dependabot[bot] opened a new pull request, #37325: URL: https://github.com/apache/beam/pull/37325
Bumps [keras](https://github.com/keras-team/keras) from 3.12.0 to 3.13.1. <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/keras-team/keras/releases";>keras's releases</a>.</em></p> <blockquote> <h2>v3.13.1</h2> <h3>Bug Fixes & Improvements</h3> <ul> <li><strong>General</strong> <ul> <li>Removed a persistent warning triggered during <code>import keras</code> when using NumPy 2.0 or higher. (<a href="https://redirect.github.com/keras-team/keras/issues/21949";>#21949</a>)</li> </ul> </li> <li><strong>Backends</strong> <ul> <li><strong>JAX:</strong> Fixed an issue where CUDNN flash attention was broken when using JAX versions greater than 0.6.2. (<a href="https://redirect.github.com/keras-team/keras/issues/21970";>#21970</a>)</li> </ul> </li> <li><strong>Export & Serialization</strong> <ul> <li>Resolved a regression in the export pipeline that incorrectly forced batch sizes to be dynamic. The export process now correctly respects static batch sizes when defined. (<a href="https://redirect.github.com/keras-team/keras/issues/21944";>#21944</a>)</li> </ul> </li> </ul> <p><strong>Full Changelog</strong>: <a href="https://github.com/keras-team/keras/compare/v3.13.0...v3.13.1";>https://github.com/keras-team/keras/compare/v3.13.0...v3.13.1</a></p> <h2>v3.13.0</h2> <h2>BREAKING changes</h2> <p>Starting with version 3.13.0, Keras now requires <code>Python 3.11</code> or higher. Please ensure your environment is updated to Python 3.11+ to install the latest version.</p> <h2>Highlights</h2> <h3>LiteRT Export</h3> <p>You can now export Keras models directly to the LiteRT format (formerly TensorFlow Lite) for on-device inference. This changes comes with improvements to input signature handling and export utility documentation. The changes ensure that LiteRT export is only available when TensorFlow is installed, update the export API and documentation, and enhance input signature inference for various model types.</p> <p>Example:</p> <pre lang="python"><code>import keras import numpy as np <h1>1. Define a simple model</h1> <p>model = keras.Sequential([<br /> keras.layers.Input(shape=(10,)),<br /> keras.layers.Dense(10, activation="relu"),<br /> keras.layers.Dense(1, activation="sigmoid")<br /> ])</p> <h1>2. Compile and train (optional, but recommended before export)</h1> <p>model.compile(optimizer="adam", loss="binary_crossentropy")<br /> model.fit(np.random.rand(100, 10), np.random.randint(0, 2, 100), epochs=1)</p> <h1>3. Export the model to LiteRT format</h1> <p>model.export("my_model.tflite", format="litert")</p> <p>print("Model exported successfully to 'my_model.tflite' using LiteRT format.")<br /> </code></pre></p> <h3>GPTQ Quantization</h3> <ul> <li>Introduced <code>keras.quantizers.QuantizationConfig</code> API that allows for customizable weight and activation quantizers, providing greater flexibility in defining quantization schemes.</li> </ul> <!-- raw HTML omitted --> </blockquote> <p>... (truncated)</p> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/keras-team/keras/commit/8914427b7fa9d90f3c476cb2ee65d55d4f808e65";><code>8914427</code></a> Patch release commits for 3.13.1 (<a href="https://redirect.github.com/keras-team/keras/issues/22005";>#22005</a>)</li> <li><a href="https://github.com/keras-team/keras/commit/986ff971d98e216a89fba38d48a337ed09d6dc44";><code>986ff97</code></a> Update release version and comment orbax checkpoint (<a href="https://redirect.github.com/keras-team/keras/issues/21934";>#21934</a>)</li> <li><a href="https://github.com/keras-team/keras/commit/ca23fceb51f7de40dba51908a0a5297f1f425651";><code>ca23fce</code></a> Refactors AbsMaxQuantizer to accept axis in <strong>call</strong> (<a href="https://redirect.github.com/keras-team/keras/issues/21931";>#21931</a>)</li> <li><a href="https://github.com/keras-team/keras/commit/1a9893f04e081ba7f233f59b0669bbde2d83143f";><code>1a9893f</code></a> Adds Serialization Support for QuantizationConfig based quantized models (<a href="https://redirect.github.com/keras-team/keras/issues/21";>#21</a>...</li> <li><a href="https://github.com/keras-team/keras/commit/86bfab4375d17ed41e56dc25ab1fdb6e324377a8";><code>86bfab4</code></a> More OpenVINO Numpy Operations (<a href="https://redirect.github.com/keras-team/keras/issues/21925";>#21925</a>)</li> <li><a href="https://github.com/keras-team/keras/commit/f48f4805fd462893f68e5f6cc632d320d4198a52";><code>f48f480</code></a> Add adaptive pooling (1D, 2D, 3D) support across JAX, NumPy, TensorFlow, and ...</li> <li><a href="https://github.com/keras-team/keras/commit/0771c8044a5c183e3cd8c1c4e8756a607579d4a0";><code>0771c80</code></a> Fix ops.tile shape inference issue on TensorFlow backend (<a href="https://redirect.github.com/keras-team/keras/issues/21860";>#21860</a>)</li> <li><a href="https://github.com/keras-team/keras/commit/024c96db1705f2d4fc0c03fde9158e33554b5304";><code>024c96d</code></a> Extended fix OOM Issue <a href="https://redirect.github.com/keras-team/keras/issues/21634";>#21634</a> on Keras side (<a href="https://redirect.github.com/keras-team/keras/issues/21755";>#21755</a>)</li> <li><a href="https://github.com/keras-team/keras/commit/71f499735f543871d296474c48f56a2e9758a73c";><code>71f4997</code></a> Introduces QuantizationConfig for fine-grained quantization control (<a href="https://redirect.github.com/keras-team/keras/issues/21896";>#21896</a>)</li> <li><a href="https://github.com/keras-team/keras/commit/3989d64176363f69a593554d5950ccde84a510b7";><code>3989d64</code></a> Fix fake quant gradient output shape and use <code>jax.grad</code> for tests. (<a href="https://redirect.github.com/keras-team/keras/issues/21927";>#21927</a>)</li> <li>Additional commits viewable in <a href="https://github.com/keras-team/keras/compare/v3.12.0...v3.13.1";>compare view</a></li> </ul> </details> <br /> [](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores) Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting `@dependabot rebase`. [//]: # (dependabot-automerge-start) [//]: # (dependabot-automerge-end) --- <details> <summary>Dependabot commands and options</summary> <br /> You can trigger Dependabot actions by commenting on this PR: - `@dependabot rebase` will rebase this PR - `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it - `@dependabot merge` will merge this PR after your CI passes on it - `@dependabot squash and merge` will squash and merge this PR after your CI passes on it - `@dependabot cancel merge` will cancel a previously requested merge and block automerging - `@dependabot reopen` will reopen this PR if it is closed - `@dependabot close` will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually - `@dependabot show <dependency name> ignore conditions` will show all of the ignore conditions of the specified dependency - `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself) - `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself) - `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself) You can disable automated security fix PRs for this repo from the [Security Alerts page](https://github.com/apache/beam/network/alerts). </details> -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
