wandb 0.17.8rc1__py3-none-any.whl → 0.17.9__py3-none-any.whl

package_readme.md

@@ -4,92 +4,86 @@
 
 # Weights and Biases [![PyPI](https://img.shields.io/pypi/v/wandb)](https://pypi.python.org/pypi/wandb) [![Conda (channel only)](https://img.shields.io/conda/vn/conda-forge/wandb)](https://anaconda.org/conda-forge/wandb) [![CircleCI](https://img.shields.io/circleci/build/github/wandb/wandb/main)](https://circleci.com/gh/wandb/wandb) [![Codecov](https://img.shields.io/codecov/c/gh/wandb/wandb)](https://codecov.io/gh/wandb/wandb)
 
-Use W&B to build better models faster. Track and visualize all the pieces of your machine learning pipeline, from datasets to production models.
+Use W&B to build better models faster. Track and visualize all the pieces of your machine learning pipeline, from datasets to production machine learning models. Get started with W&B today, [sign up for an account!](https://wandb.com?utm_source=github&utm_medium=code&utm_campaign=wandb&utm_content=readme)
 
-- Quickly identify model regressions. Use W&B to visualize results in real time, all in a central dashboard.
-- Focus on the interesting ML. Spend less time manually tracking results in spreadsheets and text files.
-- Capture dataset versions with W&B Artifacts to identify how changing data affects your resulting models.
-- Reproduce any model, with saved code, hyperparameters, launch commands, input data, and resulting model weights.
 
-[Sign up for a free account →](https://wandb.ai/login?signup=true)
 
-## Features
+See the [W&B Developer Guide](https://docs.wandb.ai/?utm_source=github&utm_medium=code&utm_campaign=wandb&utm_content=documentation) and [API Reference Guide](https://docs.wandb.ai/ref?utm_source=github&utm_medium=code&utm_campaign=wandb&utm_content=documentation) for a full technical description of the W&B platform.
 
--   Store hyper-parameters used in a training run
--   Search, compare, and visualize training runs
--   Analyze system usage metrics alongside runs
--   Collaborate with team members
--   Replicate historic results
--   Run parameter sweeps
--   Keep records of experiments available forever
+ 
 
-[Documentation →](https://docs.wandb.com)
+# Quickstart
 
-## Quickstart
+Get started with W&B in four steps:
+
+1. First, sign up for a [W&B account](https://wandb.ai/login?utm_source=github&utm_medium=code&utm_campaign=wandb&utm_content=quickstart).
+
+2. Second, install the W&B SDK with [pip](https://pip.pypa.io/en/stable/). Navigate to your terminal and type the following command:
 
 ```shell
 pip install wandb
 ```
 
-In your training script:
+3. Third, log into W&B:
 
 ```python
-import wandb
+wandb.login()
+```
 
-# Your custom arguments defined here
-args = ...
+4. Use the example code snippet below as a template to integrate W&B to your Python script:
 
-wandb.init(config=args, project="my-project")
-wandb.config["more"] = "custom"
+```python
+import wandb
 
+# Start a W&B Run with wandb.init
+run = wandb.init(project="my_first_project")
 
-def training_loop():
-    while True:
-        # Do some machine learning
-        epoch, loss, val_loss = ...
-        # Framework agnostic / custom metrics
-        wandb.log({"epoch": epoch, "loss": loss, "val_loss": val_loss})
-```
+# Save model inputs and hyperparameters in a wandb.config object
+config = run.config
+config.learning_rate = 0.01
 
-If you're already using Tensorboard or [TensorboardX](https://github.com/lanpa/tensorboardX), you can integrate with one line:
+# Model training code here ...
 
-```python
-wandb.init(sync_tensorboard=True)
+# Log metrics over time to visualize performance with wandb.log
+for i in range(10):
+    run.log({"loss": ...})
+
+# Mark the run as finished, and finish uploading all data
+run.finish()
 ```
 
-## Running your script
+For example, if the preceding code was stored in a script called train.py:
 
-Run `wandb login` from your terminal to signup or authenticate your machine (we store your api key in ~/.netrc). You can also set the `WANDB_API_KEY` environment variable with a key from your [settings](https://app.wandb.ai/settings).
+```shell
+python train.py
+```
 
-Run your script with `python my_script.py` and all metadata will be synced to the cloud. You will see a url in your terminal logs when your script starts and finishes. Data is staged locally in a directory named _wandb_ relative to your script. If you want to test your script without syncing to the cloud you can set the environment variable `WANDB_MODE=dryrun`.
+You will see a URL in your terminal logs when your script starts and finishes. Data is staged locally in a directory named _wandb_ relative to your script. Navigate to the W&B App to view a dashboard of your first W&B Experiment. Use the W&B App to compare multiple experiments in a unified place, dive into the results of a single run, and much more!
 
-If you are using [docker](https://docker.com) to run your code, we provide a wrapper command `wandb docker` that mounts your current directory, sets environment variables, and ensures the wandb library is installed. Training your models in docker gives you the ability to restore the exact code and environment with the `wandb restore` command.
+ 
 
-## Web Interface
+# Integrations
 
-[Sign up for a free account →](https://wandb.com)
-[![Watch the video](https://i.imgur.com/PW0Ejlc.png)](https://youtu.be/EeqhOSvNX-A)
-[Introduction video →](https://youtu.be/EeqhOSvNX-A)
+Use your favorite framework with W&B. W&B integrations make it fast and easy to set up experiment tracking and data versioning inside existing projects. For more information on how to integrate W&B with the framework of your choice, see [W&B Integrations](https://docs.wandb.ai/guides/integrations) in the W&B Developer Guide.
 
-## Detailed Usage
+ 
 
-Framework specific and detailed usage can be found in our [documentation](http://docs.wandb.com/).
+# Contribution guidelines
+Weights & Biases ❤️ open source, and we welcome contributions from the community! See the [Contribution guide](https://github.com/wandb/wandb/blob/main/CONTRIBUTING.md) for more information on the development workflow and the internals of the wandb library. For wandb bugs and feature requests, visit [GitHub Issues](https://github.com/wandb/wandb/issues) or contact support@wandb.com.
 
-## Testing
+ 
 
-To run basic test use `make test`.  More detailed information can be found at CONTRIBUTING.md.
+# Academic Researchers
+Reach out to W&B Support at support@wandb.com to get a [free academic license](https://www.wandb.com/academic) for you and your research group.
 
-We use [circleci](https://circleci.com) for CI.
+ 
 
-# Academic Researchers
-If you'd like a free academic account for your research group, [reach out to us →](https://www.wandb.com/academic)
+# W&B Community
 
-We make it easy to cite W&B in your published paper. [Learn more →](https://www.wandb.com/academic)
-[![](https://i.imgur.com/loKLiez.png)](https://www.wandb.com/academic)
+Be a part of the growing W&B Community and interact with the W&B team in our [Discord](https://wandb.me/discord). Stay connected with the latest ML updates and tutorials with [W&B Fully Connected](https://wandb.ai/fully-connected).
 
-## Community
-Got questions, feedback or want to join a community of ML engineers working on exciting projects?
+ 
 
-<a href="https://bit.ly/wb-slack"><img src="https://svgshare.com/i/M93.svg" alt="slack" width="55"/></a> Join our [slack](https://bit.ly/wb-slack) community.
+# License
 
-[![Twitter](https://img.shields.io/twitter/follow/weights_biases?style=social)](https://twitter.com/weights_biases) Follow us on [Twitter](https://twitter.com/weights_biases).
+[MIT License](https://github.com/wandb/wandb/blob/main/LICENSE)

wandb/__init__.py

@@ -8,7 +8,7 @@ For scripts and interactive notebooks, see https://github.com/wandb/examples.
 
 For reference documentation, see https://docs.wandb.com/ref/python.
 """
-__version__ = "0.17.8rc1"
+__version__ = "0.17.9"
 
 from typing import Optional
 
@@ -76,8 +76,7 @@ from wandb.data_types import JoinedTable
 
 from wandb.wandb_agent import agent
 
-# from wandb.core import *
-from wandb.viz import visualize
+from wandb.sdk.lib.viz import visualize
 from wandb import plot
 from wandb.integration.sagemaker import sagemaker_auth
 from wandb.sdk.internal import profiler
@@ -199,14 +198,18 @@ if wandb_sdk.lib.ipython.in_notebook():
 from .analytics import Sentry as _Sentry
 
 if "dev" in __version__:
+    import wandb.env
     import os
 
     # disable error reporting in dev versions for the python client
-    os.environ["WANDB_ERROR_REPORTING"] = os.environ.get(
-        "WANDB_ERROR_REPORTING", "false"
+    os.environ[wandb.env.ERROR_REPORTING] = os.environ.get(
+        wandb.env.ERROR_REPORTING, "false"
     )
+
     # turn on wandb-core for dev versions
-    os.environ["WANDB__REQUIRE_CORE"] = os.environ.get("WANDB__REQUIRE_CORE", "true")
+    if not wandb.env.is_require_legacy_service():
+        os.environ[wandb.env._REQUIRE_CORE] = os.environ.get(
+            wandb.env._REQUIRE_CORE, "true")
 
 _sentry = _Sentry()
 _sentry.setup()
@@ -215,6 +218,7 @@ _sentry.setup()
 __all__ = (
     "__version__",
     "init",
+    "finish",
     "setup",
     "save",
     "sweep",
@@ -237,6 +241,8 @@ __all__ = (
     "Molecule",
     "Histogram",
     "ArtifactTTL",
+    "log_artifact",
+    "use_artifact",
     "log_model",
     "use_model",
     "link_model",

wandb/__init__.pyi

@@ -12,7 +12,9 @@ For reference documentation, see https://docs.wandb.com/ref/python.
 __all__ = (
     "__version__",
     "init",
+    "finish",
     "setup",
+    "login",
     "save",
     "sweep",
     "controller",
@@ -33,6 +35,8 @@ __all__ = (
     "Molecule",
     "Histogram",
     "ArtifactTTL",
+    "log_artifact",
+    "use_artifact",
     "log_model",
     "use_model",
     "link_model",
@@ -48,7 +52,7 @@ __all__ = (
 )
 
 import os
-from typing import Any, Callable, Dict, List, Optional, Sequence, Union
+from typing import Any, Callable, Dict, List, Literal, Optional, Sequence, Union
 
 from wandb.analytics import Sentry as _Sentry
 from wandb.apis import InternalApi, PublicApi
@@ -75,7 +79,7 @@ from wandb.sdk.wandb_run import Run
 from wandb.sdk.wandb_setup import _WandbSetup
 from wandb.wandb_controller import _WandbController
 
-__version__: str = "0.17.8rc1"
+__version__: str = "0.17.9"
 
 run: Optional[Run] = None
 config = wandb_config.Config
@@ -393,6 +397,54 @@ def init(
     """
     ...
 
+def finish(exit_code: Optional[int] = None, quiet: Optional[bool] = None) -> None:
+    """Mark a run as finished, and finish uploading all data.
+
+    This is used when creating multiple runs in the same process.
+    We automatically call this method when your script exits.
+
+    Arguments:
+        exit_code: Set to something other than 0 to mark a run as failed
+    quiet: Set to true to minimize log output
+    """
+    ...
+
+def login(
+    anonymous: Optional[Literal["must", "allow", "never"]] = None,
+    key: Optional[str] = None,
+    relogin: Optional[bool] = None,
+    host: Optional[str] = None,
+    force: Optional[bool] = None,
+    timeout: Optional[int] = None,
+    verify: bool = False,
+) -> bool:
+    """Set up W&B login credentials.
+
+    By default, this will only store credentials locally without
+    verifying them with the W&B server. To verify credentials, pass
+    `verify=True`.
+
+    Arguments:
+        anonymous: (string, optional) Can be "must", "allow", or "never".
+            If set to "must", always log a user in anonymously. If set to
+            "allow", only create an anonymous user if the user
+            isn't already logged in. If set to "never", never log a
+            user anonymously. Default set to "never".
+        relogin: (bool, optional) If true, will re-prompt for API key.
+        host: (string, optional) The host to connect to.
+        force: (bool, optional) If true, will force a relogin.
+        timeout: (int, optional) Number of seconds to wait for user input.
+        verify: (bool) Verify the credentials with the W&B server.
+
+    Returns:
+        bool: if key is configured
+
+    Raises:
+        AuthenticationError - if api_key fails verification with the server
+    UsageError - if api_key cannot be configured and no tty
+    """
+    ...
+
 def log(
     data: Dict[str, Any],
     step: Optional[int] = None,
@@ -805,6 +857,64 @@ def define_metric(
     """
     ...
 
+def log_artifact(
+    artifact_or_path: Union[Artifact, StrPath],
+    name: Optional[str] = None,
+    type: Optional[str] = None,
+    aliases: Optional[List[str]] = None,
+) -> Artifact:
+    """Declare an artifact as an output of a run.
+
+    Arguments:
+        artifact_or_path: (str or Artifact) A path to the contents of this artifact,
+            can be in the following forms:
+                - `/local/directory`
+                - `/local/directory/file.txt`
+                - `s3://bucket/path`
+            You can also pass an Artifact object created by calling
+            `wandb.Artifact`.
+        name: (str, optional) An artifact name. Valid names can be in the following forms:
+                - name:version
+                - name:alias
+                - digest
+            This will default to the basename of the path prepended with the current
+            run id  if not specified.
+        type: (str) The type of artifact to log, examples include `dataset`, `model`
+        aliases: (list, optional) Aliases to apply to this artifact,
+            defaults to `["latest"]`
+
+    Returns:
+    An `Artifact` object.
+    """
+    ...
+
+def use_artifact(
+    artifact_or_name: Union[str, Artifact],
+    type: Optional[str] = None,
+    aliases: Optional[List[str]] = None,
+    use_as: Optional[str] = None,
+) -> Artifact:
+    """Declare an artifact as an input to a run.
+
+    Call `download` or `file` on the returned object to get the contents locally.
+
+    Arguments:
+        artifact_or_name: (str or Artifact) An artifact name.
+            May be prefixed with entity/project/. Valid names
+            can be in the following forms:
+                - name:version
+                - name:alias
+            You can also pass an Artifact object created by calling `wandb.Artifact`
+        type: (str, optional) The type of artifact to use.
+        aliases: (list, optional) Aliases to apply to this artifact
+        use_as: (string, optional) Optional string indicating what purpose the artifact was used with.
+                                   Will be shown in UI.
+
+    Returns:
+    An `Artifact` object.
+    """
+    ...
+
 def log_model(
     path: StrPath,
     name: Optional[str] = None,

wandb/data_types.py

@@ -1420,6 +1420,7 @@ class Graph(Media):
 
     @classmethod
     def from_keras(cls, model):
+        # TODO: his method requires a refactor to work with the keras 3.
         graph = cls()
         # Shamelessly copied (then modified) from keras/keras/utils/layer_utils.py
         sequential_like = cls._is_sequential(model)

wandb/env.py

@@ -91,6 +91,7 @@ LAUNCH_QUEUE_NAME = "WANDB_LAUNCH_QUEUE_NAME"
 LAUNCH_QUEUE_ENTITY = "WANDB_LAUNCH_QUEUE_ENTITY"
 LAUNCH_TRACE_ID = "WANDB_LAUNCH_TRACE_ID"
 _REQUIRE_CORE = "WANDB__REQUIRE_CORE"
+_REQUIRE_LEGACY_SERVICE = "WANDB__REQUIRE_LEGACY_SERVICE"
 
 # For testing, to be removed in future version
 USE_V1_ARTIFACTS = "_WANDB_USE_V1_ARTIFACTS"
@@ -154,9 +155,21 @@ def _env_as_bool(
 
 
 def is_require_core(env: Optional[Env] = None) -> bool:
+    """Return whether wandb.require("core") was used.
+
+    Note that this may contradict wandb.require("legacy-service").
+    """
     return _env_as_bool(_REQUIRE_CORE, default="False", env=env)
 
 
+def is_require_legacy_service(env: Optional[Env] = None) -> bool:
+    """Return whether wandb.require("legacy-service") was used.
+
+    Note that this may contradict wandb.require("core").
+    """
+    return _env_as_bool(_REQUIRE_LEGACY_SERVICE, default="False", env=env)
+
+
 def is_debug(default: Optional[str] = None, env: Optional[Env] = None) -> bool:
     return _env_as_bool(DEBUG, default=default, env=env)
 

wandb/integration/keras/__init__.py

@@ -1,7 +1,4 @@
-"""Tools for integrating `wandb` with [`Keras`](https://keras.io/).
-
-Keras is a deep learning API for [`TensorFlow`](https://www.tensorflow.org/).
-"""
+"""Tools for integrating `wandb` with [`Keras`](https://keras.io/)."""
 
 __all__ = (
     "WandbCallback",
@@ -11,4 +8,4 @@ __all__ = (
 )
 
 from .callbacks import WandbEvalCallback, WandbMetricsLogger, WandbModelCheckpoint
-from .keras import WandbCallback  # todo: legacy callback to be deprecated
+from .keras import WandbCallback  # TODO: legacy callback to be deprecated

wandb/integration/keras/callbacks/metrics_logger.py

@@ -2,7 +2,7 @@ import sys
 from typing import Any, Dict, Optional, Union
 
 import tensorflow as tf  # type: ignore
-from tensorflow.keras import callbacks  # type: ignore
+from tensorflow.keras import callbacks
 
 import wandb
 from wandb.integration.keras.keras import patch_tf_keras
@@ -86,14 +86,20 @@ class WandbMetricsLogger(callbacks.Callback):
             wandb.define_metric("epoch/*", step_metric="epoch/epoch")
 
     def _get_lr(self) -> Union[float, None]:
-        if isinstance(self.model.optimizer.learning_rate, tf.Variable):
+        if isinstance(
+            self.model.optimizer.learning_rate,
+            (tf.Variable, tf.Tensor),
+        ) or (
+            hasattr(self.model.optimizer.learning_rate, "shape")
+            and self.model.optimizer.learning_rate.shape == ()
+        ):
             return float(self.model.optimizer.learning_rate.numpy().item())
         try:
             return float(
                 self.model.optimizer.learning_rate(step=self.global_step).numpy().item()
             )
-        except Exception:
-            wandb.termerror("Unable to log learning rate.", repeat=False)
+        except Exception as e:
+            wandb.termerror(f"Unable to log learning rate: {e}", repeat=False)
             return None
 
     def on_epoch_end(self, epoch: int, logs: Optional[Dict[str, Any]] = None) -> None:

wandb/integration/keras/callbacks/model_checkpoint.py

@@ -71,9 +71,6 @@ class WandbModelCheckpoint(callbacks.ModelCheckpoint):
             Note that when monitoring validation metrics such as `val_acc` or `val_loss`,
             save_freq must be set to "epoch" as those metrics are only available at the
             end of an epoch.
-        options: (Optional[str]) Optional `tf.train.CheckpointOptions` object if
-            `save_weights_only` is true or optional `tf.saved_model.SaveOptions`
-            object if `save_weights_only` is false.
         initial_value_threshold: (Optional[float]) Floating point initial "best" value of the metric
             to be monitored.
     """
@@ -87,7 +84,6 @@ class WandbModelCheckpoint(callbacks.ModelCheckpoint):
         save_weights_only: bool = False,
         mode: Mode = "auto",
         save_freq: Union[SaveStrategy, int] = "epoch",
-        options: Optional[str] = None,
         initial_value_threshold: Optional[float] = None,
         **kwargs: Any,
     ) -> None:
@@ -99,7 +95,6 @@ class WandbModelCheckpoint(callbacks.ModelCheckpoint):
             save_weights_only=save_weights_only,
             mode=mode,
             save_freq=save_freq,
-            options=options,
             initial_value_threshold=initial_value_threshold,
             **kwargs,
         )

wandb/integration/keras/keras.py

@@ -417,6 +417,17 @@ class WandbCallback(tf.keras.callbacks.Callback):
     ):
         if wandb.run is None:
             raise wandb.Error("You must call wandb.init() before WandbCallback()")
+
+        deprecate(
+            field_name=Deprecated.keras_callback,
+            warning_message=(
+                "WandbCallback is deprecated and will be removed in a future release. "
+                "Please use the WandbMetricsLogger, WandbModelCheckpoint, and WandbEvalCallback "
+                "callbacks instead. "
+                "See https://docs.wandb.ai/guides/integrations/keras for more information."
+            ),
+        )
+
         with wandb.wandb_lib.telemetry.context(run=wandb.run) as tel:
             tel.feature.keras = True
         self.validation_data = None
@@ -552,7 +563,7 @@ class WandbCallback(tf.keras.callbacks.Callback):
         self.params = params
 
     def set_model(self, model):
-        self.model = model
+        super().set_model(model)
         if self.input_type == "auto" and len(model.inputs) == 1:
             self.input_type = wandb.util.guess_data_type(
                 model.inputs[0].shape, risky=True

wandb/integration/openai/fine_tuning.py

@@ -37,23 +37,23 @@ from openai.types.fine_tuning.fine_tuning_job import (  # noqa: E402
 np = util.get_module(
     name="numpy",
     required="`numpy` not installed >> This integration requires numpy!  To fix, please `pip install numpy`",
-    lazy="False",
+    lazy=False,
 )
 
 pd = util.get_module(
     name="pandas",
     required="`pandas` not installed >> This integration requires pandas!  To fix, please `pip install pandas`",
-    lazy="False",
+    lazy=False,
 )
 
 
 class WandbLogger:
     """Log OpenAI fine-tunes to [Weights & Biases](https://wandb.me/openai-docs)."""
 
-    _wandb_api: wandb.Api = None
+    _wandb_api: Optional[wandb.Api] = None
     _logged_in: bool = False
-    openai_client: OpenAI = None
-    _run: Run = None
+    openai_client: Optional[OpenAI] = None
+    _run: Optional[Run] = None
 
     @classmethod
     def sync(

wandb/integration/tensorboard/log.py

@@ -6,7 +6,7 @@ from typing import TYPE_CHECKING, Any, Dict, List, Optional, Union
 import wandb
 import wandb.util
 from wandb.sdk.lib import telemetry
-from wandb.viz import custom_chart
+from wandb.sdk.lib.viz import custom_chart
 
 if TYPE_CHECKING:
     import numpy as np