a dn@sddlZddlZddlZddlZddlmZddlmmmZ ddl m Z ddl m Z mZmZmZmZmZmZmZddlmZmZddlmZmZddlmZmZmZmZm Z m!Z!ddl"m#Z#dd l$m%Z%dd l$m%Z&gd Z'ej(ej)j(ej*ej)j*iej)j(ejj(ej)j*ejj*id Z+d d Z,d7ddZ-d8ddZ.ddZ/ddZ0d9ddZ1d:ddZ2ddZ3ddZ4d;d d!Z5d"d#Z6d$d%Z7dd,d-Zd3d4Z?dAd5d6Z@dS)BN) _FusedModule))get_default_dynamic_quant_module_mappings(get_default_static_quant_module_mappings2get_default_static_quant_reference_module_mappingsget_default_qat_module_mappings$get_default_qconfig_propagation_listno_observer_set_has_special_act_post_process_get_special_act_post_process)get_qparam_dict)has_no_children_ignoring_parametrizations) DeQuantStub QuantWrapper)_add_module_to_qconfig_obs_ctrdefault_dynamic_qconfigfloat16_dynamic_qconfig!float_qparams_weight_only_qconfig&float_qparams_weight_only_qconfig_4bit_activation_is_memoryless)type_before_parametrizations)_is_activation_post_process) get_default_custom_config_dictpropagate_qconfig_add_quant_dequantpreparequantizequantize_dynamic prepare_qat quantize_qatconvert swap_module)%float_to_observed_custom_module_class)observed_to_quantized_custom_module_classcCstS)z,Defines the default custom config dict. )_DEFAULT_CUSTOM_CONFIG_DICTr%r%g/var/www/html/stable-diffusion-webui/venv/lib/python3.9/site-packages/torch/ao/quantization/quantize.pyr=src Cs|t||}|||}t|d|}tjjj||t||}||_| D]V\}}|rj|d|n|} |dus||dgvsRt ||dgvsRt |||| qRdS)aThis is a helper function for `propagate_qconfig_` Args: module: input module qconfig_dict: dictionary that maps from name of submodule to quantization configuration qconfig_parent: quantization config of parent module, we will fallback to this config when there is no specified config for current module prefix: corresponding prefix of the current module, used as key in qconfig_dict prepare_custom_config_dict: dictionary for custom handling of modules see docs for :func:`~torch.ao.quantization.prepare_fx` Return: None, module is modified inplace with qconfig attached qconfig.NZnon_traceable_module_nameZnon_traceable_module_class) getrgetattrtorchaoZ quantizationr(Z_assert_valid_qconfigrnamed_childrentype_propagate_qconfig_helper) module qconfig_dictZqconfig_parentprefixprepare_custom_config_dictZmodule_qconfigZqconfig_with_device_checknamechild module_prefixr%r%r&r0Bs    r0cCs*|dur i}|duri}t|||ddS)aPropagate qconfig through the module hierarchy and assign `qconfig` attribute on each leaf module Args: module: input module qconfig_dict: dictionary that maps from name or type of submodule to quantization configuration, qconfig applies to all submodules of a given module unless qconfig for the submodules are specified (when the submodule already has qconfig attribute) prepare_custom_config_dict: dictionary for custom handling of modules see docs for :func:`~torch.ao.quantization.prepare_fx` Return: None, module is modified inplace with qconfig attached N)r4)r0)r1r2r4r%r%r&rjs rcCs ||S)z3Forward hook that calls observer on the output activation_post_process)selfinputoutputr%r%r&_observer_forward_hooksr=cCs||dS)z7Forward pre hook that calls observer on the output rr8)r:r;r%r%r&_observer_forward_pre_hooksr>FcCs8t|dsJd|r&|jtdd}n|jtdd}dS)Nr9zGExpect activation_post_process attribute already attached to the moduleT)prepend)hasattrregister_forward_pre_hookr>register_forward_hookr=)r1pre_hookhandler%r%r&&_register_activation_post_process_hooks rEc s|durt}|duri}dur`t|}t|dksDJd|t|dkr\tt|ndd ddddd fd d }|D]\}}t|tj fvrqqt|t j t j fvrֈ|rԈ|j |_qt|tr|r||q|durt||vr|r||qt|r.get_activation_post_processcSst|do|jduS)Nr(r@r()mr%r%r&needs_observationsz)_add_observer_..needs_observationcs>|r:t|ts:|d|j|t|t|jddS)zn Adds an activation post process module and register a pre or post hook that calls the module r9rCN) isinstancerZ add_moduler(rEr)rMrJrIrKrNr%r&insert_activation_post_processs z6_add_observer_..insert_activation_post_process)N)N)r_get_unique_devices_lenformatnextiterr.rnnZDropoutnnqZFloatFunctionalZ QFunctionalr(r9rPrr r from_floatsetattrr_add_observer_r r,Z Sequential) r1qconfig_propagation_listZnon_leaf_module_listrIcustom_module_class_mappingdevicesrRr5r6rJZobserved_childr%rQr&r\sP           r\cCs$dd|Ddd|DBS)NcSsh|] }|jqSr%rI.0pr%r%r& z'_get_unique_devices_..cSsh|] }|jqSr%r`rar%r%r&rdre) parametersbuffersr1r%r%r&rSsrScCsDt|r t|dr |jr t|S|D]\}}t||j|<q(|S)a{Wrap the leaf child module in QuantWrapper if it has a valid qconfig Note that this function will modify the children of module inplace and it can return a new module which wraps the input module as well. Args: module: input module with qconfig attributes for all the leaf modules that we want to quantize Return: Either the inplace modified module with submodules wrapped in `QuantWrapper` based on qconfig or a new `QuantWrapper` module which wraps the input module, the latter case only happens when the input module is a leaf module and we want to quantize it. r()r r@r(rr.r_modules)r1r5r6r%r%r&rs rcCstjd|durt}|di}|s4t|}|}|durFt}t|ddt dd| Dsrt dt ||||d|S) aPrepares a copy of the model for quantization calibration or quantization-aware training. Quantization configuration should be assigned preemptively to individual submodules in `.qconfig` attribute. The model will be attached with observer or fake quant modules, and qconfig will be propagated. Args: `model`: input model to be modified in-place `inplace`: carry out model transformations in-place, the original module is mutated `allow_list`: list of quantizable modules `observer_non_leaf_module_list`: list of non-leaf modules we want to add observer `prepare_custom_config_dict`: customization configuration dictionary for prepare function .. code-block:: python # Example of prepare_custom_config_dict: prepare_custom_config_dict = { # user will manually define the corresponding observed # module class which has a from_float class method that converts # float custom module to observed custom module "float_to_observed_custom_module_class": { CustomModule: ObservedCustomModule } } z!quantization_api.quantize.prepareNr"r2css|]}t|do|jVqdS)r(NrL)rbrMr%r%r& 5rezprepare..zNone of the submodule got qconfig applied. Make sure you passed correct configuration through `qconfig_dict` or by assigning the `.qconfig` attribute directly on submodules)r^)r,_C_log_api_usage_oncerr*copydeepcopyrranymoduleswarningswarnr\)modelinplaceZ allow_listobserver_non_leaf_module_listr4r^r]r%r%r&rs"     rcsDtdrtjrtddfdd }|dd|dddS)Nr9Fcs^|r jnj}|rtnt}t}|D]\}}||ur*||q*|D]}||qJdSrF)_forward_pre_hooks_forward_hooksr>r=setitemsaddpop)rCZhook_mapZ observer_hookZhandle_ids_to_removeZ handle_idhook_fnrhr%r& remove_hooksGs  z5_remove_activation_post_process..remove_hooksTrO)F)r@rr9delattr)r1r~r%rhr&_remove_activation_post_process?s   rcCs0|D] }t|qt|dr$|`t|dS)zClean up the qconfig left in the module so that new qconfig can be propagated. Args: module: module to be cleaned up r(N)children_remove_qconfigr@r(r)r1r6r%r%r&rUs    rcCs^tjd|durt}|s(t|}|t|dd||g|Rt||dd|S)aQuantize the input float model with post training static quantization. First it will prepare the model for calibration, then it calls `run_fn` which will run the calibration step, after that we will convert the model to a quantized model. Args: model: input float model run_fn: a calibration function for calibrating the prepared model run_args: positional arguments for `run_fn` inplace: carry out model transformations in-place, the original module is mutated mapping: correspondence between original module types and quantized counterparts Return: Quantized model. z"quantization_api.quantize.quantizeNTru) r,rlrmrrnroevalrr )rtrun_fnrun_argsmappingrur%r%r&rds   rc Csrtjd|dur|tjkrHtjttjttjttj ttj ttj ti}nt|tj kr|tjt tjt tjt tj t tj t tj t i}n@|tjkrtjttjti}n$|tjkrtjti}ntd|npt|tr.|tjurt}n@|tj urt }n0|tjurt}n |tjur t}ntdt|tt|t|}|dur>t}|sNt |}|!t"||t#||dd|S)avConverts a float model to dynamic (i.e. weights-only) quantized model. Replaces specified modules with dynamic weight-only quantized versions and output the quantized model. For simplest usage provide `dtype` argument that can be float16 or qint8. Weight-only quantization by default is performed for layers with large weights size - i.e. Linear and RNN variants. Fine grained control is possible with `qconfig` and `mapping` that act similarly to `quantize()`. If `qconfig` is provided, the `dtype` argument is ignored. Args: model: input model qconfig_spec: Either: - A dictionary that maps from name or type of submodule to quantization configuration, qconfig applies to all submodules of a given module unless qconfig for the submodules are specified (when the submodule already has qconfig attribute). Entries in the dictionary need to be QConfig instances. - A set of types and/or submodule names to apply dynamic quantization to, in which case the `dtype` argument is used to specify the bit-width inplace: carry out model transformations in-place, the original module is mutated mapping: maps type of a submodule to a type of corresponding dynamically quantized version with which the submodule needs to be replaced z*quantization_api.quantize.quantize_dynamicNzTDon't know how to quantize with default settings for {}. Provide full qconfig pleasez.Unknown dtype specified for quantize_dynamic: Tr)$r,rlrmqint8rXZLinearrLSTMZGRUZLSTMCellZRNNCellZGRUCellfloat16rZquint8Z EmbeddingBagrZ EmbeddingZquint4x2r ValueErrorrUrPry RuntimeErrorstrdictzip itertoolsrepeatrrnrorrr )rtZ qconfig_specZdtyperruZdefault_qconfigr%r%r&rs^             rcCsltjd|jsJd|dur(t}|s6t|}t|ddt||dddt |t | dd|S) a Prepares a copy of the model for quantization calibration or quantization-aware training and converts it to quantized version. Quantization configuration should be assigned preemptively to individual submodules in `.qconfig` attribute. Args: model: input model to be modified in-place mapping: dictionary that maps float modules to quantized modules to be replaced. inplace: carry out model transformations in-place, the original module is mutated z%quantization_api.quantize.prepare_qatz1prepare_qat only works on models in training modeNrjTF)rruremove_qconfig)rvru) r,rlrmZtrainingrrnrorr rryvalues)rtrrur%r%r&rs   rcCsNtjd|st|}|t|dd||g|Rt|dd|S)agDo quantization aware training and output a quantized model Args: model: input model run_fn: a function for evaluating the prepared model, can be a function that simply runs the prepared model or a training loop run_args: positional arguments for `run_fn` Return: Quantized model. z&quantization_api.quantize.quantize_qatTr)r,rlrmrnroZtrainrr )rtrrrur%r%r&rs    rTcCs<tjd|st|}t||d||d|r8t||S)agConverts submodules in input module to a different module according to `mapping` by calling `from_float` method on the target module class. And remove qconfig at the end if remove_qconfig is set to True. Args: `module`: prepared and calibrated module `mapping`: a dictionary that maps from source module type to target module type, can be overwritten to allow swapping user defined Modules `inplace`: carry out model transformations in-place, the original module is mutated `convert_custom_config_dict`: custom configuration dictionary for convert function .. code-block:: python # Example of convert_custom_config_dict: convert_custom_config_dict = { # user will manually define the corresponding quantized # module class which has a from_observed class method that converts # observed custom module to quantized custom module "observed_to_quantized_custom_module_class": { ObservedCustomModule: QuantizedCustomModule } } z!quantization_api.quantize.convertT)ru is_referenceconvert_custom_config_dict)r,rlrmrnro_convertr)r1rrurrrr%r%r&r s  r c Cs|dur|rtnt}|dur&t}|di}|s@t|}i}|D]>\}}t|tszt ||vrzt ||d||t |||||<qL| D]\} } | |j | <q|S)aConverts submodules in input module to a different module according to `mapping` by calling `from_float` method on the target module class Args: module: input module mapping: a dictionary that maps from source module type to target module type, can be overwritten to allow swapping user defined Modules inplace: carry out model transformations in-place, the original module is mutated is_reference: a flag to enable quantized reference module Nr#T)rrrr*rnror.rPrrrr!rzri) r1rrurrr^Zreassignr5modkeyvaluer%r%r&r.s*      rc CsJ|}t|drF|jdurFd}t||vrD|t||}d}nnt||vr|t|}t|dr|jr|jduszJ|j}||jt|}|||}n ||}d}|rF|j D]}| |q|j D]} | t ur| | qt|} t| dksJd| t| dkr2tt| nd} | rF|| |S) a Swaps the module if it has a quantized counterpart and it has an `observer` attached. Args: mod: input module mapping: a dictionary that maps from nn module to nnq module Return: The corresponding quantized module of `mod` r(NFT _IS_REFERENCEr zQswap_module only works with cpu or single-device CUDA modules, but got devices {}r)r@r(rZ from_observedrZweightr rZrwrrArxr=rBrSrTrUrVrWrH) rrr^new_modZswappedZqmodZweight_post_processZweight_qparamsZ pre_hook_fnr}r_rIr%r%r&r!Vs@          r!cCsZdd}t|dr$|j|||d<|D](\}}|rD|||n|}t|||q,dS)a,Traverse the modules and save all observers into dict. This is mainly used for quantization accuracy debug Args: mod: the top module we want to save all observers prefix: the prefix for the current module target_dict: the dictionary used to save all the observers cSs|dkr |S|dS)Nr'r)r%)r3r%r%r& get_prefixsz&_get_observer_dict..get_prefixr9N)r@r9r._get_observer_dict)r target_dictr3rr5r6r7r%r%r&rs  r)Nr'N)NN)F)NNNN)FNNN)NF)NF)F)NFTFN)NFFN)r')Arnrrrr,Ztorch.nnrXZtorch.ao.nn.quantizedr-Z quantizedrYZtorch.ao.nn.intrinsicrZ+torch.ao.quantization.quantization_mappingsrrrrrrr r utilsr r Ztorch.ao.quantization.stubsrrZtorch.ao.quantization.qconfigrrrrrrZtorch.nn.utils.parametrizerZtorch.ao.quantization.observerrZis_activation_post_process__all__rZ quantizableZMultiheadAttentionr$rr0rr=r>rEr\rSrrrrrrrrrr rr!rr%r%r%r&sf  (           (  V 8  U   ( (2