There will be 2 Keras versions: – The new one, distributed as `keras`, fully multi-backend
– The legacy, TF-only one, distributed as `tf-keras` tf.keras will be the new one by default, but can be set to the old one if you need.
OPEN SOURCE
-
Keras: Two versions, a new multi-backend and a legacy TensorFlow
By
–
-
Keras Multi-Backend Support in TensorFlow with Backward Compatibility
By
–
In the future, importing Keras from TF (`tf.keras`) will give you multi-backend Keras. However we will provide a way to revert back to the legacy tf-only Keras (exact same setup as before), so as to enable folks to upgrade their TF version without worrying about minor breakages.
-
OpenAI Cookbook Gets New Design for Better Discovery
By
–
New skin for the OpenAI cookbook, our open-source collection of examples and guides for building with the OpenAI API. Easier to use and discover than ever before: https://
cookbook.openai.com Great work led by @simonpfish
. -
BTLM-3B Outperforms Larger 7B Models in Long Context
By
–
The paper explores long context performance in detail. We found BTLM-3B-8K outperforms other 7B-8K models despite being trained with less than a fifth the pretraining compute and less than half the size.
-
BTLM: World’s Most Accurate 3B Parameter Open Source Model
By
–
BTLM is the world’s most accurate 3B parameter model, with performance that rivals many 7B models. It’s fully open source and has been downloaded >1M times.
-
BTLM-3B-8K: Distilling SOTA LLM Training Recipe
By
–
We just dropped the BTLM-3B-8K paper on arXiv! It distills our recipe for training SOTA LLMs:
– Extensively deduplicated dataset (SlimPajama)
– Hyperparameter search using muP
– Variable sequence length training + ALiBi
– Aggressive LR decay https://
arxiv.org/abs/2309.11568 -
Running AI Models Directly in Browser with Transformers.js
By
–
Use Hugging Chat, but run the model directly in your browser with transformers.js
-
PyDataset in Keras Core: New Sequence Class for Data Loading
By
–
There's a new Sequence class in Keras Core, it's called PyDataset. It's basically the same functionality (in particular parallel loading), no major improvement. I would still recommend using http://
tf.data for anything performance critical. -
Keras Core Adds Sparse Tensor Support for Recommender Systems
By
–
Keras Core now has support for sparse tensors (with the TF backend only for now — we'll add JAX next). This makes it possible to use Keras Core to build recommender systems.
-
Data Validation for PySpark Applications Using Pandera
By
–
Data Validation for PySpark Applications using Pandera: New features and concepts. https://
kdnuggets.com/2023/08/data-v
alidation-pyspark-applications-pandera.html?utm_source=dlvr.it&utm_medium=twitter&utm_campaign=data-validation-for-pyspark-applications-using-pandera
…