|---| | [ View source on GitHub](https://github.com/tensorflow/models/blob/v2.20.0/official/core/config_definitions.py#L139-L207) | High-level configurations for Runtime. Inherits From: [`Config`](https://www.tensorflow.org/api_docs/python/tfm/hyperparams/Config), [`ParamsDict`](https://www.tensorflow.org/api_docs/python/tfm/hyperparams/ParamsDict) #### View aliases **Main aliases** [`tfm.core.config_definitions.RuntimeConfig`](https://www.tensorflow.org/api_docs/python/tfm/core/base_task/RuntimeConfig)
tfm.core.base_task.RuntimeConfig( default_params: dataclasses.InitVar[Optional[Mapping[str, Any]]] = None, restrictions: dataclasses.InitVar[Optional[List[str]]] = None, distribution_strategy: str = 'mirrored', enable_xla: bool = False, gpu_thread_mode: Optional[str] = None, dataset_num_private_threads: Optional[int] = None, per_gpu_thread_count: int = 0, tpu: Optional[str] = None, num_gpus: int = 0, worker_hosts: Optional[str] = None, task_index: int = -1, all_reduce_alg: Optional[str] = None, num_packs: int = 1, mixed_precision_dtype: Optional[str] = None, loss_scale: Optional[Union[str, float]] = None, run_eagerly: bool = False, batchnorm_spatial_persistent: bool = False, tpu_enable_xla_dynamic_padder: Optional[bool] = None, num_cores_per_replica: int = 1, default_shard_dim: int = -1, use_tpu_mp_strategy: bool = False ) These include parameters that are not directly related to the experiment, e.g. directories, accelerator type, etc.


| ## Args || |---|---| | `default_params` | a Python dict or another ParamsDict object including the default parameters to initialize. | | `restrictions` | a list of strings, which define a list of restrictions to ensure the consistency of different parameters internally. Each restriction string is defined as a binary relation with a set of operators, including {'==', '!=', '\<', '\<=', '\>', '\>='}. |



| ## Attributes || |---|---| | `distribution_strategy` | e.g. 'mirrored', 'tpu', etc. | | `enable_xla` | Whether or not to enable XLA. | | `per_gpu_thread_count` | thread count per GPU. | | `gpu_thread_mode` | Whether and how the GPU device uses its own threadpool. | | `dataset_num_private_threads` | Number of threads for a private threadpool created for all datasets computation. | | `tpu` | The address of the TPU to use, if any. | | `num_gpus` | The number of GPUs to use, if any. | | `worker_hosts` | comma-separated list of worker ip:port pairs for running multi-worker models with DistributionStrategy. | | `task_index` | If multi-worker training, the task index of this worker. | | `all_reduce_alg` | Defines the algorithm for performing all-reduce. | | `num_packs` | Sets `num_packs` in the cross device ops used in MirroredStrategy. For details, see tf.distribute.NcclAllReduce. | | `mixed_precision_dtype` | dtype of mixed precision policy. It can be 'float32', 'float16', or 'bfloat16'. | | `loss_scale` | The type of loss scale, or 'float' value. This is used when setting the mixed precision policy. | | `run_eagerly` | Whether or not to run the experiment eagerly. | | `batchnorm_spatial_persistent` | Whether or not to enable the spatial persistent mode for CuDNN batch norm kernel for improved GPU performance. | | `BUILDER` |

| | `default_params` | Dataclass field | | `restrictions` | Dataclass field | | `tpu_enable_xla_dynamic_padder` | Dataclass field | | `num_cores_per_replica` | Dataclass field | | `default_shard_dim` | Dataclass field | | `use_tpu_mp_strategy` | Dataclass field |
## Methods ### `as_dict` [View source](https://github.com/tensorflow/models/blob/v2.20.0/official/modeling/hyperparams/base_config.py#L315-L324) as_dict() Returns a dict representation of params_dict.ParamsDict. For the nested params_dict.ParamsDict, a nested dict will be returned. ### `from_args` [View source](https://github.com/tensorflow/models/blob/v2.20.0/official/modeling/hyperparams/base_config.py#L349-L358) @classmethod from_args( *args, **kwargs ) Builds a config from the given list of arguments. ### `from_json` [View source](https://github.com/tensorflow/models/blob/v2.20.0/official/modeling/hyperparams/base_config.py#L344-L347) @classmethod from_json( file_path: str ) Wrapper for `from_yaml`. ### `from_yaml` [View source](https://github.com/tensorflow/models/blob/v2.20.0/official/modeling/hyperparams/base_config.py#L335-L342) @classmethod from_yaml( file_path: str ) ### `get` [View source](https://github.com/tensorflow/models/blob/v2.20.0/official/modeling/hyperparams/params_dict.py#L142-L144) get( key, value=None ) Accesses through built-in dictionary get method. ### `lock` [View source](https://github.com/tensorflow/models/blob/v2.20.0/official/modeling/hyperparams/params_dict.py#L204-L206) lock() Makes the ParamsDict immutable. ### `model_parallelism` [View source](https://github.com/tensorflow/models/blob/v2.20.0/official/core/config_definitions.py#L204-L207) model_parallelism() ### `override` [View source](https://github.com/tensorflow/models/blob/v2.20.0/official/modeling/hyperparams/params_dict.py#L165-L181) override( override_params, is_strict=True ) Override the ParamsDict with a set of given params.


| Args || |---|---| | `override_params` | a dict or a ParamsDict specifying the parameters to be overridden. | | `is_strict` | a boolean specifying whether override is strict or not. If True, keys in `override_params` must be present in the ParamsDict. If False, keys in `override_params` can be different from what is currently defined in the ParamsDict. In this case, the ParamsDict will be extended to include the new keys. |
### `replace` [View source](https://github.com/tensorflow/models/blob/v2.20.0/official/modeling/hyperparams/base_config.py#L326-L333) replace( **kwargs ) Overrides/returns a unlocked copy with the current config unchanged. ### `validate` [View source](https://github.com/tensorflow/models/blob/v2.20.0/official/modeling/hyperparams/params_dict.py#L222-L329) validate() Validate the parameters consistency based on the restrictions. This method validates the internal consistency using the pre-defined list of restrictions. A restriction is defined as a string which specifies a binary operation. The supported binary operations are {'==', '!=', '\<', '\<=', '\>', '\>='}. Note that the meaning of these operators are consistent with the underlying Python immplementation. Users should make sure the define restrictions on their type make sense. For example, for a ParamsDict like the following a: a1: 1 a2: 2 b: bb: bb1: 10 bb2: 20 ccc: a1: 1 a3: 3 one can define two restrictions like this \['a.a1 == b.ccc.a1', 'a.a2 \<= b.bb.bb2'\]

| What it enforces are || |---|---| |
- a.a1 = 1 == b.ccc.a1 = 1 - a.a2 = 2 \<= b.bb.bb2 = 20 ||



| Raises || |---|---| | `KeyError` | if any of the following happens (1) any of parameters in any of restrictions is not defined in ParamsDict, (2) any inconsistency violating the restriction is found. | | `ValueError` | if the restriction defined in the string is not supported. |
### `__contains__` [View source](https://github.com/tensorflow/models/blob/v2.20.0/official/modeling/hyperparams/params_dict.py#L138-L140) __contains__( key ) Implements the membership test operator. ### `__eq__` __eq__( other ) Return self==value.


| ## Class Variables || |---|---| | IMMUTABLE_TYPES | `(, , , , )` | | RESERVED_ATTR | `['_locked', '_restrictions']` | | SEQUENCE_TYPES | `(, )` | | all_reduce_alg | `None` | | batchnorm_spatial_persistent | `False` | | dataset_num_private_threads | `None` | | default_params | `None` | | default_shard_dim | `-1` | | distribution_strategy | `'mirrored'` | | enable_xla | `False` | | gpu_thread_mode | `None` | | loss_scale | `None` | | mixed_precision_dtype | `None` | | num_cores_per_replica | `1` | | num_gpus | `0` | | num_packs | `1` | | per_gpu_thread_count | `0` | | restrictions | `None` | | run_eagerly | `False` | | task_index | `-1` | | tpu | `None` | | tpu_enable_xla_dynamic_padder | `None` | | use_tpu_mp_strategy | `False` | | worker_hosts | `None` |