check_dim_consistency
      ~DmxPreTrainedModel.children
      ~DmxPreTrainedModel.compile
      ~DmxPreTrainedModel.compute_transition_scores
      ~DmxPreTrainedModel.configure
      ~DmxPreTrainedModel.counting_flops
      ~DmxPreTrainedModel.cpu
      ~DmxPreTrainedModel.create_extended_attention_mask_for_decoder
      ~DmxPreTrainedModel.create_submod_transform_forward
      ~DmxPreTrainedModel.cuda
      ~DmxPreTrainedModel.deepcopy_args
      ~DmxPreTrainedModel.delete_adapter
      ~DmxPreTrainedModel.dequantize
      ~DmxPreTrainedModel.disable_adapters
      ~DmxPreTrainedModel.disable_input_require_grads
      ~DmxPreTrainedModel.double
      ~DmxPreTrainedModel.enable_adapters
      ~DmxPreTrainedModel.enable_input_require_grads
      ~DmxPreTrainedModel.estimate_tokens
      ~DmxPreTrainedModel.eval
      ~DmxPreTrainedModel.extra_repr
      ~DmxPreTrainedModel.float
      ~DmxPreTrainedModel.floating_point_ops
      ~DmxPreTrainedModel.fold_weights_and_biases
      ~DmxPreTrainedModel.forward
      ~DmxPreTrainedModel.forward_weight_hypernets
      ~DmxPreTrainedModel.freeze
      ~DmxPreTrainedModel.from_pretrained
      ~DmxPreTrainedModel.generate
      ~DmxPreTrainedModel.get_adapter_state_dict
      ~DmxPreTrainedModel.get_buffer
      ~DmxPreTrainedModel.get_compiled_call
      ~DmxPreTrainedModel.get_extended_attention_mask
      ~DmxPreTrainedModel.get_extra_state
      ~DmxPreTrainedModel.get_head_mask
      ~DmxPreTrainedModel.get_input_embeddings
      ~DmxPreTrainedModel.get_memory_footprint
      ~DmxPreTrainedModel.get_monitoring_records
      ~DmxPreTrainedModel.get_output_embeddings
      ~DmxPreTrainedModel.get_parameter
      ~DmxPreTrainedModel.get_position_embeddings
      ~DmxPreTrainedModel.get_runtime_records
      ~DmxPreTrainedModel.get_submodule
      ~DmxPreTrainedModel.gradient_checkpointing_disable
      ~DmxPreTrainedModel.gradient_checkpointing_enable
      ~DmxPreTrainedModel.half
      ~DmxPreTrainedModel.heal_tokens
      ~DmxPreTrainedModel.init_weights
      ~DmxPreTrainedModel.invert_attention_mask
      ~DmxPreTrainedModel.ipu
      ~DmxPreTrainedModel.is_backend_compatible
      ~DmxPreTrainedModel.is_same_signature
      ~DmxPreTrainedModel.keep_dmx_config
      ~DmxPreTrainedModel.load_adapter
      ~DmxPreTrainedModel.load_state_dict
      ~DmxPreTrainedModel.measure_runtimes
      ~DmxPreTrainedModel.modules
      ~DmxPreTrainedModel.monitoring
      ~DmxPreTrainedModel.mtia
      ~DmxPreTrainedModel.named_buffers
      ~DmxPreTrainedModel.named_children
      ~DmxPreTrainedModel.named_dmx_modules
      ~DmxPreTrainedModel.named_modules
      ~DmxPreTrainedModel.named_parameters
      ~DmxPreTrainedModel.num_parameters
      ~DmxPreTrainedModel.parameters
      ~DmxPreTrainedModel.post_init
      ~DmxPreTrainedModel.post_process_gm
      ~DmxPreTrainedModel.prepare_inputs_for_generation
      ~DmxPreTrainedModel.print_model_tree
      ~DmxPreTrainedModel.process_inputs_for_export
      ~DmxPreTrainedModel.prune_heads
      ~DmxPreTrainedModel.push_to_hub
      ~DmxPreTrainedModel.register_backward_hook
      ~DmxPreTrainedModel.register_buffer
      ~DmxPreTrainedModel.register_for_auto_class
      ~DmxPreTrainedModel.register_forward_hook
      ~DmxPreTrainedModel.register_forward_pre_hook
      ~DmxPreTrainedModel.register_full_backward_hook
      ~DmxPreTrainedModel.register_full_backward_pre_hook
      ~DmxPreTrainedModel.register_load_state_dict_post_hook
      ~DmxPreTrainedModel.register_load_state_dict_pre_hook
      ~DmxPreTrainedModel.register_module
      ~DmxPreTrainedModel.register_parameter
      ~DmxPreTrainedModel.register_state_dict_post_hook
      ~DmxPreTrainedModel.register_state_dict_pre_hook
      ~DmxPreTrainedModel.requires_grad_
      ~DmxPreTrainedModel.reset_memory_hooks_state
      ~DmxPreTrainedModel.resize_position_embeddings
      ~DmxPreTrainedModel.resize_token_embeddings
      ~DmxPreTrainedModel.retrieve_modules_from_names
      ~DmxPreTrainedModel.reverse_bettertransformer
      ~DmxPreTrainedModel.save_pretrained
      ~DmxPreTrainedModel.set_adapter
      ~DmxPreTrainedModel.set_extra_state
      ~DmxPreTrainedModel.set_input_embeddings
      ~DmxPreTrainedModel.set_submodule
      ~DmxPreTrainedModel.share_memory
      ~DmxPreTrainedModel.state_dict
      ~DmxPreTrainedModel.tensor_parallel
      ~DmxPreTrainedModel.thaw
      ~DmxPreTrainedModel.tie_weights
      ~DmxPreTrainedModel.to
      ~DmxPreTrainedModel.to_baseline_mode
      ~DmxPreTrainedModel.to_basic_mode
      ~DmxPreTrainedModel.to_bettertransformer
      ~DmxPreTrainedModel.to_empty
      ~DmxPreTrainedModel.to_fp8_mode
      ~DmxPreTrainedModel.to_old_forward
      ~DmxPreTrainedModel.to_signature_key
      ~DmxPreTrainedModel.to_transformed_forward
      ~DmxPreTrainedModel.train
      ~DmxPreTrainedModel.transform
      ~DmxPreTrainedModel.type
      ~DmxPreTrainedModel.warn_if_padding_and_no_attention_mask
      ~DmxPreTrainedModel.xpu
      ~DmxPreTrainedModel.zero_grad
   
   

   
   
   .. rubric:: Attributes

   .. autosummary::
   
      ~DmxPreTrainedModel.T_destination
      ~DmxPreTrainedModel.base_model
      ~DmxPreTrainedModel.base_model_prefix
      ~DmxPreTrainedModel.call_super_init
      ~DmxPreTrainedModel.config_class
      ~DmxPreTrainedModel.device
      ~DmxPreTrainedModel.dmx_config
      ~DmxPreTrainedModel.dmx_module_names
      ~DmxPreTrainedModel.dtype
      ~DmxPreTrainedModel.dummy_inputs
      ~DmxPreTrainedModel.dump_patches
      ~DmxPreTrainedModel.framework
      ~DmxPreTrainedModel.is_gradient_checkpointing
      ~DmxPreTrainedModel.is_parallelizable
      ~DmxPreTrainedModel.loss_function
      ~DmxPreTrainedModel.main_input_name
      ~DmxPreTrainedModel.model_tags
      ~DmxPreTrainedModel.op_set
      ~DmxPreTrainedModel.supports_gradient_checkpointing
      ~DmxPreTrainedModel.supports_pp_plan
      ~DmxPreTrainedModel.supports_tp_plan
      ~DmxPreTrainedModel.training
      ~DmxPreTrainedModel.transformed
      ~DmxPreTrainedModel.additional_dmx_aware_mappings