keras-nlp

keras-nlp copied to clipboard

tf2.16.1 isn't compatibale with tensorflow text 2.10

4

**Describe the bug** import tensorflow_text --------------------------------------------------------------------------- ImportError Traceback (most recent call last) ~\AppData\Local\Temp\ipykernel_17348\101723867.py in ----> 1 import tensorflow_text C:\ProgramData\Anaconda3\lib\site-packages\tensorflow_text\__init__.py in 18 19 # pylint: disable=wildcard-import ---> 20 from tensorflow_text.core.pybinds import...

deutschoo

Add DINOv2 to KerasHub

1

divyashreepathihalli

Removing batch dimension from default layout maps for Gemma and Llama

17

This is to align with [Keras PR 20674](https://github.com/keras-team/keras/pull/20674) which fixes data sharding in the JAX trainer but does not support the model to be sharded on the "batch" dimension. To...

martin-gorner

divyashreepathihalli

Add a high-level speech to text task with generate() support.

2

Add a high-level Whisper text to speech task with generate() support. Add high level API for Whisper streaming - realtime transcription

divyashreepathihalli

add DeepSeek R1 to KerasHub

2

divyashreepathihalli

Support RLHF and other instruction fine-tuning options beyond supervised fine-tuning.

3

divyashreepathihalli

Keras newbie: Why doesn't loss decrease when fine-tuning LLaMA 3.2 3B on TPU v3-8 with Keras?

7

I'm opening this issue after asking the question in the *[Discussions](https://github.com/keras-team/keras/discussions/20659#discussioncomment-11735333)* @innat wrote that it is better to open a ticket here… So this is an *issue* ! I'm trying...

aeltorio

I want to help Keras Hub be compatible with the Qwen and Interlm models. Before that, I have some questions I'd like to ask.

1

[Qwen](https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct/tree/main) and [interlm](MathGenie/MathGenie-InterLM-20B) is chinese llm Their architectures are basically the same as Llama3, with just a few more biases in the attention layer. If I want to support them...

pass-lin

Compare Numeric with HF Model

2

Hi, I wanted to find issues/todo's to contribute to Keras, so I searched for TODO in the codebase. I found this specific todo appearing about 10 times - mostly in...

omkar-334