a OSicWk@sddlZddlZddlZddlZddlmZddlmmZddl m Z ddl m Z m Z mZmZmZmZmZmZddlmZmZddlmZmZddlmZmZmZmZmZm Z ddl!m"Z"d d Z#d4d d Z$d5ddZ%ddZ&ddZ'd6ddZ(d7ddZ)ddZ*ddZ+d8ddZ,dd Z-d!d"Z.d9d#d$Z/dej0ddfd%d&Z1d:d'd(Z2d;d)d*Z3dd2d3Z7dS)?N) _FusedModule))get_default_dynamic_quant_module_mappings(get_default_static_quant_module_mappings2get_default_static_quant_reference_module_mappingsget_default_qat_module_mappings$get_default_qconfig_propagation_listno_observer_set_has_special_act_post_process_get_special_act_post_process)get_qparam_dict)has_no_children_ignoring_parametrizations) DeQuantStub QuantWrapper)add_module_to_qconfig_obs_ctrdefault_dynamic_qconfigfloat16_dynamic_qconfig!float_qparams_weight_only_qconfig&float_qparams_weight_only_qconfig_4bitactivation_is_memoryless)type_before_parametrizationscCs t|tjjjpt|tjjjSN) isinstancetorchao quantization ObserverBaseFakeQuantizeBasemoduler Z/var/www/html/django/DPS/env/lib/python3.9/site-packages/torch/ao/quantization/quantize.pyis_activation_post_processsr"c Cs|t||}|||}t|d|}tjjj||t||}||_| D]V\}}|rj|d|n|} |dus||dgvsRt ||dgvsRt |||| qRdS)aThis is a helper function for `propagate_qconfig_` Args: module: input module qconfig_dict: dictionary that maps from name of submodule to quantization configuration qconfig_parent: quantization config of parent module, we will fallback to this config when there is no specified config for current module prefix: corresponding prefix of the current module, used as key in qconfig_dict prepare_custom_config_dict: dictionary for custom handling of modules see docs for :func:`~torch.ao.quantization.prepare_fx` Return: None, module is modified inplace with qconfig attached qconfig.NZnon_traceable_module_nameZnon_traceable_module_class) getrgetattrrrrr$assert_valid_qconfigrnamed_childrentype_propagate_qconfig_helper) r qconfig_dictZqconfig_parentprefixprepare_custom_config_dictZmodule_qconfigZqconfig_with_device_checknamechild module_prefixr r r!r+$s    r+cCs*|dur i}|duri}t|||ddS)aPropagate qconfig through the module hierarchy and assign `qconfig` attribute on each leaf module Args: module: input module qconfig_dict: dictionary that maps from name or type of submodule to quantization configuration, qconfig applies to all submodules of a given module unless qconfig for the submodules are specified (when the submodule already has qconfig attribute) prepare_custom_config_dict: dictionary for custom handling of modules see docs for :func:`~torch.ao.quantization.prepare_fx` Return: None, module is modified inplace with qconfig attached N)r.)r+)rr,r.r r r!propagate_qconfig_Ls r2cCs ||S)z3Forward hook that calls observer on the output activation_post_process)selfinputoutputr r r!_observer_forward_hookbsr8cCs||dS)z7Forward pre hook that calls observer on the output rr3)r5r6r r r!_observer_forward_pre_hookgsr9FcCsTt|dsJd|r4|t}|jj|jddn|t}|jj|jdddS)Nr4zGExpect activation_post_process attribute already attached to the moduleF)last) hasattrregister_forward_pre_hookr9_forward_pre_hooks move_to_endidregister_forward_hookr8_forward_hooks)rpre_hookhandler r r!%register_activation_post_process_hookls   rDc s|durt}|duri}dur`t|}t|dksDJd|t|dkr\tt|ndd ddddd fd d }|D]\}}t|tj fvrqqt|t j t j fvrֈ|rԈ|j |_qt|tr|r||qt|rt|} ||| q|dur.get_activation_post_processcSst|do|jduS)Nr$r;r$)mr r r!needs_observationsz(add_observer_..needs_observationcs>|r:t|ts:|d|j|t|t|jddS)zn Adds an activation post process module and register a pre or post hook that calls the module r4rBN)rr add_moduler$rDr)rKrHrGrIrLr r!insert_activation_post_processs z5add_observer_..insert_activation_post_process)N)N)rget_unique_devices_lenformatnextiterr)rnnDropoutnnqFloatFunctional QFunctionalr$r4rrr r from_floatsetattrr add_observer_r r Sequential) rqconfig_propagation_listZnon_leaf_module_listrGcustom_module_class_mappingdevicesrPr/r0rHZobserved_childr rOr!r]wsP           r]cCs$dd|Ddd|DBS)NcSsh|] }|jqSr rG.0pr r r! z&get_unique_devices_..cSsh|] }|jqSr rbrcr r r!rfrg) parametersbuffersrr r r!rQsrQcCsDt|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)add_quant_dequant_modules)rr/r0r r r!rjs rjcCstjd|duri}|di}|s2t|}|}|durDt}t|ddtdd| Dspt 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.prepareNZ%float_to_observed_custom_module_classr,css|]}t|do|jVqdS)r$NrJ)rdrKr r r! rgzprepare..zNone of the submodule got qconfig applied. Make sure you passed correct configuration through `qconfig_dict` or by assigning the `.qconfig` attribute directly on submodules)r`) r_C_log_api_usage_oncer&copydeepcopyrr2anymoduleswarningswarnr])modelinplaceZ allow_listobserver_non_leaf_module_listr.r`r_r r r!prepares"     rycsDtdrtjrtddfdd }|dd|dddS)Nr4Fcs^|r jnj}|rtnt}t}|D]\}}||ur*||q*|D]}||qJdSr)r=rAr9r8setitemsaddpop)rBhook_mapZ observer_hookZhandle_ids_to_remove handle_idhook_fnrr r! remove_hooks's  z5_remove_activation_post_process..remove_hooksTrM)F)r;r"r4delattr)rrr rr!_remove_activation_post_processs   rcCs0|D] }t|qt|dr$|`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)rr0r r r!r5s    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.quantizeNTrw) rrnrorrprqevalryconvert)rvrun_fnrun_argsmappingrwr r r!quantizeDs   rc Csrtjd|dur|tjkrHtjttjttjttj ttj ttj ti}nt|tj kr|tjt tjt tjt tj t tj t tj t i}n@|tjkrtjttjti}n$|tjkrtjti}ntd|npt|tr.|tjurt}n@|tj urt }n0|tjurt}n |tjur t}ntdt|tt|t|}|dur>t}|sNt |}|!t"||t#||dd|S)avConverts a float model to dynamic (i.e. weights-only) quantized model. Replaces specified modules with dynamic weight-only quantized versions and output the quantized model. For simplest usage provide `dtype` argument that can be float16 or qint8. Weight-only quantization by default is performed for layers with large weights size - i.e. Linear and RNN variants. Fine grained control is possible with `qconfig` and `mapping` that act similarly to `quantize()`. If `qconfig` is provided, the `dtype` argument is ignored. Args: model: input model qconfig_spec: Either: - A dictionary that maps from name or type of submodule to quantization configuration, qconfig applies to all submodules of a given module unless qconfig for the submodules are specified (when the submodule already has qconfig attribute). Entries in the dictionary need to be QConfig instances. - A set of types and/or submodule names to apply dynamic quantization to, in which case the `dtype` argument is used to specify the bit-width inplace: carry out model transformations in-place, the original module is mutated mapping: maps type of a submodule to a type of corresponding dynamically quantized version with which the submodule needs to be replaced z*quantization_api.quantize.quantize_dynamicNzTDon't know how to quantize with default settings for {}. Provide full qconfig pleasez.Unknown dtype specified for quantize_dynamic: Tr)$rrnroqint8rVLinearrLSTMGRULSTMCellRNNCellGRUCellfloat16rquint8 EmbeddingBagr Embeddingquint4x2r ValueErrorrSrrz RuntimeErrorstrdictzip itertoolsrepeatrrprqrr2r)rvZ qconfig_specdtyperrwdefault_qconfigr r r!quantize_dynamic`s^             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 modeNrlTF)rrwremove_qconfig)rxrw) rrnrotrainingrrprqr2rryrzvalues)rvrrwr r r! prepare_qats   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)rrnrorprqtrainrr)rvrrrwr r r! quantize_qats    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)rw is_referenceconvert_custom_config_dict)rrnrorprq_convertr)rrrwrrrr r r!rs  rc Cs|dur|rtnt}|dur$i}|di}|s>t|}i}|D]>\}}t|tsxt||vrxt ||d||t |||||<qJ| D]\} } | |j | <q|S)aConverts submodules in input module to a different module according to `mapping` by calling `from_float` method on the target module class Args: module: input module mapping: a dictionary that maps from source module type to target module type, can be overwritten to allow swapping user defined Modules inplace: carry out model transformations in-place, the original module is mutated is_reference: a flag to enable quantized reference module NZ)observed_to_quantized_custom_module_classT) rrr&rprqr)rrrr swap_moduler{rk) rrrwrrr`Zreassignr/modkeyvaluer r r!rs*      rc CsJ|}t|drF|jdurFd}t||vrD|t||}d}nnt||vr|t|}t|dr|jr|jduszJ|j}||jt|}|||}n ||}d}|rF|j D]}| |q|j D]} | t ur| | qt|} t| dksJd| t| dkr2tt| nd} | rF|| |S) a Swaps the module if it has a quantized counterpart and it has an `observer` attached. Args: mod: input module mapping: a dictionary that maps from nn module to nnq module Return: The corresponding quantized module of `mod` r$NFT _IS_REFERENCEr zQswap_module only works with cpu or single-device CUDA modules, but got devices {}r)r;r$r from_observedrweightr r[r=rr<rAr8r@rQrRrSrTrUrF) rrr`new_modswappedqmodweight_post_processweight_qparams pre_hook_fnrrarGr r r!r6s@          rcCsZdd}t|dr$|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 )r-r r r! get_prefixpsz%get_observer_dict..get_prefixr4N)r;r4r)get_observer_dict)r target_dictr-rr/r0r1r r r!rhs  r)Nr#N)NN)F)NNNN)FNNN)NF)NF)F)NFTFN)NFFN)r#)8rprrtrtorch.nnrVtorch.nn.quantized quantizedrXtorch.nn.intrinsicrZ+torch.ao.quantization.quantization_mappingsrrrrrrr r utilsr r torch.ao.quantization.stubsrrtorch.ao.quantization.qconfigrrrrrrZtorch.nn.utils.parametrizerr"r+r2r8r9rDr]rQrjryrrrrrrrrrrrr r r r!sR  (    (  V 8  U   ( (2