From 8581ab165a7850441a06f362f1063f7889a82aa8 Mon Sep 17 00:00:00 2001 From: co63oc Date: Thu, 15 Feb 2024 13:37:33 +0800 Subject: [PATCH] Fix --- python/paddle/optimizer/adadelta.py | 6 +++--- python/paddle/optimizer/adagrad.py | 6 +++--- python/paddle/optimizer/adam.py | 6 +++--- python/paddle/optimizer/adamax.py | 6 +++--- python/paddle/optimizer/adamw.py | 8 ++++---- python/paddle/optimizer/asgd.py | 4 ++-- python/paddle/optimizer/lamb.py | 6 +++--- python/paddle/optimizer/momentum.py | 6 +++--- python/paddle/optimizer/optimizer.py | 2 +- python/paddle/optimizer/rmsprop.py | 6 +++--- python/paddle/optimizer/rprop.py | 6 +++--- python/paddle/optimizer/sgd.py | 4 ++-- 12 files changed, 33 insertions(+), 33 deletions(-) diff --git a/python/paddle/optimizer/adadelta.py b/python/paddle/optimizer/adadelta.py index f9c4aa3f4d87db..e334c95f0843de 100644 --- a/python/paddle/optimizer/adadelta.py +++ b/python/paddle/optimizer/adadelta.py @@ -149,7 +149,7 @@ def _create_accumulators(self, block, parameters): parameters = parameters.get('params') for p in parameters: - if p.name in self._already_create_accumulater: + if p.name in self._already_create_accumulator: continue if self._multi_precision and self._is_dtype_fp16_or_bf16(p.dtype): master_p = self._create_master_weight(p) @@ -157,7 +157,7 @@ def _create_accumulators(self, block, parameters): self._add_accumulator( self._avg_squared_update_acc_str, master_p ) - self._already_create_accumulater.add(p.name) + self._already_create_accumulator.add(p.name) continue if ( self._is_dtype_fp16_or_bf16(p.dtype) @@ -169,7 +169,7 @@ def _create_accumulators(self, block, parameters): ) self._add_accumulator(self._avg_squared_grad_acc_str, p) self._add_accumulator(self._avg_squared_update_acc_str, p) - self._already_create_accumulater.add(p.name) + self._already_create_accumulator.add(p.name) def _append_optimize_op(self, block, param_and_grad): if isinstance(param_and_grad, dict): diff --git a/python/paddle/optimizer/adagrad.py b/python/paddle/optimizer/adagrad.py index eae910870ae53a..2ac15cbc692a39 100644 --- a/python/paddle/optimizer/adagrad.py +++ b/python/paddle/optimizer/adagrad.py @@ -142,7 +142,7 @@ def _create_accumulators(self, block, parameters): parameters = self._update_param_group(parameters) for p in parameters: - if p.name in self._already_create_accumulater: + if p.name in self._already_create_accumulator: continue if self._multi_precision and self._is_dtype_fp16_or_bf16(p.dtype): master_p = self._create_master_weight(p) @@ -151,7 +151,7 @@ def _create_accumulators(self, block, parameters): master_p, fill_value=self.initial_accumulator_value, ) - self._already_create_accumulater.add(p.name) + self._already_create_accumulator.add(p.name) continue if ( self._is_dtype_fp16_or_bf16(p.dtype) @@ -166,7 +166,7 @@ def _create_accumulators(self, block, parameters): p, fill_value=self.initial_accumulator_value, ) - self._already_create_accumulater.add(p.name) + self._already_create_accumulator.add(p.name) def _append_optimize_op(self, block, param_and_grad): assert isinstance(block, framework.Block) diff --git a/python/paddle/optimizer/adam.py b/python/paddle/optimizer/adam.py index cce61f0e89b5fa..8bf6a3f6519a6d 100644 --- a/python/paddle/optimizer/adam.py +++ b/python/paddle/optimizer/adam.py @@ -270,12 +270,12 @@ def _create_accumulators(self, block, parameters): # Create accumulator tensors for first and second moments for p in parameters: - if p.name in self._already_create_accumulater: + if p.name in self._already_create_accumulator: continue if self._multi_precision and self._is_dtype_fp16_or_bf16(p.dtype): master_p = self._create_master_weight(p) self._add_moments_pows(master_p) - self._already_create_accumulater.add(p.name) + self._already_create_accumulator.add(p.name) continue if ( self._is_dtype_fp16_or_bf16(p.dtype) @@ -286,7 +286,7 @@ def _create_accumulators(self, block, parameters): "Consider using multi_precision=True option of the Adam optimizer." ) self._add_moments_pows(p) - self._already_create_accumulater.add(p.name) + self._already_create_accumulator.add(p.name) def _append_optimize_op(self, block, param_and_grad): assert isinstance(block, (framework.Block, paddle.pir.Block)) diff --git a/python/paddle/optimizer/adamax.py b/python/paddle/optimizer/adamax.py index 3d65df04bb62be..cc61dbce86573b 100644 --- a/python/paddle/optimizer/adamax.py +++ b/python/paddle/optimizer/adamax.py @@ -197,12 +197,12 @@ def _create_accumulators(self, block, parameters): # Create accumulator tensors for first moment and infinity norm for p in parameters: - if p.name in self._already_create_accumulater: + if p.name in self._already_create_accumulator: continue if self._multi_precision and self._is_dtype_fp16_or_bf16(p.dtype): master_p = self._create_master_weight(p) self._add_moments_pows(master_p) - self._already_create_accumulater.add(p.name) + self._already_create_accumulator.add(p.name) continue if ( self._is_dtype_fp16_or_bf16(p.dtype) @@ -213,7 +213,7 @@ def _create_accumulators(self, block, parameters): "Consider using multi_precision=True option of the Adam optimizer." ) self._add_moments_pows(p) - self._already_create_accumulater.add(p.name) + self._already_create_accumulator.add(p.name) def _append_optimize_op(self, block, param_and_grad): assert isinstance(block, framework.Block) diff --git a/python/paddle/optimizer/adamw.py b/python/paddle/optimizer/adamw.py index 729991285d63ad..e665a43a53b47a 100644 --- a/python/paddle/optimizer/adamw.py +++ b/python/paddle/optimizer/adamw.py @@ -293,7 +293,7 @@ def __init__( self._use_multi_tensor = None self.regularization = None self._auxiliary_vars = {} - self._already_create_accumulater = set() + self._already_create_accumulator = set() self._create_master_grad_states() @@ -400,12 +400,12 @@ def _create_accumulators(self, block, parameters): # Create accumulator tensors for first and second moments for p in parameters: - if p.name in self._already_create_accumulater: + if p.name in self._already_create_accumulator: continue if self._multi_precision and self._is_dtype_fp16_or_bf16(p.dtype): master_p = self._create_master_weight(p) self._add_moments_pows(master_p) - self._already_create_accumulater.add(p.name) + self._already_create_accumulator.add(p.name) continue if ( self._is_dtype_fp16_or_bf16(p.dtype) @@ -416,7 +416,7 @@ def _create_accumulators(self, block, parameters): "Consider using multi_precision=True option of the Adam optimizer." ) self._add_moments_pows(p) - self._already_create_accumulater.add(p.name) + self._already_create_accumulator.add(p.name) def _append_optimize_op(self, block, param_and_grad): assert isinstance(block, (framework.Block, pir.Block)) diff --git a/python/paddle/optimizer/asgd.py b/python/paddle/optimizer/asgd.py index 36e08a6b12057f..45b35fd9e0e521 100644 --- a/python/paddle/optimizer/asgd.py +++ b/python/paddle/optimizer/asgd.py @@ -128,7 +128,7 @@ def _create_accumulators(self, block, parameters): parameters = self._update_param_group(parameters) for p in parameters: - if p.name in self._already_create_accumulater: + if p.name in self._already_create_accumulator: continue p_new = p if self._multi_precision and self._is_dtype_fp16_or_bf16(p.dtype): @@ -167,7 +167,7 @@ def _create_accumulators(self, block, parameters): [1], ) - self._already_create_accumulater.add(p.name) + self._already_create_accumulator.add(p.name) def _assign_accumulator_master( self, block, name, param, assign_value, index diff --git a/python/paddle/optimizer/lamb.py b/python/paddle/optimizer/lamb.py index 105bddf07b839f..540ea508da726f 100644 --- a/python/paddle/optimizer/lamb.py +++ b/python/paddle/optimizer/lamb.py @@ -164,15 +164,15 @@ def _create_accumulators(self, block, parameters): # Create accumulator tensors for first and second moments for p in parameters: - if p.name in self._already_create_accumulater: + if p.name in self._already_create_accumulator: continue if self._multi_precision and self._is_dtype_fp16_or_bf16(p.dtype): master_p = self._create_master_weight(p) self._add_moments_pows(master_p) - self._already_create_accumulater.add(p.name) + self._already_create_accumulator.add(p.name) else: self._add_moments_pows(p) - self._already_create_accumulater.add(p.name) + self._already_create_accumulator.add(p.name) def _add_moments_pows(self, p): acc_dtype = p.dtype diff --git a/python/paddle/optimizer/momentum.py b/python/paddle/optimizer/momentum.py index 6ae9c4dcf5f4ae..d8d94b19ad6472 100644 --- a/python/paddle/optimizer/momentum.py +++ b/python/paddle/optimizer/momentum.py @@ -210,12 +210,12 @@ def _create_accumulators(self, block, parameters): parameters = self._update_param_group(parameters) for p in parameters: - if p.name in self._already_create_accumulater: + if p.name in self._already_create_accumulator: continue if self._multi_precision and self._is_dtype_fp16_or_bf16(p.dtype): master_p = self._create_master_weight(p) self._add_accumulator(self._velocity_acc_str, master_p) - self._already_create_accumulater.add(p.name) + self._already_create_accumulator.add(p.name) continue if ( self._is_dtype_fp16_or_bf16(p.dtype) @@ -226,7 +226,7 @@ def _create_accumulators(self, block, parameters): "Consider using multi_precision=True option of the Momentum optimizer." ) self._add_accumulator(self._velocity_acc_str, p) - self._already_create_accumulater.add(p.name) + self._already_create_accumulator.add(p.name) def _create_regularization_of_grad(self, param, grad, regularization=None): """Create and add backward regularization Operators diff --git a/python/paddle/optimizer/optimizer.py b/python/paddle/optimizer/optimizer.py index 760b2b93c3d681..3eea3c6675f418 100644 --- a/python/paddle/optimizer/optimizer.py +++ b/python/paddle/optimizer/optimizer.py @@ -284,7 +284,7 @@ def __init__( self._param_dict = self._create_multi_tensor_dict() self._auxiliary_vars = {} - self._already_create_accumulater = set() + self._already_create_accumulator = set() self._master_weights = {} # create master gradients' states diff --git a/python/paddle/optimizer/rmsprop.py b/python/paddle/optimizer/rmsprop.py index 21daba3400efa8..e2568e9f7e8216 100644 --- a/python/paddle/optimizer/rmsprop.py +++ b/python/paddle/optimizer/rmsprop.py @@ -204,7 +204,7 @@ def _create_accumulators(self, block, parameters): parameters = parameters.get('params') for p in parameters: - if p.name in self._already_create_accumulater: + if p.name in self._already_create_accumulator: continue if self._multi_precision and self._is_dtype_fp16_or_bf16(p.dtype): @@ -212,7 +212,7 @@ def _create_accumulators(self, block, parameters): self._add_accumulator(self._momentum_acc_str, master_p) self._add_accumulator(self._mean_square_acc_str, master_p) self._add_accumulator(self._mean_grad_acc_str, master_p) - self._already_create_accumulater.add(p.name) + self._already_create_accumulator.add(p.name) continue if ( self._is_dtype_fp16_or_bf16(p.dtype) @@ -225,7 +225,7 @@ def _create_accumulators(self, block, parameters): self._add_accumulator(self._momentum_acc_str, p) self._add_accumulator(self._mean_square_acc_str, p) self._add_accumulator(self._mean_grad_acc_str, p) - self._already_create_accumulater.add(p.name) + self._already_create_accumulator.add(p.name) def _append_optimize_op(self, block, param_and_grad): if not isinstance(block, framework.Block): diff --git a/python/paddle/optimizer/rprop.py b/python/paddle/optimizer/rprop.py index 25b4be7170be20..6a1a76daff7908 100644 --- a/python/paddle/optimizer/rprop.py +++ b/python/paddle/optimizer/rprop.py @@ -149,7 +149,7 @@ def _create_accumulators(self, block, parameters): # Create accumulator tensors for first and second moments for p in parameters: - if p.name in self._already_create_accumulater: + if p.name in self._already_create_accumulator: continue if self._multi_precision and self._is_dtype_fp16_or_bf16(p.dtype): master_p = self._create_master_weight(p) @@ -165,7 +165,7 @@ def _create_accumulators(self, block, parameters): p.dtype, self._initial_learning_rate, ) - self._already_create_accumulater.add(p.name) + self._already_create_accumulator.add(p.name) continue if ( self._is_dtype_fp16_or_bf16(p.dtype) @@ -187,7 +187,7 @@ def _create_accumulators(self, block, parameters): p.dtype, fill_value=self._initial_learning_rate, ) - self._already_create_accumulater.add(p.name) + self._already_create_accumulator.add(p.name) @no_grad def _append_optimize_op(self, block, param_and_grad): diff --git a/python/paddle/optimizer/sgd.py b/python/paddle/optimizer/sgd.py index 233261f7f769b1..faffffa2807d11 100644 --- a/python/paddle/optimizer/sgd.py +++ b/python/paddle/optimizer/sgd.py @@ -99,11 +99,11 @@ def _create_accumulators(self, block, parameters): # Create accumulator tensors for first and second moments for p in parameters: - if p.name in self._already_create_accumulater: + if p.name in self._already_create_accumulator: continue if self._multi_precision and self._is_dtype_fp16_or_bf16(p.dtype): master_p = self._create_master_weight(p) - self._already_create_accumulater.add(p.name) + self._already_create_accumulator.add(p.name) continue if ( self._is_dtype_fp16_or_bf16(p.dtype)