From b0dbeabc64caf4b42de124455cd7975af9b1bfbc Mon Sep 17 00:00:00 2001 From: sincekmori Date: Sat, 1 Apr 2023 04:44:00 +0900 Subject: [PATCH] fix typo --- common/checkpointing/snapshot.py | 2 +- core/config/config_load.py | 2 +- core/custom_training_loop.py | 2 +- projects/home/recap/data/config.py | 6 +++--- projects/home/recap/data/dataset.py | 2 +- projects/home/recap/data/generate_random_data.py | 2 +- projects/home/recap/model/config.py | 4 ++-- projects/twhin/data/edges.py | 2 +- 8 files changed, 11 insertions(+), 11 deletions(-) diff --git a/common/checkpointing/snapshot.py b/common/checkpointing/snapshot.py index 2703efd..37b54bb 100644 --- a/common/checkpointing/snapshot.py +++ b/common/checkpointing/snapshot.py @@ -101,7 +101,7 @@ class Snapshot: weight_tensor, ) -> None: """Loads pretrained embedding from the snapshot to the model. - Utilise partial lodaing meachanism from torchsnapshot. + Utilise partial lodaing mechanism from torchsnapshot. Args: embedding_snapshot: Path to the snapshot containing pretrained embeddings (EBC). snapshot_emb_name: Name of the layer in the *snapshot* model, containing the EBC. diff --git a/core/config/config_load.py b/core/config/config_load.py index 709da41..e2fac34 100644 --- a/core/config/config_load.py +++ b/core/config/config_load.py @@ -11,7 +11,7 @@ def load_config_from_yaml(config_type: Type[BaseConfig], yaml_path: str): """Recommend method to load a config file (a yaml file) and parse it. Because we have a shared filesystem the recommended route to running jobs it put modified config - files with the desired parameters somewhere on the filesytem and run jobs pointing to them. + files with the desired parameters somewhere on the filesystem and run jobs pointing to them. """ def _substitute(s): diff --git a/core/custom_training_loop.py b/core/custom_training_loop.py index 0241145..8a82b1c 100644 --- a/core/custom_training_loop.py +++ b/core/custom_training_loop.py @@ -115,7 +115,7 @@ def train( dataset: data iterator for the training set evaluation_iterators: data iterators for the different evaluation sets scheduler: optional learning rate scheduler - output_transform_for_metrics: optional transformation functions to transorm the model + output_transform_for_metrics: optional transformation functions to transform the model output and labels into a format the metrics can understand """ diff --git a/projects/home/recap/data/config.py b/projects/home/recap/data/config.py index 27ef3ed..20d7a44 100644 --- a/projects/home/recap/data/config.py +++ b/projects/home/recap/data/config.py @@ -50,7 +50,7 @@ class DatasetConfig(base_config.BaseConfig): None, description="Number of shards to keep." ) repeat_files: bool = pydantic.Field( - True, description="DEPRICATED. Files are repeated no matter what this is set to." + True, description="DEPRECATED. Files are repeated no matter what this is set to." ) file_batch_size: pydantic.PositiveInt = pydantic.Field(16, description="File batch size") @@ -211,7 +211,7 @@ class Sampler(base_config.BaseConfig): Only use this for quick experimentation. If samplers are useful, we should sample from upstream data generation. - DEPRICATED, DO NOT USE. + DEPRECATED, DO NOT USE. """ name: str @@ -234,7 +234,7 @@ class RecapDataConfig(DatasetConfig): sampler: Sampler = pydantic.Field( None, - description="""DEPRICATED, DO NOT USE. Sampling function for offline experiments.""", + description="""DEPRECATED, DO NOT USE. Sampling function for offline experiments.""", ) @pydantic.root_validator() diff --git a/projects/home/recap/data/dataset.py b/projects/home/recap/data/dataset.py index 3478c68..2312ede 100644 --- a/projects/home/recap/data/dataset.py +++ b/projects/home/recap/data/dataset.py @@ -398,7 +398,7 @@ class RecapDataset(torch.utils.data.IterableDataset): ) else: raise ValueError( - "Must specifiy either `inputs`, `explicit_datetime_inputs`, or `explicit_date_inputs` in data_config" + "Must specify either `inputs`, `explicit_datetime_inputs`, or `explicit_date_inputs` in data_config" ) num_files = len(filenames) diff --git a/projects/home/recap/data/generate_random_data.py b/projects/home/recap/data/generate_random_data.py index 049385a..121e2f1 100644 --- a/projects/home/recap/data/generate_random_data.py +++ b/projects/home/recap/data/generate_random_data.py @@ -9,7 +9,7 @@ from tml.core import config as tml_config_mod import tml.projects.home.recap.config as recap_config_mod flags.DEFINE_string("config_path", None, "Path to hyperparameters for model.") -flags.DEFINE_integer("n_examples", 100, "Numer of examples to generate.") +flags.DEFINE_integer("n_examples", 100, "Number of examples to generate.") FLAGS = flags.FLAGS diff --git a/projects/home/recap/model/config.py b/projects/home/recap/model/config.py index 47d0640..39c2de8 100644 --- a/projects/home/recap/model/config.py +++ b/projects/home/recap/model/config.py @@ -18,7 +18,7 @@ class DropoutConfig(base_config.BaseConfig): class LayerNormConfig(base_config.BaseConfig): - """Configruation for the layer normalization.""" + """Configuration for the layer normalization.""" epsilon: float = pydantic.Field( 1e-3, description="Small float added to variance to avoid dividing by zero." @@ -91,7 +91,7 @@ class ZScoreLogConfig(base_config.BaseConfig): False, description="Option to use batch normalization on the inputs." ) use_renorm: bool = pydantic.Field( - False, description="Option to use batch renormalization for trainig and serving consistency." + False, description="Option to use batch renormalization for training and serving consistency." ) use_bq_stats: bool = pydantic.Field( False, description="Option to load the partitioned json files from BQ as statistics." diff --git a/projects/twhin/data/edges.py b/projects/twhin/data/edges.py index f7864b1..42aebfb 100644 --- a/projects/twhin/data/edges.py +++ b/projects/twhin/data/edges.py @@ -96,7 +96,7 @@ class EdgesDataset(Dataset): Returns a KeyedJaggedTensor used to look up all embeddings. - Note: We treat the lhs and rhs as though they're separate lookups: `len(lenghts) == 2 * bsz * len(tables)`. + Note: We treat the lhs and rhs as though they're separate lookups: `len(lengths) == 2 * bsz * len(tables)`. This differs from the DLRM pattern where we have `len(lengths) = bsz * len(tables)`. For the example above: