Bump keras from 3.14.0 to 3.15.0

[dependabot]

Bumps keras from 3.14.0 to 3.15.0.

Release notes

Sourced from keras's releases.

v3.15.0

Highlights

• Keras-to-Torch Export: New export_torch enables exporting Keras models to native PyTorch nn.Module format, along with LiteRT (TFLite) export support for the PyTorch backend.
• Sliding Window Attention: Added sliding_window parameter to MultiHeadAttention and GroupedQueryAttention for efficient long-context attention.
• Flash / Fused SDPA: Causal-only MHA/GQA now automatically dispatches to Flash Attention (cuDNN SDPA), and the manual attention path correctly applies causal masking.
• Multi-Optimizer Training: New MultiOptimizer supports assigning different optimizers to sub-networks.
• New Math Operations: Added unique, pinv, matrix_rank, fabs, fmax, fmin, erfc, dsplit, percentile, nanpercentile, sobel_edges, and ssim (structural similarity) to keras.ops.
• Security Hardening: Comprehensive hardening of model reloading against HDF5 exploits, tar/zip traversal attacks, insecure deserialization.

---

New Features and Operations

Multi-Backend Operations

• New NumPy Operations: Added unique, fabs, fmax, fmin, dsplit, erfc, percentile, nanpercentile in keras.ops.numpy.
• New Linear Algebra Operations: Added pinv (pseudo-inverse) and matrix_rank in keras.ops.linalg.
• New Image Operations: Added sobel_edges for edge detection and ssim (structural similarity) in keras.ops.image.
• Negative Axes in Transpose: keras.ops.transpose now supports negative axis values.

Layers and Attention

• Sliding Window Attention: MultiHeadAttention and GroupedQueryAttention layers support the sliding_window parameter for efficient long-sequence processing.
• Flash Attention Engagement: Causal-only attention in MHA/GQA now uses Flash SDPA for significant speedups.
• Fused Bidirectional LSTM/GRU: JAX backend now fuses Bidirectional LSTM into a single cuDNN call; fused bidirectional GRU added for Torch backend.
• CTC Beam Search Decoder: Added CTC beam search decoding for the Torch backend.

Training and Optimizers

• MultiOptimizer: Supports training sub-networks with different optimizers.
• SKLearn Classifier: Added predict_proba method to SKLearnClassifier.

---

Export and Deployment

• Keras-to-Torch Export: Export Keras models to native PyTorch nn.Module via model.export(..., format="torch").
• LiteRT (TFLite) Export for PyTorch: Added LiteRT export support for models using the PyTorch backend.
• LiteRT Compatibility Fix: Fixed LiteRT export for Keras 3 + TF 2.20 + Python 3.13.
• ONNX Export: Support for dict/list inputs in Torch ONNX export; documented static input signature requirement for LiteRT PyTorch export.

---

Distribution and Parallelism

• ModelParallel Improvements: Defined contiguous replica-group data shard ID convention; added distribution information (num_processes, num_model_replicas, data_shard_id).
• Initializer Distribution Layout: Initializers can now handle the distribution layout directly with JAX.
• TF Dataset Distribution: Refactored TF dataset distribution with centralized sharding routing; fixed data distribution for model training in JAX.

---

OpenVINO Backend Support

... (truncated)

Commits

• 9d1bbf9 Add erfinv tests and TensorFlow CPU fallback for half precision (#23086)
• b34af30 Fix pad_sequences string dtype check (np.bytes_ instead of duplicate np.str_)...
• f512e93 Fix wrong parameter names in ops.select and ops.argpartition docstrings (#23091)
• 915ec7e Support native Grouped-Query Attention (GQA) Key/Value head broadcast. (#23081)
• 49d668d Add JAX multi-process distribution tests (#23105)
• b49fde7 Refactor rematerialization logic into Operation and support more modes (#23107)
• e0bfa2d Add to the list of APIs that should not be part of a reloaded model. (#23115)
• 69d7fd6 Suppress spurious 'Skipping nested container' warning on freshly-saved files ...
• 048d142 Delete temp files after loading model from remote path. (#23113)
• 46f5eac Bump GitHub Actions versions to the latest (#23104)
• Additional commits viewable in compare view