AVhdZddlZddlmZddlmZddlmZddlm Z ddl m Z ddlmZdd l mZdd l mZdd l mZdd lmZdd lmZddlmZddlmZm Z m!Z!ddl"m#Z#edZ$dde#e!ejJfde#e!e$ffdZ&edejNe&Z(de#e!ejJfde#e!e$ffdZ)edZ*dde#e!ejJfde#e!e*ffdZ+edejNe+Z,de#e!ejJfde#e!e*ffdZ-edZ.dde#e!ejJfde#e!e.fde/fdZ0edejNe0Z1de#e!ejJfde#e!e.fde/fdZ2dd e#e!ejfffd!Z4ed"ejNe4Z5d e#e!ejfffd#Z6dde#e!ejJfd$e/fd%Z7ed&ejNe7Z8de#e!ejJfd$e/fd'Z9dde#e!ejJffd(Z:ed)ejNe:Z;de#e!ejJffd*Zd+e#e!ejJfd,e#e!ejfffd/Z?dd1e@d2e@d,e#e!ejJffd3ZAed4ejNeAZBd1e@d2e@d,e#e!ejJffd5ZCed6ZDdde#e!ejJfd7eDd,e#e!eDffd8ZEed9ejNeEZFde#e!ejJfd7eDd,e#e!eDffd:ZGed;ZHedZIdde#e!ejJfd?e#e!eIfd7eHd@eJdAe/d,e#e!eHff dBZKedCejNeKZLde#e!ejJfd?e#e!eIfd7eHd@eJdAe/d,e#e!eHff dDZMedEZNedFd=d>ZOdde#e!ejJfd?e#e!eOfd7eNd,e#e!eNffdGZPedHejNePZQde#e!ejJfd?e#e!eOfd7eNd,e#e!eNffdIZRedJdKdLdMdNdOdPdQd=d>dRdSdTdUdVdWdXdYdZd[ZSed\d=d>ZTdde#e!ejJfd?e#e!eTfd]e#e!eSffd^ZUed_ejNeUZVde#e!ejJfd?e#e!eTfd]e#e!eSffd`ZWedadKdLdMdNdOdPdQd=d>dRdSdTdUdVdWdXdYdZd[ZXedbd=d>ZYdde#e!ejJfd?e#e!eYfd]e#e!eXffdcZZeddejNeZZ[de#e!ejJfd?e#e!eYfd]e#e!eXffdeZ\edfdKdLdMdNdOdPdQd=d>dRdSdTdUdVdWdXdYdZd[Z]edgd=d>Z^dde#e!ejJfd?e#e!e^fd]e#e!e]ffdhZ_ediejNe_Z`de#e!ejJfd?e#e!e^fd]e#e!e]ffdjZaedkdKdLdMdNdOdPdQd=d>dRdSdTdUdVdWdXdYdZd[Zbedld=d>Zcdde#e!ejJfd?e#e!ecfd]e#e!ebffdmZdednejNedZede#e!ejJfd?e#e!ecfd]e#e!ebffdoZfedpdKdLdMdNdOdPdQd=d>dRdSdTdUdVdWdXdYdZd[Zgedqd=d>Zhdde#e!ejJfd?e#e!ehfd]e#e!egffdrZiedsejNeiZjde#e!ejJfd?e#e!ehfd]e#e!egffdtZkedudKdLdMdNdOdPdQd=d>dRdSdTdUdVdWdXdYdZd[Zledvd=d>Zmdde#e!ejJfd?e#e!emfd]e#e!elffdwZnedxejNenZode#e!ejJfd?e#e!emfd]e#e!elffdyZpedzZqed{d=d>Zrdde#e!ejJfd?e#e!erfd]e#e!eqffd|Zsed}ejNesZtde#e!ejJfd?e#e!erfd]e#e!eqffd~ZuedZvd0d0d0gdfd7evd1e@d2e@de@d,e#e!ejJff dZwedejNewZxd7evd1e@d2e@de@d,e#e!ejJff dZydde#e!ejJfd,e#e!ejffdZ{edejNe{Z|de#e!ejJfd,e#e!ejffdZ}edd=d>Z~e jdfde#e!ejJfde~d,e#e!e~ffdZedejNeZde#e!ejJfde~d,e#e!e~ffdZy)zUPython wrappers around TensorFlow ops. This file is MACHINE GENERATED! Do not edit. N) pywrap_tfe)context)core)execute)dtypes)annotation_types)op_def_registry)ops)op_def_library)deprecated_endpoints)dispatch) tf_export)TypeVarListAny) Annotated) TV_AssignAddVariableOp_dtype_atypes.BFloat16 _atypes.Bool_atypes.Complex128_atypes.Complex64_atypes.Float16_atypes.Float32_atypes.Float64_atypes.Float8e4m3b11fnuz_atypes.Float8e4m3fn_atypes.Float8e4m3fnuz_atypes.Float8e5m2_atypes.Float8e5m2fnuz _atypes.Half _atypes.Int16 _atypes.Int32 _atypes.Int4 _atypes.Int64 _atypes.Int8_atypes.QInt16_atypes.QInt32 _atypes.QInt8_atypes.QUInt16_atypes.QUInt8_atypes.Resource_atypes.String_atypes.UInt16_atypes.UInt32 _atypes.UInt4_atypes.UInt64 _atypes.UInt8_atypes.Variantresourcevaluectjxstj}|j}|jr t j |d|||}|Stjd|||\}}}} |S#t j$r }tj||Yd}~nd}~wt j$rYnwxYw t||||S#t j$rYwxYw)aAdds a value to the current value of a variable. Any ReadVariableOp with a control dependency on this op is guaranteed to see the incremented value or a subsequent newer one. Args: resource: A `Tensor` of type `resource`. handle to the resource in which to store the variable. value: A `Tensor`. the value by which the variable will be incremented. name: A name for the operation (optional). Returns: The created Operation. AssignAddVariableOpNnamectxr3r4r8)_contextr_thread_local_datais_eagerrTFE_Py_FastPathExecute_core_NotOkStatusException_opsraise_from_not_ok_status_FallbackException%assign_add_variable_op_eager_fallback_SymbolicException_op_def_library_apply_op_helper r3r4r8_ctxtld_resulte__op_outputss _/home/dcms/DCMS/lib/python3.12/site-packages/tensorflow/python/ops/gen_resource_variable_ops.pyassign_add_variable_oprQ    0h..0$ #\\ 11 #T8U.> ?(E", [ !&   3Q|#)s ?' ' .) TV_AssignSubVariableOp_dtyperrrrrrrrrrrrr r!r"r#r$r%r&r'r(r)r*r+r,r-r.r/r0r1r2ctjxstj}|j}|jr t j |d|||}|Stjd|||\}}}} |S#t j$r }tj||Yd}~nd}~wt j$rYnwxYw t||||S#t j$rYwxYw)aSubtracts a value from the current value of a variable. Any ReadVariableOp with a control dependency on this op is guaranteed to see the decremented value or a subsequent newer one. Args: resource: A `Tensor` of type `resource`. handle to the resource in which to store the variable. value: A `Tensor`. the value by which the variable will be incremented. name: A name for the operation (optional). Returns: The created Operation. AssignSubVariableOpNr7r:)r;rr<r=rr>r?r@rArBrC%assign_sub_variable_op_eager_fallbackrErFrGrHs rPassign_sub_variable_oprhNrRrSzraw_ops.AssignSubVariableOpctj|g|g\}\}tj|tj }||g}d|f}tj dd||||}d}|S)NrUsAssignSubVariableOprrVrYr^s rPrgrgtrbrc) TV_AssignVariableOp_dtyperrrrrrrrrrrrr r!r"r#r$r%r&r'r(r)r*r+r,r-r.r/r0r1r2validate_shapec tjxstj}|j}|jr t j |d|||d|}|S|d}tj|d}tj d||||\}}} } | S#t j$r }tj||Yd}~nd}~wt j$rYnwxYw t|||||S#t j$rYwxYw)aAssigns a new value to a variable. Any ReadVariableOp with a control dependency on this op is guaranteed to return this value or a subsequent newer value of the variable. Args: resource: A `Tensor` of type `resource`. handle to the resource in which to store the variable. value: A `Tensor`. the value to set the new tensor to use. validate_shape: An optional `bool`. Defaults to `False`. name: A name for the operation (optional). Returns: The created Operation. AssignVariableOprkN)rkr8r9F)r3r4rkr8)r;rr<r=rr>r?r@rArBrC!assign_variable_op_eager_fallbackrErZ make_boolrFrG) r3r4rkr8rIrJrKrLrMrNrOs rPassign_variable_oprps    0h..0$ #\\ 11  $%9IgnN%%n6FG.'88XU+9F!QX *!  & &- ##At,,  # #    . E.tOO  # #   s0BC+CCC#C33D D zraw_ops.AssignVariableOpc |d}tj|d}tj|g|g\}\}tj|t j }||g}d|d|f}tjdd||||}d}|S)NFrkrUsAssignVariableOprrV)rZror[rAr\r]r3r) r3r4rkr8r9r_r`rarKs rPrnrnsN%%n6FG."995'3K+x  # #Hg.>.> ?(E", ["2N C&   0!L#)s ?' ' .rc mutex_lockctjxstj}|j}|jr t j |d||}|Stjd||\}}}}|S#t j$r }tj||Yd}~nd}~wt j$rYnwxYw t|||S#t j$rYwxYw)aThis op consumes a lock created by `MutexLock`. This op exists to consume a tensor created by `MutexLock` (other than direct control dependencies). It should be the only that consumes the tensor, and will raise an error if it is not. Its only purpose is to keep the mutex lock tensor alive until it is consumed by this op. **NOTE**: This operation must run on the same device as its input. This may be enforced via the `colocate_with` mechanism. Args: mutex_lock: A `Tensor` of type `variant`. A tensor returned by `MutexLock`. name: A name for the operation (optional). Returns: The created Operation. ConsumeMutexLockNr7)rrr8)r;rr<r=rr>r?r@rArBrC!consume_mutex_lock_eager_fallbackrErFrG) rrr8rIrJrKrLrMrNrOs rPconsume_mutex_lockrvs&    0h..0$ #\\ 11  $ 4g n(88z>!QX *  & &- ##At,,  # #    . 4T++  # #   0A99C B''C?C CC('C(zraw_ops.ConsumeMutexLockctj|tj}|g}d}t j dd||||}d}|S)NsConsumeMutexLockrrV)rAr\r]variantrZr)rrr8r9r`rarKs rPrurusL%%j'//B*, &   0!L#)s ?' ' .rcignore_lookup_errorctjxstj}|j}|jr t j |d||d|}|S|d}tj|d}tj d|||\}}}} |S#t j$r }tj||Yd}~nd}~wt j$rYnwxYw t||||S#t j$rYwxYw)aDeletes the resource specified by the handle. All subsequent operations using the resource will result in a NotFound error status. Args: resource: A `Tensor` of type `resource`. handle to the resource to delete. ignore_lookup_error: An optional `bool`. Defaults to `True`. whether to ignore the error when the resource doesn't exist. name: A name for the operation (optional). Returns: The created Operation. DestroyResourceOprzN)rzr8r9T)r3rzr8)r;rr<r=rr>r?r@rArBrC"destroy_resource_op_eager_fallbackrErZrorFrG) r3rzr8rIrJrKrLrMrNrOs rPdestroy_resource_opr~s    0h..0$ #\\ 11 !43Hgn  **+>@UV'88h1D"&(!QX *%  & &- ##At,,  # #    / (;$  # #   s0BC)CCC!C00DDzraw_ops.DestroyResourceOpc|d}tj|d}tj|tj }|g}d|f}tj dd||||}d}|S)NTrzsDestroyResourceOprrV)rZrorAr\r]r3r)r3rzr8r9r`rarKs rPr}r}su  **+>@UV  # #Hg.>.> ?(, !#6 7&   11\#)s ?' ' .rcctjxstj}|j}|jr t j |d||}|Stjd||\}}}}|S#t j$r }tj||Yd}~nd}~wt j$rYnwxYw t|||S#t j$rYwxYw)abTurns off the copy-on-read mode. Turns off the copy-on-read mode of a resource variable. If the variable is not in copy-on-read mode, this op has no effect. Args: resource: A `Tensor` of type `resource`. The resource handle of the resource variable. name: A name for the operation (optional). Returns: The created Operation. DisableCopyOnReadNr7r3r8)r;rr<r=rr>r?r@rArBrC#disable_copy_on_read_eager_fallbackrErFrG) r3r8rIrJrKrLrMrNrOs rPdisable_copy_on_readr*s    0h..0$ #\\ 11 !43g n(88hT;!QX *  & &- ##At,,  # #    0 4))  # #   rwzraw_ops.DisableCopyOnReadctj|tj}|g}d}t j dd||||}d}|S)NsDisableCopyOnReadrrV)rAr\r]r3rZrr3r8r9r`rarKs rPrrNsN  # #Hg.>.> ?(, &   11\#)s ?' ' .rcmutexreturnc\tjxstj}|j}|jr t j |d||}|Stjd||\}}}}|dd}tj r&d} |j"} tj$d| | ||\}|S#t j$r }tj||Yd}~nd}~wt j$rYnwxYw t|||S#t j$rYwxYw)aJLocks a mutex resource. The output is the lock. So long as the lock tensor is alive, any other request to use `MutexLock` with this mutex will wait. This is particularly useful for creating a critical section when used in conjunction with `MutexLockIdentity`: ```python mutex = mutex_v2( shared_name=handle_name, container=container, name=name) def execute_in_critical_section(fn, *args, **kwargs): lock = gen_resource_variable_ops.mutex_lock(mutex) with ops.control_dependencies([lock]): r = fn(*args, **kwargs) with ops.control_dependencies(nest.flatten(r)): with ops.colocate_with(mutex): ensure_lock_exists = mutex_lock_identity(lock) # Make sure that if any element of r is accessed, all of # them are executed together. r = nest.map_structure(tf.identity, r) with ops.control_dependencies([ensure_lock_exists]): return nest.map_structure(tf.identity, r) ``` While `fn` is running in the critical section, no other functions which wish to use this critical section may run. Often the use case is that two executions of the same graph, in parallel, wish to run `fn`; and we wish to ensure that only one of them executes at a time. This is especially important if `fn` modifies one or more variables at a time. It is also useful if two separate functions must share a resource, but we wish to ensure the usage is exclusive. Args: mutex: A `Tensor` of type `resource`. The mutex resource to lock. name: A name for the operation (optional). Returns: A `Tensor` of type `variant`. MutexLockNr7)rr8)r;rr<r=rr>r?r@rArBrCmutex_lock_eager_fallbackrErFrGrZmust_record_gradientrWrecord_gradient) rr8rIrJrKrLrMrNrOrar`s rPrrrrXs(b    0h..0$ #\\ 11 k4(g n(885t-!QX QK' ""$ F::L \674 (' .'  & &- ##At,,  # #    & d&&  # #   0B<<DC**DD DD+*D+zraw_ops.MutexLockctj|tj}|g}d}t j dd||||}t j rt jd||||\}|S)Ns MutexLockrVrrAr\r]r3rZrrr)rr8r9r`rarKs rPrrsr  (8(8 9%, &   \1\#)s ?' ""$ \674 (' .rc container shared_namec tjxstj}|j}|jr t j |d|d|d|}|S|d}tj|d}|d}tj|d}tj d|||\}}}} | dd}tj"rHd|j%dd|j%df} |j&} tj(d| | ||\}|S#t j$r }tj||Yd}~nd}~wt j$rYnwxYw t||||S#t j$rY)wxYw)a Creates a Mutex resource that can be locked by `MutexLock`. Args: container: An optional `string`. Defaults to `""`. If non-empty, this variable is placed in the given container. Otherwise, a default container is used. shared_name: An optional `string`. Defaults to `""`. If non-empty, this variable is named in the given bucket with this shared_name. Otherwise, the node name is used instead. name: A name for the operation (optional). Returns: A `Tensor` of type `resource`. MutexV2rrN)rrr8r9r)rrr8)r;rr<r=rr>r?r@rArBrCmutex_v2_eager_fallbackrErZmake_strrFrGrget_attrrWr) rrr8rIrJrKrLrMrNrOrar`s rPmutex_v2rs    0h..0$ #\\ 11 i{I}gnI ;7)K!!+}=+'88YKdL!QX QK' ""$3<< 4mll=)+F::L <2 (' .5  & &- ##At,,  # #    $;TtMM  # #   s0DE)EEE!E00FFzraw_ops.MutexV2c|d}tj|d}|d}tj|d}g}d|d|f}tjdd||||}tjrtjd||||\}|S)NrrrsMutexV2rrVr)rZrrrr)rrr8r9r`rarKs rPrrsI ;7)K!!+}=+, M; ?&   Zr?r@rArBrCread_variable_op_eager_fallbackrErZ make_typerFrGr_get_attr_typerWr) r3rUr8rIrJrKrLrMrNrOrar`s rPread_variable_oprsT&    0h..0$ #\\ 11 h@g n   UG ,%'8885tE!QX QK' ""$s))'2 3F::L ,9 (' .)  & &- ##At,,  # #    , %d66  # #   s0C&&D-9DD-,D-1EEEzraw_ops.ReadVariableOpctj|d}tj|tj }|g}d|f}tj dd||||}tjrtjd||||\}|S)NrUsReadVariableOprrVr) rZrrAr\r]r3rrr)r3rUr8r9r`rarKs rPrr2s   UG ,%  # #Hg.>.> ?(, U &   .,#)s ?' ""$ ,9 (' .rc) TV_ResourceGather_dtyperrrrrrrrrrrrr r!r"r#r$r%r&r'r(r)r*r+r,r-r.r/r0r1r2TV_ResourceGather_Tindicesr"r$indices batch_dimsvalidate_indicesc tjxstj}|j}|jr" t j |d|||d|d|d| }|Stj|d}|d}tj|d}|d}tj |d}t#j$d|||||| \} } } } | dd}tj&rjd| j)dd| j+dd| j-dd | j-d f} | j.}tj0d|| ||\}|S#t j$r } tj| |Yd} ~ nd} ~ wt j$rYnwxYw t|||||||S#t j$rYgwxYw) atGather slices from the variable pointed to by `resource` according to `indices`. `indices` must be an integer tensor of any dimension (usually 0-D or 1-D). Produces an output tensor with shape `indices.shape + params.shape[1:]` where: ```python # Scalar indices output[:, ..., :] = params[indices, :, ... :] # Vector indices output[i, :, ..., :] = params[indices[i], :, ... :] # Higher rank indices output[i, ..., j, :, ... :] = params[indices[i, ..., j], :, ..., :] ``` Args: resource: A `Tensor` of type `resource`. indices: A `Tensor`. Must be one of the following types: `int32`, `int64`. dtype: A `tf.DType`. batch_dims: An optional `int`. Defaults to `0`. validate_indices: An optional `bool`. Defaults to `True`. name: A name for the operation (optional). Returns: A `Tensor` of type `dtype`. ResourceGatherrrrUN)rrrUr8r9rT)r3rrUrrr8Tindices)r;rr<r=rr>r?r@rArBrCresource_gather_eager_fallbackrErZrmake_introrFrGr _get_attr_int_get_attr_boolrrWr)r3rrUrrr8rIrJrKrLrMrNrOrar`s rPresource_gatherrCs8    0h..0$ #\\ 11 h&(8'5Jgn   UG ,%J  \:*''(8:LM'888WE%/+;$H!QX QK' ""$C--l; #"4"45G"Hs))'2J  ,.F::L ,9 (' .A  & &- ##At,,  # #    + G +5tOO  # #   s0 EF(FFF F22G G zraw_ops.ResourceGatherc tj|d}|d}tj|d}|d}tj|d}tj|g|t j t jg\}\}tj|t j}||g}d|d|d|d|f} tjdd|| || } tjrtjd || | | \} | S) NrUrrTrrsResourceGatherrrVr)rZrrror[r]int32int64rAr\r3rrr) r3rrUrrr8r9_attr_Tindicesr`rarKs rPrrs    UG ,%J  \:*''(8:LM'>>y#PWP]P]_f_l_lOop.*7  # #Hg.>.> ?(G$, *&8:J 5*n .&   .,#)s ?' ""$ ,9 (' .rc) TV_ResourceGatherNd_dtyperrrrrrrrrrrrr r!r"r#r$r%r&r'r(r)r*r+r,r-r.r/r0r1r2TV_ResourceGatherNd_Tindicesc tjxstj}|j}|jr t j |d|||d|}|Stj|d}tj d||||\}}} } | dd}tj"rHd| j%dd| j%df} | j&} tj(d| | ||\}|S#t j$r }tj||Yd}~nd}~wt j$rYnwxYw t|||||S#t j$rY wxYw)a TODO: add doc. Args: resource: A `Tensor` of type `resource`. indices: A `Tensor`. Must be one of the following types: `int32`, `int64`. dtype: A `tf.DType`. name: A name for the operation (optional). Returns: A `Tensor` of type `dtype`. ResourceGatherNdrUNr)r3rrUr8r)r;rr<r=rr>r?r@rArBrC!resource_gather_nd_eager_fallbackrErZrrFrGrrrWr) r3rrUr8rIrJrKrLrMrNrOrar`s rPresource_gather_ndrsn    0h..0$ #\\ 11  $'7EKg n   UG ,%'88Xwe!%'!QX QK' ""$s))'2J  ,.F::L L&'; (' .-  & &- ##At,,  # #    . G5t??  # #   s0C99E D''E?EEE+*E+zraw_ops.ResourceGatherNdctj|d}tj|g|tjtj g\}\}t j|tj}||g}d|d|f}tjdd||||}tjrtjd||||\}|S)NrUrsResourceGatherNdrrVr) rZrr[r]rrrAr\r3rrr) r3rrUr8r9rr`rarKs rPrrs   UG ,%'>>y#PWP]P]_f_l_lOop.*7  # #Hg.>.> ?(G$, UJ 7&   0!L#)s ?' ""$ L&'; (' .rcTV_ResourceScatterAdd_dtyperrrrrr r!r%r&r'r(r)r*r-r.r0r1TV_ResourceScatterAdd_Tindicesupdatesctjxstj}|j}|jr t j |d||||}|Stjd||||\}}} } | S#t j$r }tj||Yd}~nd}~wt j$rYnwxYw t|||||S#t j$rYwxYw)aBAdds sparse updates to the variable referenced by `resource`. This operation computes # Scalar indices ref[indices, ...] += updates[...] # Vector indices (for each i) ref[indices[i], ...] += updates[i, ...] # High rank indices (for each i, ..., j) ref[indices[i, ..., j], ...] += updates[i, ..., j, ...] Duplicate entries are handled correctly: if multiple `indices` reference the same location, their contributions add. Requires `updates.shape = indices.shape + ref.shape[1:]` or `updates.shape = []`.
Args: resource: A `Tensor` of type `resource`. Should be from a `Variable` node. indices: A `Tensor`. Must be one of the following types: `int32`, `int64`. A tensor of indices into the first dimension of `ref`. updates: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `uint8`, `int16`, `int8`, `complex64`, `int64`, `qint8`, `quint8`, `qint32`, `bfloat16`, `qint16`, `quint16`, `uint16`, `complex128`, `half`, `uint32`, `uint64`. A tensor of updated values to add to `ref`. name: A name for the operation (optional). Returns: The created Operation. ResourceScatterAddNr7r3rrr8)r;rr<r=rr>r?r@rArBrC#resource_scatter_add_eager_fallbackrErFrG r3rrr8rIrJrKrLrMrNrOs rPresource_scatter_addrD    0h..0$ #\\ 11 "D(GWFg n(88x&-D:!QX *  & &- ##At,,  # #    0 GW4T;;  # #   0A==CB++CCCC.-C.zraw_ops.ResourceScatterAddctj|g|tjtjtj tj tjtjtjtjtjtjtjtjtjtj tj"tj$tj&tj(tj*g\}\}tj|g|tj tjg\}\}t-j.|tj0}|||g}d|d|f}tj2dd||||} d} | S)NrUrsResourceScatterAddrrVrZr[r]float32float64ruint8int16int8 complex64rqint8quint8qint32bfloat16qint16quint16uint16 complex128halfuint32uint64rAr\r3r r3rrr8r9r_rr`rarKs rPrr$;;WIsW__^e^m^movo|o|FLLNUN[N[]d]i]ikrk|k|~E~K~KMTMZMZ\c\j\jlslzlz|C|L|LNUN\N\^e^m^movo}o}FQQSZS_S_ahaoaoqxqqMBC+z'>>y#PWP]P]_f_l_lOop.*7  # #Hg.>.> ?(GW-, [*n =&   2Al#)s ?' ' .rcTV_ResourceScatterDiv_dtypeTV_ResourceScatterDiv_Tindicesctjxstj}|j}|jr t j |d||||}|Stjd||||\}}} } | S#t j$r }tj||Yd}~nd}~wt j$rYnwxYw t|||||S#t j$rYwxYw)aLDivides sparse updates into the variable referenced by `resource`. This operation computes # Scalar indices ref[indices, ...] /= updates[...] # Vector indices (for each i) ref[indices[i], ...] /= updates[i, ...] # High rank indices (for each i, ..., j) ref[indices[i, ..., j], ...] /= updates[i, ..., j, ...] Duplicate entries are handled correctly: if multiple `indices` reference the same location, their contributions multiply. Requires `updates.shape = indices.shape + ref.shape[1:]` or `updates.shape = []`.
Args: resource: A `Tensor` of type `resource`. Should be from a `Variable` node. indices: A `Tensor`. Must be one of the following types: `int32`, `int64`. A tensor of indices into the first dimension of `ref`. updates: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `uint8`, `int16`, `int8`, `complex64`, `int64`, `qint8`, `quint8`, `qint32`, `bfloat16`, `qint16`, `quint16`, `uint16`, `complex128`, `half`, `uint32`, `uint64`. A tensor of updated values to add to `ref`. name: A name for the operation (optional). Returns: The created Operation. ResourceScatterDivNr7r)r;rr<r=rr>r?r@rArBrC#resource_scatter_div_eager_fallbackrErFrGrs rPresource_scatter_divr.rrzraw_ops.ResourceScatterDivctj|g|tjtjtj tj tjtjtjtjtjtjtjtjtjtj tj"tj$tj&tj(tj*g\}\}tj|g|tj tjg\}\}t-j.|tj0}|||g}d|d|f}tj2dd||||} d} | S)NrUrsResourceScatterDivrrVrrs rPrrhrrcTV_ResourceScatterMax_dtypeTV_ResourceScatterMax_Tindicesctjxstj}|j}|jr t j |d||||}|Stjd||||\}}} } | S#t j$r }tj||Yd}~nd}~wt j$rYnwxYw t|||||S#t j$rYwxYw)aReduces sparse updates into the variable referenced by `resource` using the `max` operation. This operation computes # Scalar indices ref[indices, ...] = max(ref[indices, ...], updates[...]) # Vector indices (for each i) ref[indices[i], ...] = max(ref[indices[i], ...], updates[i, ...]) # High rank indices (for each i, ..., j) ref[indices[i, ..., j], ...] = max(ref[indices[i, ..., j], ...], updates[i, ..., j, ...]) Duplicate entries are handled correctly: if multiple `indices` reference the same location, their contributions are combined. Requires `updates.shape = indices.shape + ref.shape[1:]` or `updates.shape = []`.
Args: resource: A `Tensor` of type `resource`. Should be from a `Variable` node. indices: A `Tensor`. Must be one of the following types: `int32`, `int64`. A tensor of indices into the first dimension of `ref`. updates: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `uint8`, `int16`, `int8`, `complex64`, `int64`, `qint8`, `quint8`, `qint32`, `bfloat16`, `qint16`, `quint16`, `uint16`, `complex128`, `half`, `uint32`, `uint64`. A tensor of updated values to add to `ref`. name: A name for the operation (optional). Returns: The created Operation. ResourceScatterMaxNr7r)r;rr<r=rr>r?r@rArBrC#resource_scatter_max_eager_fallbackrErFrGrs rPresource_scatter_maxrwrrzraw_ops.ResourceScatterMaxctj|g|tjtjtj tj tjtjtjtjtjtjtjtjtjtj tj"tj$tj&tj(tj*g\}\}tj|g|tj tjg\}\}t-j.|tj0}|||g}d|d|f}tj2dd||||} d} | S)NrUrsResourceScatterMaxrrVrrs rPrrrrcTV_ResourceScatterMin_dtypeTV_ResourceScatterMin_Tindicesctjxstj}|j}|jr t j |d||||}|Stjd||||\}}} } | S#t j$r }tj||Yd}~nd}~wt j$rYnwxYw t|||||S#t j$rYwxYw)aReduces sparse updates into the variable referenced by `resource` using the `min` operation. This operation computes # Scalar indices ref[indices, ...] = min(ref[indices, ...], updates[...]) # Vector indices (for each i) ref[indices[i], ...] = min(ref[indices[i], ...], updates[i, ...]) # High rank indices (for each i, ..., j) ref[indices[i, ..., j], ...] = min(ref[indices[i, ..., j], ...], updates[i, ..., j, ...]) Duplicate entries are handled correctly: if multiple `indices` reference the same location, their contributions are combined. Requires `updates.shape = indices.shape + ref.shape[1:]` or `updates.shape = []`.
Args: resource: A `Tensor` of type `resource`. Should be from a `Variable` node. indices: A `Tensor`. Must be one of the following types: `int32`, `int64`. A tensor of indices into the first dimension of `ref`. updates: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `uint8`, `int16`, `int8`, `complex64`, `int64`, `qint8`, `quint8`, `qint32`, `bfloat16`, `qint16`, `quint16`, `uint16`, `complex128`, `half`, `uint32`, `uint64`. A tensor of updated values to add to `ref`. name: A name for the operation (optional). Returns: The created Operation. ResourceScatterMinNr7r)r;rr<r=rr>r?r@rArBrC#resource_scatter_min_eager_fallbackrErFrGrs rPresource_scatter_minrrrzraw_ops.ResourceScatterMinctj|g|tjtjtj tj tjtjtjtjtjtjtjtjtjtj tj"tj$tj&tj(tj*g\}\}tj|g|tj tjg\}\}t-j.|tj0}|||g}d|d|f}tj2dd||||} d} | S)NrUrsResourceScatterMinrrVrrs rPrrrrcTV_ResourceScatterMul_dtypeTV_ResourceScatterMul_Tindicesctjxstj}|j}|jr t j |d||||}|Stjd||||\}}} } | S#t j$r }tj||Yd}~nd}~wt j$rYnwxYw t|||||S#t j$rYwxYw)aOMultiplies sparse updates into the variable referenced by `resource`. This operation computes # Scalar indices ref[indices, ...] *= updates[...] # Vector indices (for each i) ref[indices[i], ...] *= updates[i, ...] # High rank indices (for each i, ..., j) ref[indices[i, ..., j], ...] *= updates[i, ..., j, ...] Duplicate entries are handled correctly: if multiple `indices` reference the same location, their contributions multiply. Requires `updates.shape = indices.shape + ref.shape[1:]` or `updates.shape = []`.
Args: resource: A `Tensor` of type `resource`. Should be from a `Variable` node. indices: A `Tensor`. Must be one of the following types: `int32`, `int64`. A tensor of indices into the first dimension of `ref`. updates: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `uint8`, `int16`, `int8`, `complex64`, `int64`, `qint8`, `quint8`, `qint32`, `bfloat16`, `qint16`, `quint16`, `uint16`, `complex128`, `half`, `uint32`, `uint64`. A tensor of updated values to add to `ref`. name: A name for the operation (optional). Returns: The created Operation. ResourceScatterMulNr7r)r;rr<r=rr>r?r@rArBrC#resource_scatter_mul_eager_fallbackrErFrGrs rPresource_scatter_mulr rrzraw_ops.ResourceScatterMulctj|g|tjtjtj tj tjtjtjtjtjtjtjtjtjtj tj"tj$tj&tj(tj*g\}\}tj|g|tj tjg\}\}t-j.|tj0}|||g}d|d|f}tj2dd||||} d} | S)NrUrsResourceScatterMulrrVrrs rPrrCrrcTV_ResourceScatterSub_dtypeTV_ResourceScatterSub_Tindicesctjxstj}|j}|jr t j |d||||}|Stjd||||\}}} } | S#t j$r }tj||Yd}~nd}~wt j$rYnwxYw t|||||S#t j$rYwxYw)aISubtracts sparse updates from the variable referenced by `resource`. This operation computes # Scalar indices ref[indices, ...] -= updates[...] # Vector indices (for each i) ref[indices[i], ...] -= updates[i, ...] # High rank indices (for each i, ..., j) ref[indices[i, ..., j], ...] -= updates[i, ..., j, ...] Duplicate entries are handled correctly: if multiple `indices` reference the same location, their contributions add. Requires `updates.shape = indices.shape + ref.shape[1:]` or `updates.shape = []`.
Args: resource: A `Tensor` of type `resource`. Should be from a `Variable` node. indices: A `Tensor`. Must be one of the following types: `int32`, `int64`. A tensor of indices into the first dimension of `ref`. updates: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `uint8`, `int16`, `int8`, `complex64`, `int64`, `qint8`, `quint8`, `qint32`, `bfloat16`, `qint16`, `quint16`, `uint16`, `complex128`, `half`, `uint32`, `uint64`. A tensor of updated values to add to `ref`. name: A name for the operation (optional). Returns: The created Operation. ResourceScatterSubNr7r)r;rr<r=rr>r?r@rArBrC#resource_scatter_sub_eager_fallbackrErFrGrs rPresource_scatter_subrRrrzraw_ops.ResourceScatterSubctj|g|tjtjtj tj tjtjtjtjtjtjtjtjtjtj tj"tj$tj&tj(tj*g\}\}tj|g|tj tjg\}\}t-j.|tj0}|||g}d|d|f}tj2dd||||} d} | S)NrUrsResourceScatterSubrrVrrs rPrrrrc) TV_ResourceScatterUpdate_dtyperrrrrrrrrrrrr r!r"r#r$r%r&r'r(r)r*r+r,r-r.r/r0r1r2!TV_ResourceScatterUpdate_Tindicesctjxstj}|j}|jr t j |d||||}|Stjd||||\}}} } | S#t j$r }tj||Yd}~nd}~wt j$rYnwxYw t|||||S#t j$rYwxYw)aAssigns sparse updates to the variable referenced by `resource`. This operation computes # Scalar indices ref[indices, ...] = updates[...] # Vector indices (for each i) ref[indices[i], ...] = updates[i, ...] # High rank indices (for each i, ..., j) ref[indices[i, ..., j], ...] = updates[i, ..., j, ...] Args: resource: A `Tensor` of type `resource`. Should be from a `Variable` node. indices: A `Tensor`. Must be one of the following types: `int32`, `int64`. A tensor of indices into the first dimension of `ref`. updates: A `Tensor`. A tensor of updated values to add to `ref`. name: A name for the operation (optional). Returns: The created Operation. ResourceScatterUpdateNr7r)r;rr<r=rr>r?r@rArBrC&resource_scatter_update_eager_fallbackrErFrGrs rPresource_scatter_updaters0    0h..0$ #\\ 11 %tXwIg n(88(G)0t=!QX *  & &- ##At,,  # #    3 GW4T;;  # #   rzraw_ops.ResourceScatterUpdatecNtj|g|g\}\}tj|g|tjtjg\}\}t j |tj}|||g}d|d|f}tjdd||||} d} | S)NrUrsResourceScatterUpdaterrV) rZr[r]rrrAr\r3rrs rPrrs$;;WIsBO+z'>>y#PWP]P]_f_l_lOop.*7  # #Hg.>.> ?(GW-, [*n =&   5q#)s ?' ' .rc) TV_VarHandleOp_dtyperrrrrrrrrrrrr r!r"r#r$r%r&r'r(r)r*r+r,r-r.r/r0r1r2 debug_namectjxstj}|j}|jr& t j |d|d|d|d|d|d|d|} | Stj|d}tj|d}|d }tj |d}|d }tj |d}|d }tj |d}|g}t#|t$t&fst)d |z|D cgc]} tj | d}} t+j,d||||||| \} } } }|dd} tj.rd| j1dd| j1dd| j1dd| j3dd| j1dd| j1df }| j4}tj6d||| | \} | S#t j$r } tj| |Yd} ~ nd} ~ wt j$rYnwxYw t|||||||| S#t j$rYwxYwcc} w) aCreates a handle to a Variable resource. Args: dtype: A `tf.DType`. the type of this variable. Must agree with the dtypes of all ops using this variable. shape: A `tf.TensorShape` or list of `ints`. The (possibly partially specified) shape of this variable. container: An optional `string`. Defaults to `""`. the container this variable is placed in. shared_name: An optional `string`. Defaults to `""`. the name by which this variable is referred to. debug_name: An optional `string`. Defaults to `""`. the user-given name, which still applies in anonymous mode. allowed_devices: An optional list of `strings`. Defaults to `[]`. DEPRECATED. The allowed devices containing the resource variable. Set when the output ResourceHandle represents a per-replica/partitioned resource variable. name: A name for the operation (optional). Returns: A `Tensor` of type `resource`. VarHandleOprrr rUshapeallowed_devicesN)rrr rUr r r8r9rKExpected list for 'allowed_devices' argument to 'var_handle_op' Op, not %r.)rUr rrr r r8)r;rr<r=rr>r?r@rArBrCvar_handle_op_eager_fallbackrErZr make_shaper isinstancelisttuple TypeErrorrFrGrrrrWr)rUr rrr r r8rIrJrKrL_srMrNrOrar`s rP var_handle_oprs,    0h..0$ #\\  11 mT; =\:ww?,gn   UG ,%   eW -%I ;7)K!!+}=+J  \:*O OdE] 3  &(7 8 99IXX"X&&r+<=X/X'88U%9#.:'6TC!QX QK' ""$3<< 4mll=)<ll<('  )7CLL4I.?!@ BF ::L |VW6 (' .[  & &- ##At,,  # #    );:UO  # #   (Ys6$G7 I/7H> H%%H>=H>II,+I,zraw_ops.VarHandleOpc Ptj|d}tj|d}|d}tj|d}|d}tj|d}|d}tj|d}|g}t |t t fstd|z|Dcgc]}tj|d}}g} d|d|d|d|d|d|f } tjd d | | || } tjrtjd | | | | \} | Scc}w) NrUr rrrr rr s VarHandleOprrVr ) rZrrrrrrrrrr) rUr rrr r r8r9rr`rarKs rPrr*sQ   UG ,%   eW -%I ;7)K!!+}=+J  \:*O OdE] 3  &(7 8 99IXX"X&&r+<=X/X, M;  gugu.? &   ^Q|#)s ?' ""$ |VW6 (' .Ys'D#c\tjxstj}|j}|jr t j |d||}|Stjd||\}}}}|dd}tj r&d} |j"} tj$d| | ||\}|S#t j$r }tj||Yd}~nd}~wt j$rYnwxYw t|||S#t j$rYwxYw)zChecks whether a resource handle-based variable has been initialized. Args: resource: A `Tensor` of type `resource`. the input resource handle. name: A name for the operation (optional). Returns: A `Tensor` of type `bool`. VarIsInitializedOpNr7rr)r;rr<r=rr>r?r@rArBrC$var_is_initialized_op_eager_fallbackrErFrGrZrrWr) r3r8rIrJrKrLrMrNrOrar`s rPvar_is_initialized_oprJs(    0h..0$ #\\ 11 "D(4g n(88xd<!QX QK' ""$ F::L lFG= (' .'  & &- ##At,,  # #    1 4))  # #   rzraw_ops.VarIsInitializedOpctj|tj}|g}d}t j dd||||}t j rt jd||||\}|S)NsVarIsInitializedOprrVrrrs rPrrsss  # #Hg.>.> ?(, &   2Al#)s ?' ""$ lFG= (' .rcTV_VariableShape_out_typeinputout_typectjxstj}|j}|jr t j |d||d|}|S|tj}tj |d}t#j$d|||\}}}} | dd}tj&r7d|j)df} |j*} tj,d| | ||\}|S#t j$r }tj||Yd}~nd}~wt j$rYnwxYw t||||S#t j$rY wxYw)aReturns the shape of the variable pointed to by `resource`. This operation returns a 1-D integer tensor representing the shape of `input`. For example: ``` # 't' is [[[1, 1, 1], [2, 2, 2]], [[3, 3, 3], [4, 4, 4]]] shape(t) ==> [2, 2, 3] ``` Args: input: A `Tensor` of type `resource`. out_type: An optional `tf.DType` from: `tf.int32, tf.int64`. Defaults to `tf.int32`. name: A name for the operation (optional). Returns: A `Tensor` of type `out_type`. VariableShaperN)rr8r9)rrr8)r;rr<r=rr>r?r@rArBrCvariable_shape_eager_fallbackrEr]rrZrrFrGrrrWr) rrr8rIrJrKrLrMrNrOrar`s rPvariable_shaper#sb(    0h..0$ #\\ 11 otUJBg n}}H   * 5('88uxdD!QX QK' ""$#,,Z8 9F::L vw8 (' .-  & &- ##At,,  # #    * (499  # #   s0C88D? D&&D?>D?EE)(E)zraw_ops.VariableShapecB|tj}tj|d}t j |tj }|g}d|f}tjdd||||}tjrtjd||||\}|S)Nrs VariableShaperrVr!) r]rrZrrAr\r3rrr)rrr8r9r`rarKs rPr"r"s }}H   * 5(  (8(8 9%,  !&   -q#)s ?' ""$ vw8 (' .rc)N)FN)TN)rrN)rTN)__doc__ collectionstensorflow.pythonrtensorflow.python.eagerrr;rr?rrZtensorflow.python.frameworkrr]tensorflow.security.fuzzing.pyr_atypesr _op_def_registryr rAr rF"tensorflow.python.util.deprecationr tensorflow.python.utilr _dispatch tensorflow.python.util.tf_exportrtypingrrrtyping_extensionsrrResourcerQ to_raw_opr6rDrdrhrfrgrjboolrprmrnVariantrvrtrur~r|r}rrrrrrrstrrrrrrrrrrintrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr rBoolrrrrrr#r!r"rrcrPr:s 67179FK3IC86%%'& \ " YsG4D4D/D%E" iX[]yXyNz" F?i =>~t~~Nd?efIc7CSCS>S4T]fgjmIhI^J ' \ " YsG4D4D/D%E" iX[]yXyNz" F?i =>~t~~Nd?efIc7CSCS>S4T]fgjmIhI^J$V ( 30@0@+@!A( )TWYrTrJs( EI( R9978HZ9[\  #w?O?O:O0P YbcfiBdBZC UY & 9S'//-A#B& N9978HZ9[\)C4>>J];^_ 3@P@P;P1Q hl   9S'2B2B-B#C  B;I9:>4>>J^;_`)CAQAQ4>>*+E F  YsG4D4D/D%E U^_bdkdsds_sUt 1s1313PWP`P`K`Aa1f ')% &~t~~h'? @sIVY[b[k[kVkLl$"R .yg.>.>)>?.H_.oxy|VzVpW.`534^T^^DT5UV iW=M=M8M.N Wn HILNeIef "R $%A?TcdDiW-=-=(=>DSVXrSrIsD}TDadDy}DR[\_ax\xRyDL534^T^^O5TUYsG4>>RiCjk YsGDTDT?T5U `ijmpQkQaR ]fgjlJgJ]K L DFXZkm@BIML-LLTWLgjLR[\_ahaqaq\qRrL\/i-.~t~~m/LM (<PSbesvXabegngwgwbwXx@$Ic73C3C.C$D$T]^acjcoco^oTp$L=Y;<^T^^La=bc 9S'BRBR=R3S clmpryr~r~m~c $$?Rab_f_l_lsw1)C)9)9$9:1E^1}FGJLeGe}f1f3 12>4>>.3QR  38H8H3H)I Un HILNgIgh rc