tfm.core.train_utils.ExperimentParser
Stay organized with collections
Save and categorize content based on your preferences.
Constructs the Experiment config from Flags or equivalent object.
tfm.core.train_utils.ExperimentParser(
flags_obj
)
Most of the cases, users only need to call the parse()
function:
builder = ExperimentParser(FLAGS)
params = builder.parse()
The advanced users can modify the flow by calling the parse_*() functions
separately.
Methods
base_experiment
View source
base_experiment()
Get the base experiment config from --experiment field.
parse
View source
parse()
Overrall process of constructing Experiment config.
parse_config_file
View source
parse_config_file(
params
)
Override the configs of params from the config_file.
parse_data_service
View source
parse_data_service(
params
)
Override the data service configs of params from flags.
parse_params_override
View source
parse_params_override(
params
)
parse_runtime
View source
parse_runtime(
params
)
Override the runtime configs of params from flags.
Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates. Some content is licensed under the numpy license.
Last updated 2024-02-02 UTC.
[null,null,["Last updated 2024-02-02 UTC."],[],[],null,["# tfm.core.train_utils.ExperimentParser\n\n\u003cbr /\u003e\n\n|-------------------------------------------------------------------------------------------------------------------|\n| [View source on GitHub](https://github.com/tensorflow/models/blob/v2.15.0/official/core/train_utils.py#L300-L374) |\n\nConstructs the Experiment config from Flags or equivalent object. \n\n tfm.core.train_utils.ExperimentParser(\n flags_obj\n )\n\nMost of the cases, users only need to call the `parse()` function: \n\n builder = ExperimentParser(FLAGS)\n params = builder.parse()\n\nThe advanced users can modify the flow by calling the parse_\\*() functions\nseparately.\n\nMethods\n-------\n\n### `base_experiment`\n\n[View source](https://github.com/tensorflow/models/blob/v2.15.0/official/core/train_utils.py#L325-L329) \n\n base_experiment()\n\nGet the base experiment config from --experiment field.\n\n### `parse`\n\n[View source](https://github.com/tensorflow/models/blob/v2.15.0/official/core/train_utils.py#L316-L323) \n\n parse()\n\nOverrall process of constructing Experiment config.\n\n### `parse_config_file`\n\n[View source](https://github.com/tensorflow/models/blob/v2.15.0/official/core/train_utils.py#L331-L336) \n\n parse_config_file(\n params\n )\n\nOverride the configs of params from the config_file.\n\n### `parse_data_service`\n\n[View source](https://github.com/tensorflow/models/blob/v2.15.0/official/core/train_utils.py#L348-L363) \n\n parse_data_service(\n params\n )\n\nOverride the data service configs of params from flags.\n\n### `parse_params_override`\n\n[View source](https://github.com/tensorflow/models/blob/v2.15.0/official/core/train_utils.py#L365-L374) \n\n parse_params_override(\n params\n )\n\n### `parse_runtime`\n\n[View source](https://github.com/tensorflow/models/blob/v2.15.0/official/core/train_utils.py#L338-L346) \n\n parse_runtime(\n params\n )\n\nOverride the runtime configs of params from flags."]]