Index

_

__init__() (common.pytorch.gradient_clipper.GradientClipper method) (common.pytorch.half_dtype.HalfDType method) (common.pytorch.layers.AlibiPositionEmbeddingLayer.AlibiPositionEmbeddingLayer method) (common.pytorch.layers.AttentionLayer.MultiheadAttention method) (common.pytorch.layers.BiaslessLayerNorm.BiaslessLayerNorm method) (common.pytorch.layers.EmbeddingLayer.EmbeddingLayer method) (common.pytorch.layers.FeedForwardNetwork.FeedForwardNetwork method) (common.pytorch.layers.FeedForwardNetwork.SingleFeedForwardLayer method) (common.pytorch.layers.GaussianNLLLoss.GaussianNLLLoss method) (common.pytorch.layers.HuberLoss.HuberLoss method) (common.pytorch.layers.MultiMarginLoss.MultiMarginLoss method) (common.pytorch.layers.RelativePositionEmbeddingLayer.RelativePositionEmbeddingLayer method) (common.pytorch.layers.SmoothL1Loss.SmoothL1Loss method) (common.pytorch.metrics.cb_metric.CBMetric method) (common.pytorch.metrics.cb_metric.DeviceOutputs method) (common.pytorch.model_utils.BertPretrainModelLoss.BertPretrainModelLoss method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseCheckpointConverter method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseCheckpointConverter_HF_CS method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseCheckpointConverter_PT_PT method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseConfigConverter method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseConfigConverter_CS_CS method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseConfigConverter_HF_CS method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseDictionaryConverter method) (common.pytorch.model_utils.checkpoint_converters.base_converter.ConversionRule method) (common.pytorch.model_utils.checkpoint_converters.base_converter.EquivalentSubkey method) (common.pytorch.model_utils.checkpoint_converters.bert.ConfigConverter_Bert_CS16_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert.ConfigConverter_Bert_CS16_CS18 method) (common.pytorch.model_utils.checkpoint_converters.bert.ConfigConverter_Bert_CS17_CS18 method) (common.pytorch.model_utils.checkpoint_converters.bert.ConfigConverter_Bert_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert.ConfigConverter_Bert_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_Bert_CS17_CS18 method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertLayerNorm_HF_CS method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertModel_CS16_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertModel_CS16_CS18 method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertModel_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertModel_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertPretrainModel_CS16_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertPretrainModel_CS16_CS18 method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertPretrainModel_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertPretrainModel_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.ConfigConverter_BertForQuestionAnswering_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.ConfigConverter_BertForQuestionAnswering_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.ConfigConverter_BertForSequenceClassification_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.ConfigConverter_BertForSequenceClassification_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.ConfigConverter_BertForTokenClassification_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.ConfigConverter_BertForTokenClassification_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertFinetuneModel_CS16_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertFinetuneModel_CS16_CS18 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertForQuestionAnswering_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertForQuestionAnswering_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertForSequenceClassification_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertForSequenceClassification_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertForTokenClassification_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertForTokenClassification_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.ConfigConverter_GPT2Model_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.ConfigConverter_GPT2Model_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2_Attention_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2LMHeadModel_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2LMHeadModel_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2Model_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2Model_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.ConfigConverter_GPT_Neox_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.ConfigConverter_GPT_Neox_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_Attention_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_Headless_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_Headless_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_LMHeadModel_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_LMHeadModel_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.ConfigConverter_GPTJModel_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.ConfigConverter_GPTJModel_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.Converter_GPTJ_Attention_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.Converter_GPTJ_Headless_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.Converter_GPTJ_Headless_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.Converter_GPTJ_LMHeadModel_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.Converter_GPTJ_LMHeadModel_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.ConfigConverter_Codegen_Model_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.ConfigConverter_Codegen_Model_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_Attention_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_Headless_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_Headless_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_LMHeadModel_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_LMHeadModel_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.t5.ConfigConverter_T5_CS16_CS17 method) (common.pytorch.model_utils.checkpoint_converters.t5.ConfigConverter_T5_CS16_CS18 method) (common.pytorch.model_utils.checkpoint_converters.t5.ConfigConverter_T5_CS17_CS18 method) (common.pytorch.model_utils.checkpoint_converters.t5.ConfigConverter_T5_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.t5.ConfigConverter_T5_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_CS16_CS17 method) (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_CS16_CS18 method) (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_CS17_CS18 method) (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_HF_CS18 method) (common.pytorch.model_utils.convert_checkpoint.CheckpointConverterCLI method) (common.pytorch.model_utils.GPTLMHeadModelLoss.GPTLMHeadModelLoss method) (common.pytorch.model_utils.RotaryPositionEmbeddingHelper.RotaryPositionEmbeddingHelper method) (common.pytorch.model_utils.T5ForConditionalGenerationLoss.T5ForConditionalGenerationLoss method) (common.pytorch.optim.AdamBase.Adam method) (common.pytorch.optim.AdamBase.AdamBase method) (common.pytorch.optim.AdamBase.AdamW method) (common.pytorch.optim.CSOptimizer.CSOptimizer method) (common.pytorch.optim.lr_scheduler.ChainedScheduler method) (common.pytorch.optim.lr_scheduler.ConstantLR method) (common.pytorch.optim.lr_scheduler.CosineAnnealingLR method) (common.pytorch.optim.lr_scheduler.CosineAnnealingWarmRestarts method) (common.pytorch.optim.lr_scheduler.CosineDecayLR method) (common.pytorch.optim.lr_scheduler.CyclicLR method) (common.pytorch.optim.lr_scheduler.ExponentialLR method) (common.pytorch.optim.lr_scheduler.InverseExponentialTimeDecayLR method) (common.pytorch.optim.lr_scheduler.InverseSquareRootDecayLR method) (common.pytorch.optim.lr_scheduler.LambdaLR method) (common.pytorch.optim.lr_scheduler.LRScheduler method) (common.pytorch.optim.lr_scheduler.MultiplicativeLR method) (common.pytorch.optim.lr_scheduler.MultiStepLR method) (common.pytorch.optim.lr_scheduler.OneCycleLR method) (common.pytorch.optim.lr_scheduler.PiecewiseConstantLR method) (common.pytorch.optim.lr_scheduler.PolynomialLR method) (common.pytorch.optim.lr_scheduler.SequentialLR method) (common.pytorch.optim.lr_scheduler.StepLR method) (common.pytorch.optim.sparse.base.BaseSparsityOptimizer method) (common.pytorch.optim.sparse.dynamic.DynamicSparsityOptimizer method) (common.pytorch.optim.sparse.dynamic.GMPSparsityOptimizer method) (common.pytorch.optim.sparse.static.StaticSparsityOptimizer method) (common.pytorch.perf_utils.PerfData method) (common.pytorch.pytorch_base_cs_runner.PyTorchBaseCSRunner method) (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) (common.pytorch.pytorch_cs_compiler.PyTorchCSCompiler method) (common.pytorch.pytorch_dist_runner.PyTorchDistRunner method) (common.pytorch.pytorch_runner.PyTorchRunner method) (common.pytorch.PyTorchBaseModel.PyTorchBaseModel method) (common.pytorch.summaries.cb_summary.CBSummary method) (common.pytorch.summaries.cb_summary.DeviceOutputs method) (common.pytorch.summaries.tensor_summary.TensorDescriptor method) (common.pytorch.summaries.tensor_summary.TensorSummary method) (common.pytorch.summaries.tensor_summary.TensorSummaryReader method) (common.pytorch.summary_collection.SummaryCollection method) (common.pytorch.utils.BufferedShuffleDataset method) (common.pytorch.utils.RunConfigParamsValidator method) (common.pytorch.utils.SampleGenerator method) (common.tf.hooks.grad_accum_hooks.GradAccumLoggingTensorHook method) (common.tf.hooks.grad_accum_hooks.GradAccumStepCounterHook method) (common.tf.hooks.grad_accum_hooks.GradAccumSummarySaverHook method) (common.tf.input.TfRecordsProcessor.TfRecordsProcessor method) (common.tf.layers.ActivationLayer.ActivationLayer method) (common.tf.layers.AddLayer.AddLayer method) (common.tf.layers.AttentionLayer.AttentionLayer method) (common.tf.layers.BaseLayer.BaseLayer method) (common.tf.layers.Conv2DLayer.Conv2DLayer method) (common.tf.layers.Conv2DTransposeLayer.Conv2DTransposeLayer method) (common.tf.layers.CrossEntropyFromLogitsLayer.CrossEntropyFromLogitsLayer method) (common.tf.layers.DenseLayer.DenseLayer method) (common.tf.layers.DropoutLayer.DropoutLayer method) (common.tf.layers.EmbeddingLayer.EmbeddingLayer method) (common.tf.layers.FeedForwardNetwork.FeedForwardNetwork method) (common.tf.layers.FeedForwardNetworkV2.FeedForwardNetworkV2 method) (common.tf.layers.LayerNormalizationLayer.LayerNormalizationLayer method) (common.tf.layers.MaxPool2DLayer.MaxPool2DLayer method) (common.tf.layers.PoolerLayer.PoolerLayer method) (common.tf.layers.PoolerLayerV2.PoolerLayerV2 method) (common.tf.layers.PositionEmbeddingLayer.PositionEmbeddingLayer method) (common.tf.layers.PrePostProcessWrapper.PrePostProcessWrapper method) (common.tf.layers.ReshapeLayer.ReshapeLayer method) (common.tf.layers.SegmentEmbeddingLayer.SegmentEmbeddingLayer method) (common.tf.layers.SharedWeightsDenseLayer.SharedWeightsDenseLayer method) (common.tf.layers.SoftmaxLayer.SoftmaxLayer method) (common.tf.layers.SquaredErrorLayer.SquaredErrorLayer method) (common.tf.optimizers.AdamWOptimizer.AdamWOptimizer method) (common.tf.optimizers.GradAccumOptimizer.GradAccumOptimizer method) (common.tf.optimizers.LossScale.CSDynamicLossScale method) (common.tf.optimizers.LossScale.MixedPrecisionLossScaleOptimizerAdapter method) (common.tf.optimizers.Trainer.Trainer method) (common.tf.run_utils.GetWeights method) (common.tf.TFBaseModel.TFBaseModel method) (optim.Adadelta method) (optim.Adafactor method) (optim.Adagrad method) (optim.Adam method) (optim.Adamax method) (optim.AdamW method) (optim.ASGD method) (optim.Lamb method) (optim.lr_scheduler.ChainedScheduler method) (optim.lr_scheduler.ConstantLR method) (optim.lr_scheduler.CosineAnnealingLR method) (optim.lr_scheduler.CosineAnnealingWarmRestarts method) (optim.lr_scheduler.CosineDecayLR method) (optim.lr_scheduler.CyclicLR method) (optim.lr_scheduler.ExponentialLR method) (optim.lr_scheduler.InverseExponentialTimeDecayLR method) (optim.lr_scheduler.InverseSquareRootDecayLR method) (optim.lr_scheduler.LambdaLR method) (optim.lr_scheduler.LinearLR method) (optim.lr_scheduler.MultiplicativeLR method) (optim.lr_scheduler.MultiStepLR method) (optim.lr_scheduler.OneCycleLR method) (optim.lr_scheduler.PiecewiseConstantLR method) (optim.lr_scheduler.PolynomialLR method) (optim.lr_scheduler.SequentialLR method) (optim.lr_scheduler.StepLR method) (optim.NAdam method) (optim.Optimizer method) (optim.RAdam method) (optim.RMSprop method) (optim.Rprop method) (optim.SGD method) (vision.pytorch.input.classification.dataset_factory.Processor method) (vision.pytorch.input.classification.dataset_factory.VisionSubset method) (vision.pytorch.input.classification.mixup.RandomCutmix method) (vision.pytorch.input.classification.mixup.RandomMixup method) (vision.pytorch.input.classification.sampler.RepeatedAugSampler method) (vision.pytorch.input.utils.FastDataLoader method) (vision.pytorch.input.utils.ShardedSampler method) (vision.pytorch.layers.BatchChannelNorm.BatchChannelNorm2D method) (vision.pytorch.layers.ConvNormActBlock.ConvNormActBlock method) (vision.pytorch.layers.ConvNormActBlock.ConvNormActBlockModule method) (vision.pytorch.layers.ConvNormActBlock.ConvNormActLayers method) (vision.pytorch.layers.GroupInstanceNorm.GroupInstanceNorm method) (vision.pytorch.layers.utils.ModuleWrapperClass method) (vision.pytorch.losses.dice_loss.Dice method) (vision.pytorch.losses.dice_loss.DiceCELoss method) (vision.pytorch.losses.dice_loss.DiceScore method)

__new__() (common.pytorch.PyTorchBaseModel.Final static method)

A

AbstractRecomputeWrapper (class in common.tf.layers.AbstractRecomputeWrapper) ActivationLayer (class in common.tf.layers.ActivationLayer) Adadelta (class in optim) Adafactor (class in optim) Adagrad (class in optim) Adam (class in common.pytorch.optim.AdamBase) (class in optim) Adamax (class in optim) AdamBase (class in common.pytorch.optim.AdamBase) (in module optim) AdamW (class in common.pytorch.optim.AdamBase) (class in optim) AdamWOptimizer (class in common.tf.optimizers.AdamWOptimizer) add_param_group() (common.pytorch.optim.sparse.base.BaseSparsityOptimizer method) (common.pytorch.optim.sparse.dynamic.DynamicSparsityOptimizer method) AddLayer (class in common.tf.layers.AddLayer) adjust_channels() (in module vision.pytorch.layers.utils) adjust_depth() (in module vision.pytorch.layers.utils) after_run() (common.tf.hooks.grad_accum_hooks.GradAccumLoggingTensorHook method) (common.tf.hooks.grad_accum_hooks.GradAccumStepCounterHook method) (common.tf.hooks.grad_accum_hooks.GradAccumSummarySaverHook method) aggregate_across_replicas() (in module common.tf.metrics.utils)

AlibiPositionEmbeddingLayer (class in common.pytorch.layers.AlibiPositionEmbeddingLayer) apply_attention_bias() (common.pytorch.layers.AttentionLayer.MultiheadAttention method) apply_grad_sparsity() (common.pytorch.optim.sparse.base.BaseSparsityOptimizer method) apply_gradients() (common.tf.optimizers.GradAccumOptimizer.GradAccumOptimizer method) apply_loss_reduction() (in module common.pytorch.layers.utils) apply_param_sparsity() (common.pytorch.optim.sparse.base.BaseSparsityOptimizer method) apply_position_bias() (common.pytorch.layers.AttentionLayer.MultiheadAttention method) (in module common.pytorch.layers.utils) apply_rotary_position_embedding() (common.pytorch.layers.AttentionLayer.MultiheadAttention method) apply_sparsity() (common.pytorch.optim.sparse.base.BaseSparsityOptimizer method) arg_filter() (in module common.pytorch.run_utils) args (common.pytorch.metrics.cb_metric.DeviceOutputs attribute) (common.pytorch.summaries.cb_summary.DeviceOutputs attribute) as_appliance_key() (common.tf.appliance_utils.ExecutionStrategy class method) ASGD (class in optim) assert_already_converted() (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2_Attention_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_Attention_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_Attention_HF_CS17 method) assert_decoder_nonlinearity() (common.pytorch.model_utils.checkpoint_converters.t5.ConfigConverter_T5_HF_CS17 method) assert_factory_fn() (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseConfigConverter static method) assert_mlm_nonlinearity() (common.pytorch.model_utils.checkpoint_converters.bert.ConfigConverter_Bert_HF_CS17 method) AttentionLayer (class in common.tf.layers.AttentionLayer) autogen_loss() (in module common.pytorch.layers.utils)

B

backward() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) BaseCheckpointConverter (class in common.pytorch.model_utils.checkpoint_converters.base_converter) BaseCheckpointConverter_HF_CS (class in common.pytorch.model_utils.checkpoint_converters.base_converter) BaseCheckpointConverter_PT_PT (class in common.pytorch.model_utils.checkpoint_converters.base_converter) BaseConfigConverter (class in common.pytorch.model_utils.checkpoint_converters.base_converter) BaseConfigConverter_CS_CS (class in common.pytorch.model_utils.checkpoint_converters.base_converter) BaseConfigConverter_HF_CS (class in common.pytorch.model_utils.checkpoint_converters.base_converter) BaseDictionaryConverter (class in common.pytorch.model_utils.checkpoint_converters.base_converter) BaseLayer (class in common.tf.layers.BaseLayer) BaseSparsityOptimizer (class in common.pytorch.optim.sparse.base) batch_fn() (common.tf.input.TfRecordsProcessor.TfRecordsProcessor method) BatchChannelNorm2D (class in vision.pytorch.layers.BatchChannelNorm) BCELoss (class in common.pytorch.layers.BCELoss) BCEWithLogitsLoss (class in common.pytorch.layers.BCEWithLogitsLoss) before_run() (common.tf.hooks.grad_accum_hooks.GradAccumLoggingTensorHook method) (common.tf.hooks.grad_accum_hooks.GradAccumSummarySaverHook method) begin() (common.tf.hooks.grad_accum_hooks.GradAccumLoggingTensorHook method) (common.tf.hooks.grad_accum_hooks.GradAccumStepCounterHook method) (common.tf.hooks.grad_accum_hooks.GradAccumSummarySaverHook method)

BertPretrainModelLoss (class in common.pytorch.model_utils.BertPretrainModelLoss) BiaslessLayerNorm (class in common.pytorch.layers.BiaslessLayerNorm) bits_per_x_metric() (in module common.tf.metrics.bits_per_x) bucket_data() (in module common.input.analyze_bucketing) bucketed_batch() (in module common.pytorch.input_utils) (in module common.tf.input.utils) bucketed_cost() (in module common.input.analyze_bucketing) BufferedShuffleDataset (class in common.pytorch.utils) build() (common.tf.layers.AttentionLayer.AttentionLayer method) (common.tf.layers.EmbeddingLayer.EmbeddingLayer method) (common.tf.layers.PositionEmbeddingLayer.PositionEmbeddingLayer method) (common.tf.layers.SegmentEmbeddingLayer.SegmentEmbeddingLayer method) build_eval_metric_ops() (common.tf.TFBaseModel.TFBaseModel method) build_model() (common.tf.TFBaseModel.TFBaseModel method) build_optimizer() (common.tf.optimizers.Trainer.Trainer method) build_total_loss() (common.tf.TFBaseModel.TFBaseModel method) build_train_ops() (common.tf.optimizers.Trainer.Trainer method) (common.tf.TFBaseModel.TFBaseModel method)

C

c_attn_converter() (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2_Attention_HF_CS17 method) c_attn_converter_cs17_to_hf() (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2_Attention_HF_CS17 method) c_attn_converter_hf_to_cs17() (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2_Attention_HF_CS17 method) calculate_accuracy() (in module common.tf.metrics.accuracy) calculate_attention_logits() (common.pytorch.layers.AttentionLayer.MultiheadAttention method) calculate_attention_output() (common.pytorch.layers.AttentionLayer.MultiheadAttention method) calculate_attention_scores() (common.pytorch.layers.AttentionLayer.MultiheadAttention method) calculate_bits_per_x() (in module common.tf.metrics.bits_per_x) calculate_perplexity() (in module common.tf.metrics.perplexity) call() (common.tf.layers.AbstractRecomputeWrapper.AbstractRecomputeWrapper method) (common.tf.layers.ActivationLayer.ActivationLayer method) (common.tf.layers.AddLayer.AddLayer method) (common.tf.layers.AttentionLayer.AttentionLayer method) (common.tf.layers.AttentionLayer.SelfAttentionLayer method) (common.tf.layers.BaseLayer.BaseLayer method) (common.tf.layers.Conv2DLayer.Conv2DLayer method) (common.tf.layers.Conv2DTransposeLayer.Conv2DTransposeLayer method) (common.tf.layers.CrossEntropyFromLogitsLayer.CrossEntropyFromLogitsLayer method) (common.tf.layers.DenseLayer.DenseLayer method) (common.tf.layers.DropoutLayer.DropoutLayer method) (common.tf.layers.EmbeddingLayer.EmbeddingLayer method) (common.tf.layers.FeedForwardNetwork.FeedForwardNetwork method) (common.tf.layers.FeedForwardNetworkV2.FeedForwardNetworkV2 method) (common.tf.layers.LayerNormalizationLayer.LayerNormalizationLayer method) (common.tf.layers.MaxPool2DLayer.MaxPool2DLayer method) (common.tf.layers.PoolerLayer.PoolerLayer method) (common.tf.layers.PoolerLayerV2.PoolerLayerV2 method) (common.tf.layers.PositionEmbeddingLayer.PositionEmbeddingLayer method) (common.tf.layers.ReshapeLayer.ReshapeLayer method) (common.tf.layers.SegmentEmbeddingLayer.SegmentEmbeddingLayer method) (common.tf.layers.SharedWeightsDenseLayer.SharedWeightsDenseLayer method) (common.tf.layers.SoftmaxLayer.SoftmaxLayer method) (common.tf.layers.SquaredErrorLayer.SquaredErrorLayer method) CBMetric (class in common.pytorch.metrics.cb_metric) CBSummary (class in common.pytorch.summaries.cb_summary) CBSummaryMeta (class in common.pytorch.summaries.cb_summary) ChainedScheduler (class in common.pytorch.optim.lr_scheduler) (class in optim.lr_scheduler) check_amp() (common.pytorch.gradient_clipper.GradientClipper method) check_and_create_output_dirs() (in module common.input.utils) check_env() (in module common.tf.run_utils) check_extra_params() (common.pytorch.layers.AttentionLayer.MultiheadAttention method) check_split_valid() (vision.pytorch.input.classification.dataset_factory.Processor method) CheckpointConverterCLI (class in common.pytorch.model_utils.convert_checkpoint) clip() (common.pytorch.gradient_clipper.GradientClipper method) clip_gradients() (common.tf.optimizers.Trainer.Trainer method) collect_perf_data() (in module common.pytorch.perf_utils) combine_masks() (common.pytorch.layers.AttentionLayer.MultiheadAttention method) common.input module common.input.analyze_bucketing module common.input.utils module common.model_utils module common.model_utils.count_lines module common.pytorch module common.pytorch.gradient_clipper module common.pytorch.half_dtype module common.pytorch.input_utils module common.pytorch.layers module common.pytorch.layers.AlibiPositionEmbeddingLayer module common.pytorch.layers.AttentionHelper module common.pytorch.layers.AttentionLayer module common.pytorch.layers.BCELoss module common.pytorch.layers.BCEWithLogitsLoss module common.pytorch.layers.BiaslessLayerNorm module common.pytorch.layers.CosineEmbeddingLoss module common.pytorch.layers.CrossEntropyLoss module common.pytorch.layers.EmbeddingLayer module common.pytorch.layers.FeedForwardNetwork module common.pytorch.layers.GaussianNLLLoss module common.pytorch.layers.GPTJDecoderLayer module common.pytorch.layers.HingeEmbeddingLoss module common.pytorch.layers.HuberLoss module common.pytorch.layers.KLDivLoss module common.pytorch.layers.L1Loss module common.pytorch.layers.MarginRankingLoss module common.pytorch.layers.MSELoss module common.pytorch.layers.MultiLabelSoftMarginLoss module common.pytorch.layers.MultiMarginLoss module common.pytorch.layers.NLLLoss module common.pytorch.layers.PoissonNLLLoss module common.pytorch.layers.RelativePositionEmbeddingLayer module common.pytorch.layers.SmoothL1Loss module common.pytorch.layers.Transformer module common.pytorch.layers.TransformerDecoder module common.pytorch.layers.TransformerDecoderLayer module common.pytorch.layers.TransformerEncoder module common.pytorch.layers.TransformerEncoderLayer module common.pytorch.layers.TripletMarginLoss module common.pytorch.layers.TripletMarginWithDistanceLoss module common.pytorch.layers.utils module common.pytorch.metrics module common.pytorch.metrics.accuracy module common.pytorch.metrics.auc module common.pytorch.metrics.cb_metric module common.pytorch.metrics.dice_coefficient module common.pytorch.metrics.fbeta_score module common.pytorch.metrics.mean_iou module common.pytorch.metrics.metric_utils module common.pytorch.metrics.perplexity module common.pytorch.metrics.precision_at_k module common.pytorch.metrics.recall_at_k module common.pytorch.metrics.rouge_score module common.pytorch.model_utils module common.pytorch.model_utils.BertPretrainModelLoss module common.pytorch.model_utils.checkpoint_converters module common.pytorch.model_utils.checkpoint_converters.base_converter module common.pytorch.model_utils.checkpoint_converters.bert module common.pytorch.model_utils.checkpoint_converters.bert_finetune module common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs module common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs module common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs module common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs module common.pytorch.model_utils.checkpoint_converters.t5 module common.pytorch.model_utils.convert_checkpoint module common.pytorch.model_utils.create_initializer module common.pytorch.model_utils.GPTLMHeadModelLoss module common.pytorch.model_utils.RotaryPositionEmbeddingHelper module common.pytorch.model_utils.T5ForConditionalGenerationLoss module common.pytorch.model_utils.weight_initializers module common.pytorch.optim module common.pytorch.optim.Adadelta module common.pytorch.optim.Adafactor module common.pytorch.optim.Adagrad module common.pytorch.optim.Adamax module common.pytorch.optim.AdamBase module common.pytorch.optim.ASGD module common.pytorch.optim.CSOptimizer module common.pytorch.optim.Lamb module common.pytorch.optim.lr_scheduler module common.pytorch.optim.NAdam module common.pytorch.optim.RAdam module common.pytorch.optim.RMSprop module common.pytorch.optim.Rprop module common.pytorch.optim.SGD module common.pytorch.optim.sparse module common.pytorch.optim.sparse.base module common.pytorch.optim.sparse.dynamic module common.pytorch.optim.sparse.static module common.pytorch.optim.sparse.utils module common.pytorch.optim.utils module common.pytorch.perf_utils module common.pytorch.pytorch_base_cs_runner module common.pytorch.pytorch_base_runner module common.pytorch.pytorch_cs_appliance module common.pytorch.pytorch_cs_compiler module common.pytorch.pytorch_cs_runner module common.pytorch.pytorch_dist_runner module common.pytorch.pytorch_runner module common.pytorch.PyTorchBaseModel module common.pytorch.run_cstorch_flow module common.pytorch.run_utils module common.pytorch.summaries module common.pytorch.summaries.cb_summary module common.pytorch.summaries.scalar_summary module common.pytorch.summaries.tensor_summary module common.pytorch.summary_collection module common.pytorch.utils module common.tf module common.tf.appliance_utils module

common.tf.estimator module common.tf.hooks module common.tf.hooks.grad_accum_hooks module common.tf.input module common.tf.input.TfRecordsProcessor module common.tf.input.utils module common.tf.layers module common.tf.layers.AbstractRecomputeWrapper module common.tf.layers.ActivationLayer module common.tf.layers.AddLayer module common.tf.layers.AttentionLayer module common.tf.layers.BaseLayer module common.tf.layers.Conv2DLayer module common.tf.layers.Conv2DTransposeLayer module common.tf.layers.CrossEntropyFromLogitsLayer module common.tf.layers.DenseLayer module common.tf.layers.DropoutLayer module common.tf.layers.EmbeddingLayer module common.tf.layers.FeedForwardNetwork module common.tf.layers.FeedForwardNetworkV2 module common.tf.layers.Input module common.tf.layers.LayerNormalizationLayer module common.tf.layers.MaxPool2DLayer module common.tf.layers.PoolerLayer module common.tf.layers.PoolerLayerV2 module common.tf.layers.PositionEmbeddingLayer module common.tf.layers.PrePostProcessWrapper module common.tf.layers.ReshapeLayer module common.tf.layers.SegmentEmbeddingLayer module common.tf.layers.SharedWeightsDenseLayer module common.tf.layers.SoftmaxLayer module common.tf.layers.SquaredErrorLayer module common.tf.metrics module common.tf.metrics.accuracy module common.tf.metrics.bits_per_x module common.tf.metrics.dice_coefficient module common.tf.metrics.ece_loss_metric module common.tf.metrics.f1_score module common.tf.metrics.fbeta_score module common.tf.metrics.mcc module common.tf.metrics.perplexity module common.tf.metrics.rouge_score module common.tf.metrics.utils module common.tf.model_utils module common.tf.model_utils.create_initializer module common.tf.model_utils.gather module common.tf.model_utils.reshape_gather module common.tf.model_utils.scale_loss_value module common.tf.model_utils.shard_dataset module common.tf.model_utils.vocab_utils module common.tf.optimizers module common.tf.optimizers.AdamWOptimizer module common.tf.optimizers.GradAccumOptimizer module common.tf.optimizers.LossScale module common.tf.optimizers.Trainer module common.tf.run_utils module common.tf.TFBaseModel module compile_time (common.pytorch.perf_utils.PerfData attribute) compute() (common.pytorch.metrics.cb_metric.CBMetric method) compute_all_metrics() (in module common.pytorch.metrics.cb_metric) compute_confusion_matrix() (in module common.pytorch.metrics.metric_utils) compute_eval_metrics() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) (common.pytorch.pytorch_cs_compiler.PyTorchCSCompiler method) (common.pytorch.pytorch_dist_runner.PyTorchDistRunner method) compute_grad_norm() (in module common.pytorch.run_cstorch_flow) compute_gradients() (common.tf.optimizers.GradAccumOptimizer.GradAccumOptimizer method) compute_helper() (in module common.pytorch.metrics.dice_coefficient) (in module common.pytorch.metrics.mean_iou) compute_params_norm() (in module common.pytorch.run_cstorch_flow) compute_positional_embeddings() (common.pytorch.layers.EmbeddingLayer.EmbeddingLayer method) compute_raw_relative_positions() (common.pytorch.layers.RelativePositionEmbeddingLayer.RelativePositionEmbeddingLayer static method) compute_relative_positions() (common.pytorch.layers.RelativePositionEmbeddingLayer.RelativePositionEmbeddingLayer method) compute_segment_embeddings() (common.pytorch.layers.EmbeddingLayer.EmbeddingLayer method) compute_token_embeddings() (common.pytorch.layers.EmbeddingLayer.EmbeddingLayer method) ConfigConversionError ConfigConverter_Bert_CS16_CS17 (class in common.pytorch.model_utils.checkpoint_converters.bert) ConfigConverter_Bert_CS16_CS18 (class in common.pytorch.model_utils.checkpoint_converters.bert) ConfigConverter_Bert_CS17_CS18 (class in common.pytorch.model_utils.checkpoint_converters.bert) ConfigConverter_Bert_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.bert) ConfigConverter_Bert_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.bert) ConfigConverter_BertForQuestionAnswering_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.bert_finetune) ConfigConverter_BertForQuestionAnswering_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.bert_finetune) ConfigConverter_BertForSequenceClassification_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.bert_finetune) ConfigConverter_BertForSequenceClassification_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.bert_finetune) ConfigConverter_BertForTokenClassification_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.bert_finetune) ConfigConverter_BertForTokenClassification_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.bert_finetune) ConfigConverter_Codegen_Model_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs) ConfigConverter_Codegen_Model_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs) ConfigConverter_GPT2Model_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs) ConfigConverter_GPT2Model_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs) ConfigConverter_GPT_Neox_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs) ConfigConverter_GPT_Neox_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs) ConfigConverter_GPTJModel_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs) ConfigConverter_GPTJModel_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs) ConfigConverter_T5_CS16_CS17 (class in common.pytorch.model_utils.checkpoint_converters.t5) ConfigConverter_T5_CS16_CS18 (class in common.pytorch.model_utils.checkpoint_converters.t5) ConfigConverter_T5_CS17_CS18 (class in common.pytorch.model_utils.checkpoint_converters.t5) ConfigConverter_T5_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.t5) ConfigConverter_T5_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.t5) configure_lr_scheduler() (in module optim) configure_optimizer() (in module optim) ConstantLR (class in common.pytorch.optim.lr_scheduler) (class in optim.lr_scheduler) construct_key_vector() (common.pytorch.layers.AttentionLayer.MultiheadAttention method) construct_present_kv() (common.pytorch.layers.AttentionLayer.MultiheadAttention method) construct_query_vector() (common.pytorch.layers.AttentionLayer.MultiheadAttention method) construct_value_vector() (common.pytorch.layers.AttentionLayer.MultiheadAttention method) Conv2DLayer (class in common.tf.layers.Conv2DLayer) Conv2DTransposeLayer (class in common.tf.layers.Conv2DTransposeLayer) ConversionRule (class in common.pytorch.model_utils.checkpoint_converters.base_converter) convert() (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseCheckpointConverter class method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseConfigConverter class method) convert_all_keys() (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseDictionaryConverter method) convert_attention_type() (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.ConfigConverter_GPT2Model_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.ConfigConverter_GPTJModel_HF_CS17 method) convert_checkpoint() (in module common.pytorch.model_utils.convert_checkpoint) convert_checkpoint_from_file() (in module common.pytorch.model_utils.convert_checkpoint) convert_cls_predictions_bias() (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertPretrainModel_CS16_CS17 method) convert_config() (in module common.pytorch.model_utils.convert_checkpoint) convert_config_from_file() (in module common.pytorch.model_utils.convert_checkpoint) convert_dense_layer() (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_CS16_CS17 method) convert_embeddings() (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_HF_CS17 method) convert_helper() (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseCheckpointConverter method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseConfigConverter method) convert_key() (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseDictionaryConverter method) (common.pytorch.model_utils.checkpoint_converters.base_converter.ConversionRule method) convert_nonlinearity() (common.pytorch.model_utils.checkpoint_converters.t5.ConfigConverter_T5_HF_CS17 method) convert_position_embedding_type() (common.pytorch.model_utils.checkpoint_converters.bert.ConfigConverter_Bert_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.ConfigConverter_GPTJModel_HF_CS17 method) convert_relative_attention_bias_cs16_to_cs17() (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_CS16_CS17 method) convert_relative_attention_bias_cs17_to_cs16() (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_CS16_CS17 method) convert_relative_attention_bias_cs17_to_hf() (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_HF_CS17 method) convert_state_dict_for_checkpoint() (common.pytorch.optim.AdamBase.AdamBase method) Converter_Bert_CS17_CS18 (class in common.pytorch.model_utils.checkpoint_converters.bert) Converter_BertFinetuneModel_CS16_CS17 (class in common.pytorch.model_utils.checkpoint_converters.bert_finetune) Converter_BertFinetuneModel_CS16_CS18 (class in common.pytorch.model_utils.checkpoint_converters.bert_finetune) Converter_BertForQuestionAnswering_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.bert_finetune) Converter_BertForQuestionAnswering_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.bert_finetune) Converter_BertForSequenceClassification_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.bert_finetune) Converter_BertForSequenceClassification_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.bert_finetune) Converter_BertForTokenClassification_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.bert_finetune) Converter_BertForTokenClassification_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.bert_finetune) Converter_BertLayerNorm_HF_CS (class in common.pytorch.model_utils.checkpoint_converters.bert) Converter_BertModel_CS16_CS17 (class in common.pytorch.model_utils.checkpoint_converters.bert) Converter_BertModel_CS16_CS18 (class in common.pytorch.model_utils.checkpoint_converters.bert) Converter_BertModel_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.bert) Converter_BertModel_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.bert) Converter_BertPretrainModel_CS16_CS17 (class in common.pytorch.model_utils.checkpoint_converters.bert) Converter_BertPretrainModel_CS16_CS18 (class in common.pytorch.model_utils.checkpoint_converters.bert) Converter_BertPretrainModel_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.bert) Converter_BertPretrainModel_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.bert) Converter_Codegen_Attention_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs) Converter_Codegen_Headless_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs) Converter_Codegen_Headless_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs) Converter_Codegen_LMHeadModel_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs) Converter_Codegen_LMHeadModel_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs) Converter_GPT2_Attention_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs) Converter_GPT2LMHeadModel_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs) Converter_GPT2LMHeadModel_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs) Converter_GPT2Model_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs) Converter_GPT2Model_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs) Converter_GPT_Neox_Attention_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs) Converter_GPT_Neox_Headless_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs) Converter_GPT_Neox_Headless_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs) Converter_GPT_Neox_LMHeadModel_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs) Converter_GPT_Neox_LMHeadModel_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs) Converter_GPTJ_Attention_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs) Converter_GPTJ_Headless_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs) Converter_GPTJ_Headless_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs) Converter_GPTJ_LMHeadModel_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs) Converter_GPTJ_LMHeadModel_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs) converter_notes() (in module common.pytorch.model_utils.checkpoint_converters.base_converter) Converter_T5_CS16_CS17 (class in common.pytorch.model_utils.checkpoint_converters.t5) Converter_T5_CS16_CS18 (class in common.pytorch.model_utils.checkpoint_converters.t5) Converter_T5_CS17_CS18 (class in common.pytorch.model_utils.checkpoint_converters.t5) Converter_T5_HF_CS17 (class in common.pytorch.model_utils.checkpoint_converters.t5) Converter_T5_HF_CS18 (class in common.pytorch.model_utils.checkpoint_converters.t5) ConvNormActBlock (class in vision.pytorch.layers.ConvNormActBlock) ConvNormActBlockModule (class in vision.pytorch.layers.ConvNormActBlock) ConvNormActLayers (class in vision.pytorch.layers.ConvNormActBlock) CosineAnnealingLR (class in common.pytorch.optim.lr_scheduler) (class in optim.lr_scheduler) CosineAnnealingWarmRestarts (class in common.pytorch.optim.lr_scheduler) (class in optim.lr_scheduler) CosineDecayLR (class in common.pytorch.optim.lr_scheduler) (class in optim.lr_scheduler) CosineEmbeddingLoss (class in common.pytorch.layers.CosineEmbeddingLoss) count_lines() (in module common.model_utils.count_lines) create() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner static method) create_bytes_feature() (in module common.tf.input.utils) create_dataloader() (vision.pytorch.input.classification.dataset_factory.Processor method) create_dataset() (vision.pytorch.input.classification.dataset_factory.Processor method) create_fix_pos_embedding() (common.pytorch.layers.EmbeddingLayer.EmbeddingLayer method) create_fixed_pos_emb() (common.pytorch.model_utils.RotaryPositionEmbeddingHelper.RotaryPositionEmbeddingHelper method) create_float_feature() (in module common.tf.input.utils) create_initializer() (in module common.pytorch.model_utils.create_initializer) (in module common.tf.model_utils.create_initializer) create_int_feature() (in module common.tf.input.utils) create_metric_impl_factory() (common.pytorch.metrics.cb_metric.CBMetric class method) create_preprocessing_params_with_defaults() (in module vision.pytorch.input.classification.utils) create_shuffled_idx() (vision.pytorch.input.classification.dataset_factory.Processor method) create_tf_dataset() (common.tf.input.TfRecordsProcessor.TfRecordsProcessor method) create_transform() (in module vision.pytorch.input.transforms) create_warm_start_settings() (in module common.tf.run_utils) create_worker_cache() (in module vision.pytorch.input.utils) CrossEntropyFromLogitsLayer (class in common.tf.layers.CrossEntropyFromLogitsLayer) CrossEntropyLoss (class in common.pytorch.layers.CrossEntropyLoss) CSDynamicLossScale (class in common.tf.optimizers.LossScale) CSOptimizer (class in common.pytorch.optim.CSOptimizer) CtrlFlowWarnedOnce (common.tf.layers.AbstractRecomputeWrapper.AbstractRecomputeWrapper attribute) CyclicLR (class in common.pytorch.optim.lr_scheduler) (class in optim.lr_scheduler)

D

DenseLayer (class in common.tf.layers.DenseLayer) DeviceOutputs (class in common.pytorch.metrics.cb_metric) (class in common.pytorch.summaries.cb_summary) Dice (class in vision.pytorch.losses.dice_loss) dice_coefficient_metric() (in module common.tf.metrics.dice_coefficient) DiceCELoss (class in vision.pytorch.losses.dice_loss) DiceScore (class in vision.pytorch.losses.dice_loss)

dict_to_checkpoint() (in module common.tf.run_utils) disable_collection() (common.pytorch.summary_collection.SummaryCollection method) discard() (common.pytorch.summaries.cb_summary.CBSummary method) discard_cached_summaries() (in module common.pytorch.summaries.cb_summary) divide_no_nan() (in module common.pytorch.metrics.metric_utils) DropoutLayer (class in common.tf.layers.DropoutLayer) duplicate_params_map (common.pytorch.PyTorchBaseModel.PyTorchBaseModel property) DynamicSparsityOptimizer (class in common.pytorch.optim.sparse.dynamic)

E

ece_loss_metric() (in module common.tf.metrics.ece_loss_metric) embedding_table() (common.tf.layers.EmbeddingLayer.EmbeddingLayer method) EmbeddingLayer (class in common.pytorch.layers.EmbeddingLayer) (class in common.tf.layers.EmbeddingLayer) enable_collection() (common.pytorch.summary_collection.SummaryCollection method) end() (common.tf.hooks.grad_accum_hooks.GradAccumLoggingTensorHook method) (common.tf.hooks.grad_accum_hooks.GradAccumStepCounterHook method) (common.tf.hooks.grad_accum_hooks.GradAccumSummarySaverHook method) EquivalentSubkey (class in common.pytorch.model_utils.checkpoint_converters.base_converter) est_samples_per_sec (common.pytorch.perf_utils.PerfData attribute) eval() (common.pytorch.PyTorchBaseModel.PyTorchBaseModel method) eval_epoch() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) eval_forward() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) (common.pytorch.pytorch_cs_compiler.PyTorchCSCompiler method) (common.pytorch.pytorch_runner.PyTorchRunner method)

evaluate() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) (common.pytorch.pytorch_cs_compiler.PyTorchCSCompiler method) (common.pytorch.pytorch_cs_runner.PyTorchCSRunner method) (common.pytorch.pytorch_dist_runner.PyTorchDistRunner method) ExecutionMode (class in common.tf.run_utils) ExecutionStrategy (class in common.tf.appliance_utils) exists_in_index() (common.pytorch.model_utils.checkpoint_converters.base_converter.ConversionRule method) ExponentialLR (class in common.pytorch.optim.lr_scheduler) (class in optim.lr_scheduler) extra_repr() (vision.pytorch.layers.utils.ModuleWrapperClass method) extract_text_tokens_given_cls_indices() (in module common.pytorch.metrics.rouge_score) extract_text_words_by_token_ids() (in module common.pytorch.metrics.rouge_score) extract_text_words_given_cls_indices() (in module common.pytorch.metrics.rouge_score)

F

f1_score_metric() (in module common.tf.metrics.f1_score) FastDataLoader (class in vision.pytorch.input.utils) fbeta_score_metric() (in module common.tf.metrics.fbeta_score) FeedForwardNetwork (class in common.pytorch.layers.FeedForwardNetwork) (class in common.tf.layers.FeedForwardNetwork) FeedForwardNetworkV2 (class in common.tf.layers.FeedForwardNetworkV2) file_formats() (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseCheckpointConverter static method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseCheckpointConverter_HF_CS static method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseCheckpointConverter_PT_PT static method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseConfigConverter static method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseConfigConverter_CS_CS static method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseConfigConverter_HF_CS static method) Final (class in common.pytorch.PyTorchBaseModel) find_even_buckets() (in module common.input.analyze_bucketing) flip_use_pre_encoder_decoder_layer_norm() (common.pytorch.model_utils.checkpoint_converters.t5.ConfigConverter_T5_CS17_CS18 class method) flush() (common.pytorch.summary_collection.SummaryCollection method) formats() (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseDictionaryConverter static method) (common.pytorch.model_utils.checkpoint_converters.bert.ConfigConverter_Bert_CS16_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert.ConfigConverter_Bert_CS16_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert.ConfigConverter_Bert_CS17_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert.ConfigConverter_Bert_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert.ConfigConverter_Bert_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_Bert_CS17_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertLayerNorm_HF_CS static method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertModel_CS16_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertModel_CS16_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertModel_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertModel_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertPretrainModel_CS16_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertPretrainModel_CS16_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertPretrainModel_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertPretrainModel_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.ConfigConverter_BertForQuestionAnswering_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.ConfigConverter_BertForQuestionAnswering_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.ConfigConverter_BertForSequenceClassification_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.ConfigConverter_BertForSequenceClassification_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.ConfigConverter_BertForTokenClassification_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.ConfigConverter_BertForTokenClassification_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertFinetuneModel_CS16_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertFinetuneModel_CS16_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertForQuestionAnswering_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertForQuestionAnswering_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertForSequenceClassification_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertForSequenceClassification_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertForTokenClassification_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertForTokenClassification_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.ConfigConverter_GPT2Model_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.ConfigConverter_GPT2Model_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2_Attention_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2LMHeadModel_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2LMHeadModel_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2Model_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2Model_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.ConfigConverter_GPT_Neox_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.ConfigConverter_GPT_Neox_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_Attention_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_Headless_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_Headless_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_LMHeadModel_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_LMHeadModel_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.ConfigConverter_GPTJModel_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.ConfigConverter_GPTJModel_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.Converter_GPTJ_Attention_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.Converter_GPTJ_Headless_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.Converter_GPTJ_Headless_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.Converter_GPTJ_LMHeadModel_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.Converter_GPTJ_LMHeadModel_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.ConfigConverter_Codegen_Model_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.ConfigConverter_Codegen_Model_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_Attention_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_Headless_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_Headless_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_LMHeadModel_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_LMHeadModel_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.t5.ConfigConverter_T5_CS16_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.t5.ConfigConverter_T5_CS16_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.t5.ConfigConverter_T5_CS17_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.t5.ConfigConverter_T5_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.t5.ConfigConverter_T5_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_CS16_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_CS16_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_CS17_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_HF_CS18 static method)

forward() (common.pytorch.layers.AlibiPositionEmbeddingLayer.AlibiPositionEmbeddingLayer method) (common.pytorch.layers.AttentionLayer.MultiheadAttention method) (common.pytorch.layers.BiaslessLayerNorm.BiaslessLayerNorm method) (common.pytorch.layers.EmbeddingLayer.EmbeddingLayer method) (common.pytorch.layers.FeedForwardNetwork.FeedForwardNetwork method) (common.pytorch.layers.FeedForwardNetwork.SingleFeedForwardLayer method) (common.pytorch.layers.GaussianNLLLoss.GaussianNLLLoss method) (common.pytorch.layers.HuberLoss.HuberLoss method) (common.pytorch.layers.MultiMarginLoss.MultiMarginLoss method) (common.pytorch.layers.RelativePositionEmbeddingLayer.RelativePositionEmbeddingLayer method) (common.pytorch.layers.SmoothL1Loss.SmoothL1Loss method) (common.pytorch.model_utils.BertPretrainModelLoss.BertPretrainModelLoss method) (common.pytorch.model_utils.GPTLMHeadModelLoss.GPTLMHeadModelLoss method) (common.pytorch.model_utils.T5ForConditionalGenerationLoss.T5ForConditionalGenerationLoss method) (vision.pytorch.input.classification.mixup.RandomCutmix method) (vision.pytorch.input.classification.mixup.RandomMixup method) (vision.pytorch.layers.BatchChannelNorm.BatchChannelNorm2D method) (vision.pytorch.layers.ConvNormActBlock.ConvNormActBlockModule method) (vision.pytorch.layers.GroupInstanceNorm.GroupInstanceNorm method) (vision.pytorch.layers.StandardizedConvolutionLayer.StdConv1d method) (vision.pytorch.layers.StandardizedConvolutionLayer.StdConv2d method) (vision.pytorch.layers.StandardizedConvolutionLayer.StdConv3d method) (vision.pytorch.layers.utils.ModuleWrapperClass method) (vision.pytorch.losses.dice_loss.DiceCELoss method) from_dict() (common.pytorch.summaries.tensor_summary.TensorDescriptor static method)

G

gather_last_tokens() (in module common.tf.model_utils.gather) GaussianNLLLoss (class in common.pytorch.layers.GaussianNLLLoss) gelu() (common.tf.layers.ActivationLayer.ActivationLayer static method) generate_input_fn() (in module common.tf.input.utils) get_all_metrics() (in module common.pytorch.metrics.cb_metric) get_all_summaries() (in module common.pytorch.summaries.cb_summary) get_attention_module() (in module common.pytorch.layers.AttentionHelper) get_checkpoints() (in module common.pytorch.utils) get_config() (common.tf.optimizers.LossScale.CSDynamicLossScale method) get_config_converter_class() (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseCheckpointConverter static method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_Bert_CS17_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertLayerNorm_HF_CS static method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertModel_CS16_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertModel_CS16_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertModel_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertModel_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertPretrainModel_CS16_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertPretrainModel_CS16_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertPretrainModel_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertPretrainModel_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertFinetuneModel_CS16_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertFinetuneModel_CS16_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertForQuestionAnswering_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertForQuestionAnswering_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertForSequenceClassification_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertForSequenceClassification_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertForTokenClassification_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertForTokenClassification_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2_Attention_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2LMHeadModel_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2LMHeadModel_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2Model_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2Model_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_Attention_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_Headless_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_Headless_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_LMHeadModel_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_LMHeadModel_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.Converter_GPTJ_Attention_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.Converter_GPTJ_Headless_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.Converter_GPTJ_Headless_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.Converter_GPTJ_LMHeadModel_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.Converter_GPTJ_LMHeadModel_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_Attention_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_Headless_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_Headless_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_LMHeadModel_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_LMHeadModel_HF_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_CS16_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_CS16_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_CS17_CS18 static method) (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_HF_CS17 static method) (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_HF_CS18 static method)

get_csconfig() (in module common.tf.run_utils) get_csrunconfig_dict() (in module common.tf.run_utils) get_debug_args() (in module common.pytorch.utils) (in module common.tf.appliance_utils) get_debug_mgr_args() (in module common.tf.appliance_utils) get_embedding() (common.pytorch.layers.RelativePositionEmbeddingLayer.RelativePositionEmbeddingLayer method) get_execution_mode() (in module common.tf.run_utils) get_from_index() (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseDictionaryConverter class method) get_grad_accum_hooks() (in module common.tf.hooks.grad_accum_hooks) get_input_checkpoint_steps() (in module common.tf.run_utils) get_input_dtype() (in module common.pytorch.utils) get_input_embeddings() (common.pytorch.layers.EmbeddingLayer.EmbeddingLayer method) get_input_fn_params() (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) get_layers() (vision.pytorch.layers.ConvNormActBlock.ConvNormActLayers method) get_learning_rate() (common.tf.optimizers.Trainer.Trainer method) get_loss_value() (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) get_lr() (common.pytorch.optim.lr_scheduler.LRScheduler method) get_lr_scheduler() (common.pytorch.PyTorchBaseModel.PyTorchBaseModel method) get_normalization() (in module vision.pytorch.layers.normalizations) get_optimizer() (common.pytorch.PyTorchBaseModel.PyTorchBaseModel method) get_params() (in module common.tf.run_utils) get_predict_directory() (in module common.tf.run_utils) get_preprocess_transform() (in module vision.pytorch.input.classification.preprocessing) get_sequence_length() (common.pytorch.layers.AttentionLayer.MultiheadAttention method) get_state() (common.pytorch.PyTorchBaseModel.PyTorchBaseModel method) get_vocab_size() (in module common.tf.model_utils.vocab_utils) get_weight_dict() (in module common.tf.run_utils) GetWeights (class in common.tf.run_utils) global_start_step (common.pytorch.optim.lr_scheduler.LRScheduler attribute) GMPSparsityOptimizer (class in common.pytorch.optim.sparse.dynamic) GPTLMHeadModelLoss (class in common.pytorch.model_utils.GPTLMHeadModelLoss) grad_accum_steps (common.tf.optimizers.Trainer.Trainer property) GradAccumLoggingTensorHook (class in common.tf.hooks.grad_accum_hooks) GradAccumOptimizer (class in common.tf.optimizers.GradAccumOptimizer) GradAccumStepCounterHook (class in common.tf.hooks.grad_accum_hooks) GradAccumSummarySaverHook (class in common.tf.hooks.grad_accum_hooks) gradient_global_norm (common.tf.optimizers.Trainer.Trainer property) GradientClipper (class in common.pytorch.gradient_clipper) group_optimizer_params() (in module common.pytorch.utils) GroupInstanceNorm (class in vision.pytorch.layers.GroupInstanceNorm)

H

half_dtype (common.pytorch.half_dtype.HalfDType property) HalfDType (class in common.pytorch.half_dtype)

hard_swish() (common.tf.layers.ActivationLayer.ActivationLayer static method) HingeEmbeddingLoss (class in common.pytorch.layers.HingeEmbeddingLoss) HuberLoss (class in common.pytorch.layers.HuberLoss)

I

increment_global_step() (optim.Optimizer method) init_sparsity() (common.pytorch.optim.sparse.base.BaseSparsityOptimizer method) (common.pytorch.optim.sparse.dynamic.DynamicSparsityOptimizer method) (common.pytorch.optim.sparse.static.StaticSparsityOptimizer method) init_state() (common.pytorch.metrics.cb_metric.CBMetric method) initial_epoch (common.pytorch.optim.lr_scheduler.LRScheduler attribute) interleave_helper() (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_Attention_HF_CS17 method) InverseExponentialTimeDecayLR (class in common.pytorch.optim.lr_scheduler) (class in optim.lr_scheduler)

InverseSquareRootDecayLR (class in common.pytorch.optim.lr_scheduler) (class in optim.lr_scheduler) is_cs() (in module common.tf.run_utils) is_gpu_distributed() (in module vision.pytorch.input.utils) is_grad_accum() (common.tf.optimizers.Trainer.Trainer method) is_in_while_loop() (common.tf.layers.AbstractRecomputeWrapper.AbstractRecomputeWrapper static method) is_master_ordinal() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) (common.pytorch.pytorch_dist_runner.PyTorchDistRunner method)

K

KLDivLoss (class in common.pytorch.layers.KLDivLoss)

kwargs (common.pytorch.metrics.cb_metric.DeviceOutputs attribute) (common.pytorch.summaries.cb_summary.DeviceOutputs attribute)

L

L1Loss (class in common.pytorch.layers.L1Loss) Lamb (class in optim) LambdaLR (class in common.pytorch.optim.lr_scheduler) (class in optim.lr_scheduler) LayerNormalizationLayer (class in common.tf.layers.LayerNormalizationLayer) lecun_normal_() (in module common.pytorch.model_utils.weight_initializers) lecun_uniform_() (in module common.pytorch.model_utils.weight_initializers) LinearLR (class in optim.lr_scheduler) load() (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseCheckpointConverter class method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseCheckpointConverter_PT_PT class method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseConfigConverter class method) (common.pytorch.summaries.tensor_summary.TensorSummaryReader method) load_state_dict() (common.pytorch.optim.AdamBase.Adam method) (common.pytorch.optim.AdamBase.AdamBase method) (common.pytorch.optim.AdamBase.AdamW method) (common.pytorch.optim.CSOptimizer.CSOptimizer method) (common.pytorch.optim.lr_scheduler.LRScheduler method) (common.pytorch.optim.lr_scheduler.SequentialLR method) (common.pytorch.optim.sparse.base.BaseSparsityOptimizer method) (common.pytorch.optim.sparse.dynamic.DynamicSparsityOptimizer method) (optim.Adam method) (optim.AdamW method) (optim.lr_scheduler.ChainedScheduler method) (optim.lr_scheduler.SequentialLR method) (optim.Optimizer method)

log_summaries (common.tf.optimizers.Trainer.Trainer property) log_training_summaries() (common.tf.optimizers.Trainer.Trainer method) loss_scale (common.tf.optimizers.GradAccumOptimizer.GradAccumOptimizer property) (common.tf.optimizers.LossScale.MixedPrecisionLossScaleOptimizerAdapter property) loss_scale_value (common.tf.optimizers.Trainer.Trainer property) lr_function() (common.pytorch.optim.lr_scheduler.ChainedScheduler method) (common.pytorch.optim.lr_scheduler.LRScheduler method) (common.pytorch.optim.lr_scheduler.SequentialLR method) lr_scheduler_step() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) LRScheduler (class in common.pytorch.optim.lr_scheduler)

M

main() (in module common.input.analyze_bucketing) make_mask_top_atleast_k() (in module common.pytorch.optim.sparse.utils) make_mask_topk_k() (in module common.pytorch.optim.sparse.utils) make_mask_topk_sparsity() (in module common.pytorch.optim.sparse.utils) map_fn() (common.tf.input.TfRecordsProcessor.TfRecordsProcessor method) MarginRankingLoss (class in common.pytorch.layers.MarginRankingLoss) MaxPool2DLayer (class in common.tf.layers.MaxPool2DLayer) maybe_get_loss_value() (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) mcc_metric() (in module common.tf.metrics.mcc) merge() (common.pytorch.perf_utils.PerfData method) metric_variable() (in module common.tf.metrics.utils) MixedPrecisionLossScaleOptimizerAdapter (class in common.tf.optimizers.LossScale) module common.input common.input.analyze_bucketing common.input.utils common.model_utils common.model_utils.count_lines common.pytorch common.pytorch.gradient_clipper common.pytorch.half_dtype common.pytorch.input_utils common.pytorch.layers common.pytorch.layers.AlibiPositionEmbeddingLayer common.pytorch.layers.AttentionHelper common.pytorch.layers.AttentionLayer common.pytorch.layers.BCELoss common.pytorch.layers.BCEWithLogitsLoss common.pytorch.layers.BiaslessLayerNorm common.pytorch.layers.CosineEmbeddingLoss common.pytorch.layers.CrossEntropyLoss common.pytorch.layers.EmbeddingLayer common.pytorch.layers.FeedForwardNetwork common.pytorch.layers.GaussianNLLLoss common.pytorch.layers.GPTJDecoderLayer common.pytorch.layers.HingeEmbeddingLoss common.pytorch.layers.HuberLoss common.pytorch.layers.KLDivLoss common.pytorch.layers.L1Loss common.pytorch.layers.MarginRankingLoss common.pytorch.layers.MSELoss common.pytorch.layers.MultiLabelSoftMarginLoss common.pytorch.layers.MultiMarginLoss common.pytorch.layers.NLLLoss common.pytorch.layers.PoissonNLLLoss common.pytorch.layers.RelativePositionEmbeddingLayer common.pytorch.layers.SmoothL1Loss common.pytorch.layers.Transformer common.pytorch.layers.TransformerDecoder common.pytorch.layers.TransformerDecoderLayer common.pytorch.layers.TransformerEncoder common.pytorch.layers.TransformerEncoderLayer common.pytorch.layers.TripletMarginLoss common.pytorch.layers.TripletMarginWithDistanceLoss common.pytorch.layers.utils common.pytorch.metrics common.pytorch.metrics.accuracy common.pytorch.metrics.auc common.pytorch.metrics.cb_metric common.pytorch.metrics.dice_coefficient common.pytorch.metrics.fbeta_score common.pytorch.metrics.mean_iou common.pytorch.metrics.metric_utils common.pytorch.metrics.perplexity common.pytorch.metrics.precision_at_k common.pytorch.metrics.recall_at_k common.pytorch.metrics.rouge_score common.pytorch.model_utils common.pytorch.model_utils.BertPretrainModelLoss common.pytorch.model_utils.checkpoint_converters common.pytorch.model_utils.checkpoint_converters.base_converter common.pytorch.model_utils.checkpoint_converters.bert common.pytorch.model_utils.checkpoint_converters.bert_finetune common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs common.pytorch.model_utils.checkpoint_converters.t5 common.pytorch.model_utils.convert_checkpoint common.pytorch.model_utils.create_initializer common.pytorch.model_utils.GPTLMHeadModelLoss common.pytorch.model_utils.RotaryPositionEmbeddingHelper common.pytorch.model_utils.T5ForConditionalGenerationLoss common.pytorch.model_utils.weight_initializers common.pytorch.optim common.pytorch.optim.Adadelta common.pytorch.optim.Adafactor common.pytorch.optim.Adagrad common.pytorch.optim.Adamax common.pytorch.optim.AdamBase common.pytorch.optim.ASGD common.pytorch.optim.CSOptimizer common.pytorch.optim.Lamb common.pytorch.optim.lr_scheduler common.pytorch.optim.NAdam common.pytorch.optim.RAdam common.pytorch.optim.RMSprop common.pytorch.optim.Rprop common.pytorch.optim.SGD common.pytorch.optim.sparse common.pytorch.optim.sparse.base common.pytorch.optim.sparse.dynamic common.pytorch.optim.sparse.static common.pytorch.optim.sparse.utils common.pytorch.optim.utils common.pytorch.perf_utils common.pytorch.pytorch_base_cs_runner common.pytorch.pytorch_base_runner common.pytorch.pytorch_cs_appliance common.pytorch.pytorch_cs_compiler common.pytorch.pytorch_cs_runner common.pytorch.pytorch_dist_runner common.pytorch.pytorch_runner common.pytorch.PyTorchBaseModel common.pytorch.run_cstorch_flow common.pytorch.run_utils common.pytorch.summaries common.pytorch.summaries.cb_summary common.pytorch.summaries.scalar_summary common.pytorch.summaries.tensor_summary common.pytorch.summary_collection common.pytorch.utils common.tf common.tf.appliance_utils common.tf.estimator common.tf.hooks common.tf.hooks.grad_accum_hooks common.tf.input common.tf.input.TfRecordsProcessor common.tf.input.utils common.tf.layers common.tf.layers.AbstractRecomputeWrapper common.tf.layers.ActivationLayer common.tf.layers.AddLayer common.tf.layers.AttentionLayer common.tf.layers.BaseLayer common.tf.layers.Conv2DLayer common.tf.layers.Conv2DTransposeLayer common.tf.layers.CrossEntropyFromLogitsLayer common.tf.layers.DenseLayer common.tf.layers.DropoutLayer common.tf.layers.EmbeddingLayer common.tf.layers.FeedForwardNetwork common.tf.layers.FeedForwardNetworkV2 common.tf.layers.Input common.tf.layers.LayerNormalizationLayer common.tf.layers.MaxPool2DLayer common.tf.layers.PoolerLayer common.tf.layers.PoolerLayerV2 common.tf.layers.PositionEmbeddingLayer common.tf.layers.PrePostProcessWrapper common.tf.layers.ReshapeLayer common.tf.layers.SegmentEmbeddingLayer common.tf.layers.SharedWeightsDenseLayer common.tf.layers.SoftmaxLayer common.tf.layers.SquaredErrorLayer common.tf.metrics common.tf.metrics.accuracy common.tf.metrics.bits_per_x common.tf.metrics.dice_coefficient common.tf.metrics.ece_loss_metric common.tf.metrics.f1_score common.tf.metrics.fbeta_score common.tf.metrics.mcc common.tf.metrics.perplexity common.tf.metrics.rouge_score common.tf.metrics.utils common.tf.model_utils common.tf.model_utils.create_initializer common.tf.model_utils.gather common.tf.model_utils.reshape_gather common.tf.model_utils.scale_loss_value common.tf.model_utils.shard_dataset common.tf.model_utils.vocab_utils common.tf.optimizers common.tf.optimizers.AdamWOptimizer common.tf.optimizers.GradAccumOptimizer common.tf.optimizers.LossScale common.tf.optimizers.Trainer common.tf.run_utils common.tf.TFBaseModel optim optim.lr_scheduler vision.pytorch vision.pytorch.input vision.pytorch.input.classification vision.pytorch.input.classification.dataset_factory vision.pytorch.input.classification.mixup vision.pytorch.input.classification.preprocessing vision.pytorch.input.classification.sampler vision.pytorch.input.classification.utils vision.pytorch.input.transforms vision.pytorch.input.utils vision.pytorch.layers vision.pytorch.layers.BatchChannelNorm vision.pytorch.layers.ConvNormActBlock vision.pytorch.layers.GroupInstanceNorm vision.pytorch.layers.normalizations vision.pytorch.layers.StandardizedConvolutionLayer vision.pytorch.layers.utils vision.pytorch.losses vision.pytorch.losses.dice_loss

ModuleWrapperClass (class in vision.pytorch.layers.utils) MSELoss (class in common.pytorch.layers.MSELoss) MultiheadAttention (class in common.pytorch.layers.AttentionLayer) MultiLabelSoftMarginLoss (class in common.pytorch.layers.MultiLabelSoftMarginLoss) MultiMarginLoss (class in common.pytorch.layers.MultiMarginLoss) MultiplicativeLR (class in common.pytorch.optim.lr_scheduler) (class in optim.lr_scheduler) MultiStepLR (class in common.pytorch.optim.lr_scheduler) (class in optim.lr_scheduler)

N

NAdam (class in optim) name (common.pytorch.metrics.cb_metric.CBMetric property) (common.pytorch.summaries.cb_summary.CBSummary property) names() (common.pytorch.summaries.tensor_summary.TensorSummaryReader method)

next() (common.pytorch.utils.SampleGenerator method) NLLLoss (class in common.pytorch.layers.NLLLoss) ns_since_epoch (common.pytorch.summaries.tensor_summary.TensorDescriptor attribute) num_tasks() (in module vision.pytorch.input.utils) num_updates (common.pytorch.metrics.cb_metric.CBMetric property)

O

on_checkpoint_saved() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) on_device_state_dict() (common.pytorch.metrics.cb_metric.CBMetric method) on_eval_batch_end() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) (common.pytorch.pytorch_cs_compiler.PyTorchCSCompiler method) (common.pytorch.pytorch_dist_runner.PyTorchDistRunner method) on_eval_batch_start() (common.pytorch.pytorch_base_cs_runner.PyTorchBaseCSRunner method) (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) (common.pytorch.pytorch_dist_runner.PyTorchDistRunner method) (common.pytorch.pytorch_runner.PyTorchRunner method) on_eval_end() (common.pytorch.pytorch_base_cs_runner.PyTorchBaseCSRunner method) (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) (common.pytorch.pytorch_cs_runner.PyTorchCSRunner method) (common.pytorch.pytorch_dist_runner.PyTorchDistRunner method) (common.pytorch.pytorch_runner.PyTorchRunner method) on_eval_epoch_end() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) (common.pytorch.pytorch_cs_compiler.PyTorchCSCompiler method) (common.pytorch.pytorch_cs_runner.PyTorchCSRunner method) on_eval_epoch_start() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) on_eval_start() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) (common.pytorch.pytorch_cs_compiler.PyTorchCSCompiler method) (common.pytorch.pytorch_cs_runner.PyTorchCSRunner method) on_process_start() (common.pytorch.pytorch_dist_runner.PyTorchDistRunner method) on_train_batch_end() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) (common.pytorch.pytorch_cs_compiler.PyTorchCSCompiler method) (common.pytorch.pytorch_dist_runner.PyTorchDistRunner method) on_train_batch_start() (common.pytorch.pytorch_base_cs_runner.PyTorchBaseCSRunner method) (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) (common.pytorch.pytorch_dist_runner.PyTorchDistRunner method) (common.pytorch.pytorch_runner.PyTorchRunner method)

on_train_end() (common.pytorch.pytorch_base_cs_runner.PyTorchBaseCSRunner method) (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) (common.pytorch.pytorch_cs_runner.PyTorchCSRunner method) (common.pytorch.pytorch_dist_runner.PyTorchDistRunner method) (common.pytorch.pytorch_runner.PyTorchRunner method) on_train_epoch_end() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) (common.pytorch.pytorch_cs_compiler.PyTorchCSCompiler method) (common.pytorch.pytorch_cs_runner.PyTorchCSRunner method) (common.pytorch.pytorch_dist_runner.PyTorchDistRunner method) on_train_epoch_start() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_dist_runner.PyTorchDistRunner method) on_train_start() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) (common.pytorch.pytorch_cs_compiler.PyTorchCSCompiler method) (common.pytorch.pytorch_cs_runner.PyTorchCSRunner method) (common.pytorch.pytorch_runner.PyTorchRunner method) OneCycleLR (class in common.pytorch.optim.lr_scheduler) (class in optim.lr_scheduler) optim module optim.lr_scheduler module Optimizer (class in optim) optimizer (common.tf.optimizers.LossScale.MixedPrecisionLossScaleOptimizerAdapter property) (common.tf.optimizers.Trainer.Trainer property) optimizer_step() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) (common.pytorch.pytorch_cs_runner.PyTorchCSRunner method) optimizer_step_with_summaries() (in module common.pytorch.run_cstorch_flow) optimizer_zero_grad() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) OutsideCerebras (common.tf.run_utils.ExecutionMode attribute)

P

parse_args() (in module common.input.analyze_bucketing) parse_args_and_params() (in module common.tf.appliance_utils) PerfData (class in common.pytorch.perf_utils) perplexity_metric() (in module common.tf.metrics.perplexity) PiecewiseConstantLR (class in common.pytorch.optim.lr_scheduler) (class in optim.lr_scheduler) pipeline (common.tf.appliance_utils.ExecutionStrategy attribute) Pipeline (common.tf.run_utils.ExecutionMode attribute) PoissonNLLLoss (class in common.pytorch.layers.PoissonNLLLoss) PolynomialLR (class in common.pytorch.optim.lr_scheduler) (class in optim.lr_scheduler) PoolerLayer (class in common.tf.layers.PoolerLayer) PoolerLayerV2 (class in common.tf.layers.PoolerLayerV2) position_embedding_helper() (common.pytorch.layers.EmbeddingLayer.EmbeddingLayer method) position_embeddings_convert() (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertModel_CS16_CS17 method) PositionEmbeddingLayer (class in common.tf.layers.PositionEmbeddingLayer) post_checkpoint_convert() (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseCheckpointConverter method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseCheckpointConverter_HF_CS method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertModel_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertPretrainModel_CS16_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertPretrainModel_CS16_CS18 method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertPretrainModel_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertFinetuneModel_CS16_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertFinetuneModel_CS16_CS18 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertForSequenceClassification_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.Converter_BertForTokenClassification_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_CS16_CS17 method) (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_CS16_CS18 method) (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_HF_CS17 method) post_config_convert() (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseConfigConverter method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseConfigConverter_CS_CS method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseConfigConverter_HF_CS method) (common.pytorch.model_utils.checkpoint_converters.bert.ConfigConverter_Bert_CS16_CS18 method) (common.pytorch.model_utils.checkpoint_converters.bert.ConfigConverter_Bert_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert.ConfigConverter_Bert_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.ConfigConverter_BertForQuestionAnswering_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.ConfigConverter_BertForTokenClassification_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.ConfigConverter_GPT_Neox_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.ConfigConverter_GPTJModel_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.ConfigConverter_Codegen_Model_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.t5.ConfigConverter_T5_HF_CS17 method) post_load_state_dict() (common.pytorch.optim.CSOptimizer.CSOptimizer method) post_model_convert() (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseCheckpointConverter method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2Model_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_Headless_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.Converter_GPTJ_Headless_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_Headless_HF_CS17 method) pre_checkpoint_convert() (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseCheckpointConverter method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseCheckpointConverter_HF_CS method) pre_config_convert() (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseConfigConverter method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseConfigConverter_CS_CS method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseConfigConverter_HF_CS method) (common.pytorch.model_utils.checkpoint_converters.bert.ConfigConverter_Bert_CS16_CS18 method) (common.pytorch.model_utils.checkpoint_converters.bert.ConfigConverter_Bert_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert.ConfigConverter_Bert_HF_CS18 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.ConfigConverter_BertForQuestionAnswering_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.ConfigConverter_BertForSequenceClassification_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert_finetune.ConfigConverter_BertForTokenClassification_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.ConfigConverter_GPT2Model_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.ConfigConverter_GPT_Neox_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.ConfigConverter_GPTJModel_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.ConfigConverter_Codegen_Model_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.t5.ConfigConverter_T5_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.t5.ConfigConverter_T5_HF_CS18 method)

pre_model_convert() (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseCheckpointConverter method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertModel_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.bert.Converter_BertPretrainModel_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2LMHeadModel_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2Model_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_Headless_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_LMHeadModel_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.Converter_GPTJ_Headless_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.Converter_GPTJ_LMHeadModel_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_Headless_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_LMHeadModel_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.t5.Converter_T5_HF_CS17 method) preinitialize() (common.pytorch.optim.AdamBase.AdamBase method) (common.pytorch.optim.CSOptimizer.CSOptimizer method) (optim.Adadelta method) (optim.Adafactor method) (optim.Adagrad method) (optim.Adamax method) (optim.ASGD method) (optim.Lamb method) (optim.NAdam method) (optim.Optimizer method) (optim.RAdam method) (optim.RMSprop method) (optim.Rprop method) (optim.SGD method) PrePostProcessWrapper (class in common.tf.layers.PrePostProcessWrapper) print_eval_metrics() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) process_attention_mask() (common.pytorch.layers.AttentionLayer.MultiheadAttention method) process_k_before_logits_calc() (common.pytorch.layers.AttentionLayer.MultiheadAttention method) process_key_padding_mask() (common.pytorch.layers.AttentionLayer.MultiheadAttention method) process_past_kv() (common.pytorch.layers.AttentionLayer.MultiheadAttention method) process_q_before_logits_calc() (common.pytorch.layers.AttentionLayer.MultiheadAttention method) process_schedule() (common.pytorch.optim.sparse.dynamic.DynamicSparsityOptimizer method) process_transform() (vision.pytorch.input.classification.dataset_factory.Processor method) process_v_before_logits_calc() (common.pytorch.layers.AttentionLayer.MultiheadAttention method) Processor (class in vision.pytorch.input.classification.dataset_factory) programming_time (common.pytorch.perf_utils.PerfData attribute) PyTorchBaseCSRunner (class in common.pytorch.pytorch_base_cs_runner) PyTorchBaseModel (class in common.pytorch.PyTorchBaseModel) PyTorchBaseRunner (class in common.pytorch.pytorch_base_runner) PyTorchCSAppliance (class in common.pytorch.pytorch_cs_appliance) PyTorchCSCompiler (class in common.pytorch.pytorch_cs_compiler) PyTorchCSRunner (class in common.pytorch.pytorch_cs_runner) PyTorchDistRunner (class in common.pytorch.pytorch_dist_runner) PyTorchRunner (class in common.pytorch.pytorch_runner)

Q

qkv_converter() (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_Attention_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_Attention_HF_CS17 method) qkv_converter_cs17_to_hf() (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_Attention_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_Attention_HF_CS17 method)

qkv_converter_hf_to_cs17() (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_Attention_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_Attention_HF_CS17 method)

R

RAdam (class in optim) RandomCutmix (class in vision.pytorch.input.classification.mixup) RandomMixup (class in vision.pytorch.input.classification.mixup) relative_position_bucket() (common.pytorch.layers.RelativePositionEmbeddingLayer.RelativePositionEmbeddingLayer static method) RelativePositionEmbeddingLayer (class in common.pytorch.layers.RelativePositionEmbeddingLayer) relu6() (common.tf.layers.ActivationLayer.ActivationLayer static method) RepeatedAugSampler (class in vision.pytorch.input.classification.sampler) replace_final_norm() (common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs.Converter_GPT2Model_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_Headless_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.Converter_GPTJ_Headless_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_Headless_HF_CS17 method) replace_or_fill_masked_bias() (common.pytorch.model_utils.checkpoint_converters.gptj_hf_cs.Converter_GPTJ_Attention_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.Converter_Codegen_Attention_HF_CS17 method) replaceKey() (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseDictionaryConverter static method) reset() (common.pytorch.metrics.cb_metric.CBMetric method) reset_all_metrics() (in module common.pytorch.metrics.cb_metric) reset_parameters() (common.pytorch.layers.AlibiPositionEmbeddingLayer.AlibiPositionEmbeddingLayer method) (common.pytorch.layers.AttentionLayer.MultiheadAttention method) (common.pytorch.layers.EmbeddingLayer.EmbeddingLayer method) (common.pytorch.layers.FeedForwardNetwork.FeedForwardNetwork method) (common.pytorch.layers.RelativePositionEmbeddingLayer.RelativePositionEmbeddingLayer method) (vision.pytorch.layers.BatchChannelNorm.BatchChannelNorm2D method)

reset_state() (common.pytorch.metrics.cb_metric.CBMetric method) reshape_gather() (in module common.tf.model_utils.reshape_gather) ReshapeLayer (class in common.tf.layers.ReshapeLayer) reverse_interleave_helper() (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.Converter_GPT_Neox_Attention_HF_CS17 method) RMSprop (class in optim) rotary_dim_converter() (common.pytorch.model_utils.checkpoint_converters.gpt_neox_hf_cs.ConfigConverter_GPT_Neox_HF_CS17 method) (common.pytorch.model_utils.checkpoint_converters.salesforce_codegen_hf_cs.ConfigConverter_Codegen_Model_HF_CS17 method) RotaryPositionEmbeddingHelper (class in common.pytorch.model_utils.RotaryPositionEmbeddingHelper) rotate_tensor() (common.pytorch.model_utils.RotaryPositionEmbeddingHelper.RotaryPositionEmbeddingHelper method) rouge_score_metric() (in module common.tf.metrics.rouge_score) Rprop (class in optim) run() (in module common.pytorch.run_utils) run_appliance() (in module common.tf.appliance_utils) run_base_model_flow() (in module common.pytorch.run_utils) run_cstorch_eval() (in module common.pytorch.run_cstorch_flow) run_cstorch_flow() (in module common.pytorch.run_cstorch_flow) run_cstorch_train() (in module common.pytorch.run_cstorch_flow) run_on_device() (common.pytorch.summaries.cb_summary.CBSummary method) (common.pytorch.summaries.tensor_summary.TensorSummary method) run_on_host() (common.pytorch.summaries.cb_summary.CBSummary method) (common.pytorch.summaries.tensor_summary.TensorSummary method) run_with_params() (in module common.pytorch.run_utils) RunConfigParamsValidator (class in common.pytorch.utils)

S

SampleGenerator (class in common.pytorch.utils) samples_per_sec (common.pytorch.perf_utils.PerfData attribute) save() (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseCheckpointConverter class method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseCheckpointConverter_PT_PT class method) (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseConfigConverter class method) (common.pytorch.summaries.cb_summary.CBSummary method) save_all_summaries() (in module common.pytorch.summaries.cb_summary) save_on_host() (common.pytorch.summaries.cb_summary.CBSummary method) (common.pytorch.summaries.tensor_summary.TensorSummary method) save_params() (in module common.input.utils) (in module common.tf.run_utils) save_perf() (in module common.pytorch.perf_utils) save_predictions() (in module common.tf.run_utils) scalar_summary() (in module common.pytorch.summaries.scalar_summary) scale_loss_value() (in module common.tf.model_utils.scale_loss_value) segment_is_converter() (common.pytorch.model_utils.checkpoint_converters.base_converter.ConversionRule static method) SegmentEmbeddingLayer (class in common.tf.layers.SegmentEmbeddingLayer) SelfAttentionLayer (class in common.tf.layers.AttentionLayer) SequentialLR (class in common.pytorch.optim.lr_scheduler) (class in optim.lr_scheduler) set_epoch() (vision.pytorch.input.classification.sampler.RepeatedAugSampler method) set_input_embeddings() (common.pytorch.layers.EmbeddingLayer.EmbeddingLayer method) set_lr_lambda() (common.pytorch.optim.lr_scheduler.LambdaLR method) (common.pytorch.optim.lr_scheduler.MultiplicativeLR method) (optim.lr_scheduler.LambdaLR method) (optim.lr_scheduler.MultiplicativeLR method) set_max_gradients() (common.pytorch.gradient_clipper.GradientClipper method) set_scale_fn() (common.pytorch.optim.lr_scheduler.CyclicLR method) (optim.lr_scheduler.CyclicLR method) set_state() (common.pytorch.PyTorchBaseModel.PyTorchBaseModel method) set_transforms() (vision.pytorch.input.classification.dataset_factory.VisionSubset method) setup_environment() (in module common.tf.run_utils) setup_fixed_position_embedding() (common.tf.layers.PositionEmbeddingLayer.PositionEmbeddingLayer method) setup_logging() (in module common.pytorch.utils) (in module common.tf.appliance_utils) SetupInputTensor() (in module common.tf.layers.Input) SGD (class in optim) shard_dataset() (in module common.tf.model_utils.shard_dataset) ShardedSampler (class in vision.pytorch.input.utils) SharedWeightsDenseLayer (class in common.tf.layers.SharedWeightsDenseLayer) sideband_eval_all() (in module common.pytorch.run_utils) sideband_train_eval_all() (in module common.pytorch.run_utils) SingleFeedForwardLayer (class in common.pytorch.layers.FeedForwardNetwork) smooth_loss() (in module common.pytorch.model_utils.T5ForConditionalGenerationLoss) SmoothL1Loss (class in common.pytorch.layers.SmoothL1Loss) SoftmaxLayer (class in common.tf.layers.SoftmaxLayer) sparse_params() (common.pytorch.optim.sparse.base.BaseSparsityOptimizer method) split_dataset() (vision.pytorch.input.classification.dataset_factory.Processor method) SquaredErrorLayer (class in common.tf.layers.SquaredErrorLayer) state_dict() (common.pytorch.optim.lr_scheduler.ChainedScheduler method) (common.pytorch.optim.lr_scheduler.LRScheduler method) (common.pytorch.optim.lr_scheduler.SequentialLR method) (common.pytorch.optim.sparse.base.BaseSparsityOptimizer method) (common.pytorch.optim.sparse.dynamic.DynamicSparsityOptimizer method) (optim.lr_scheduler.ChainedScheduler method) (optim.lr_scheduler.SequentialLR method) (optim.Optimizer method)

state_names_to_sparsify() (common.pytorch.optim.AdamBase.AdamBase method) (common.pytorch.optim.CSOptimizer.CSOptimizer method) (optim.Adadelta method) (optim.Adafactor method) (optim.Adagrad method) (optim.Adamax method) (optim.ASGD method) (optim.Lamb method) (optim.NAdam method) (optim.Optimizer method) (optim.RAdam method) (optim.RMSprop method) (optim.Rprop method) (optim.SGD method) StaticSparsityOptimizer (class in common.pytorch.optim.sparse.static) StdConv1d (class in vision.pytorch.layers.StandardizedConvolutionLayer) StdConv2d (class in vision.pytorch.layers.StandardizedConvolutionLayer) StdConv3d (class in vision.pytorch.layers.StandardizedConvolutionLayer) step (common.pytorch.summaries.tensor_summary.TensorDescriptor attribute) step() (common.pytorch.optim.AdamBase.AdamBase method) (common.pytorch.optim.CSOptimizer.CSOptimizer method) (common.pytorch.optim.lr_scheduler.ChainedScheduler method) (common.pytorch.optim.lr_scheduler.LRScheduler method) (common.pytorch.optim.lr_scheduler.SequentialLR method) (common.pytorch.optim.sparse.dynamic.DynamicSparsityOptimizer method) (common.pytorch.optim.sparse.static.StaticSparsityOptimizer method) (optim.Adadelta method) (optim.Adafactor method) (optim.Adagrad method) (optim.Adamax method) (optim.ASGD method) (optim.Lamb method) (optim.NAdam method) (optim.Optimizer method) (optim.RAdam method) (optim.RMSprop method) (optim.Rprop method) (optim.SGD method) StepLR (class in common.pytorch.optim.lr_scheduler) (class in optim.lr_scheduler) strategies() (common.tf.appliance_utils.ExecutionStrategy class method) streaming_confusion_matrix() (in module common.tf.metrics.utils) streaming_rouge_matrix() (in module common.tf.metrics.rouge_score) SummaryCollection (class in common.pytorch.summary_collection) supported_cs_modes (common.pytorch.PyTorchBaseModel.PyTorchBaseModel property) supported_modes (common.pytorch.PyTorchBaseModel.PyTorchBaseModel property) supported_non_cs_modes (common.pytorch.PyTorchBaseModel.PyTorchBaseModel property) supports_conversion() (common.pytorch.model_utils.checkpoint_converters.base_converter.BaseDictionaryConverter class method) supports_mode() (common.pytorch.PyTorchBaseModel.PyTorchBaseModel method)

T

T5ForConditionalGenerationLoss (class in common.pytorch.model_utils.T5ForConditionalGenerationLoss) task_id() (in module vision.pytorch.input.utils) tensor (common.pytorch.summaries.tensor_summary.TensorDescriptor attribute) tensor_summary() (in module common.pytorch.summaries.tensor_summary) TensorDescriptor (class in common.pytorch.summaries.tensor_summary) TensorSummary (class in common.pytorch.summaries.tensor_summary) TensorSummaryReader (class in common.pytorch.summaries.tensor_summary) TFBaseModel (class in common.tf.TFBaseModel) TfRecordsProcessor (class in common.tf.input.TfRecordsProcessor) throughput_dict() (common.pytorch.perf_utils.PerfData method) tiebreak_for_topk() (in module common.pytorch.optim.sparse.utils) to() (common.pytorch.optim.CSOptimizer.CSOptimizer method) to_cpu() (in module common.pytorch.utils) to_dict() (common.pytorch.summaries.tensor_summary.TensorDescriptor method) to_one_hot() (in module vision.pytorch.losses.dice_loss) to_tensor() (in module common.pytorch.utils) total_samples (common.pytorch.perf_utils.PerfData attribute) total_time (common.pytorch.perf_utils.PerfData attribute) train() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) (common.pytorch.pytorch_cs_compiler.PyTorchCSCompiler method) (common.pytorch.pytorch_cs_runner.PyTorchCSRunner method) (common.pytorch.pytorch_dist_runner.PyTorchDistRunner method) (common.pytorch.PyTorchBaseModel.PyTorchBaseModel method)

train_and_eval() (common.pytorch.pytorch_base_cs_runner.PyTorchBaseCSRunner method) (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_dist_runner.PyTorchDistRunner method) train_epoch() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) train_forward() (common.pytorch.pytorch_base_runner.PyTorchBaseRunner method) (common.pytorch.pytorch_cs_appliance.PyTorchCSAppliance method) (common.pytorch.pytorch_runner.PyTorchRunner method) trainable_named_parameters() (common.pytorch.PyTorchBaseModel.PyTorchBaseModel method) (in module common.pytorch.utils) Trainer (class in common.tf.optimizers.Trainer) transform_dataset() (in module common.tf.input.utils) transpose_key_if_2D() (in module common.pytorch.model_utils.checkpoint_converters.gpt2_hf_cs) TripletMarginLoss (class in common.pytorch.layers.TripletMarginLoss) TripletMarginWithDistanceLoss (class in common.pytorch.layers.TripletMarginWithDistanceLoss) trunc_normal_() (in module common.pytorch.model_utils.weight_initializers) truncate_to_idx() (vision.pytorch.input.classification.dataset_factory.VisionSubset method)

U

update() (common.tf.optimizers.LossScale.CSDynamicLossScale method) update_debug_args_from_stack_params() (in module common.tf.appliance_utils) update_input_checkpoint_steps() (in module common.tf.run_utils) update_mask() (common.pytorch.optim.sparse.dynamic.DynamicSparsityOptimizer method) (common.pytorch.optim.sparse.dynamic.GMPSparsityOptimizer method) update_on_device() (common.pytorch.metrics.cb_metric.CBMetric method)

update_on_host() (common.pytorch.metrics.cb_metric.CBMetric method) update_params_from_args() (in module common.tf.run_utils) update_sideband_mode_arg() (in module common.pytorch.run_utils) uses_dynamic_loss_scaling() (common.tf.optimizers.Trainer.Trainer method) uses_loss_scaling() (common.tf.optimizers.Trainer.Trainer method) uses_static_loss_scaling() (common.tf.optimizers.Trainer.Trainer method) utctime (common.pytorch.summaries.tensor_summary.TensorDescriptor property)

V

validate() (common.pytorch.utils.RunConfigParamsValidator method) validate_segments() (common.pytorch.model_utils.checkpoint_converters.base_converter.ConversionRule method) var_names (common.tf.run_utils.GetWeights property) variance_scaling_() (in module common.pytorch.model_utils.weight_initializers) vision.pytorch module vision.pytorch.input module vision.pytorch.input.classification module vision.pytorch.input.classification.dataset_factory module vision.pytorch.input.classification.mixup module vision.pytorch.input.classification.preprocessing module vision.pytorch.input.classification.sampler module vision.pytorch.input.classification.utils module vision.pytorch.input.transforms module

vision.pytorch.input.utils module vision.pytorch.layers module vision.pytorch.layers.BatchChannelNorm module vision.pytorch.layers.ConvNormActBlock module vision.pytorch.layers.GroupInstanceNorm module vision.pytorch.layers.normalizations module vision.pytorch.layers.StandardizedConvolutionLayer module vision.pytorch.layers.utils module vision.pytorch.losses module vision.pytorch.losses.dice_loss module VisionSubset (class in vision.pytorch.input.classification.dataset_factory) visit_structure() (in module common.pytorch.utils)

W

weight_streaming (common.tf.appliance_utils.ExecutionStrategy attribute)

WeightStreaming (common.tf.run_utils.ExecutionMode attribute) wrap_optimizer() (in module common.tf.optimizers.LossScale)