Yijia Shao

Update audio_spec.py

2

The preprocess_model used by BrowserFftSpec will produce "nan" results when the input tensors contain too many zero entries. Such inputs are common because there may be delays in the recording...

Hi, I'm trying to use your API to integrate a piece of python script to my flutter app which is written by Dart. Is it possible to pass arguments to...

Process run / start crashes the built macos app

24

I'm using Process.run() / Process.start() to open tensorboard in my flutter app on macos. A weird problem is everything goes well when I directly run the flutter app on the...

How to index large corpus which cannot be loaded into the memory?

7

Hi, Thank you so much for sharing and maintaining the code! I'm using the off-the-shelf ColBERT v2 for the retrieval part of my system. So, I mainly call the `index()`...

[FYI] We disable the "Create New Article" function at present

14

Thanks everyone for trying our demo and providing feedback! Due to very high demand, we have to disable the "Create New Article" function at present since our search API has...

Support cost estimation before execution and improve configuration

1

The default configurations in the Readme.md are corresponding to our hyper-parameters setup in the [NAACL'24 paper](https://arxiv.org/abs/2402.14207). As people are using/building on top of this project, having cost estimation (maximum number...

Currently, integrating STORM requires cloning the repo. Wrapping it as a python package could make integrating STORM to other project easier.

1. Clean up `requirement.txt`. 2. Resolve #77. 3. Rename `src` into `knowledge_storm` and adjust import logic. `knowledge_storm` is wrapped as a python package, available on https://pypi.org/project/knowledge-storm. 4. Update `examples`, `frontend/demo_light`...

Add `GoogleModel` to support using Gemini models in `STORMWikiRunner`. Run the following command under the root directory to test it end-to-end: ``` python -m examples.run_storm_wiki_gemini --retriever you --do-research --do-generate-outline --do-generate-article...

The current project adopts PEP 8 standard for Python code format but misses a code formatter to avoid unnecessary file change when contributing.