-
## The problem
Right now, quickwit offers a dynamic json and object to represent deep structure.
The tokens emitted are however flattened.
Taking the example of documents holding a t-shirt sh…
-
### Summary
We can adopt a more flexible and performant event model by using `cosmossdk.io/schema/appdata` events in `core/app` and code which implements that API.
### Problem Definition
The event …
-
### What happened?
During our tests we stumbled over an issue with the current JSON mapping. The new JSON mapping is actually not an entity in GitHub yet. I could have allocated the to Methodology, b…
-
I'm trying to generate embeddings from scGPT for my single-cell data but encountering tokenization issues. Here's my scenario and the errors I'm facing:
1. Initial Setup:
```python
adata = ad…
-
### Describe the bug
Autobuild .save() doesn't record Function/Tools config information in the json
### Steps to reproduce
Following the steps in this ipynb notebook: https://github.com/krishnashed…
-
Bij het genereren van een model met imvertor 4.0 krijg ik de volgende melding, ook al heb ik geen Json output gevraagd.
 affected**: 4.0.6
**Description**
The ApiFilter attribute takes over my empty "filters" for my post counter.
**How to reproduce**
```
-
## Motivation
LLM users and existing tools most commonly use the OpenAI API. TensorZero currently has an API that maps onto our internal representations, but we should also offer an OpenAI-compatib…
-
!swift sft \
--model_type got-ocr2 \
--model_id_or_path stepfun-ai/GOT-OCR2_0 \
--sft_type lora \
--dataset /kaggle/working/output_data.json \
--output_dir /kaggle/working/hindi_got_model_3 \
--…
-
### What happened?
**Background:**
To make management of configuration files and schemas simpler, and unit testable we want to utilize the [multiple data sources](https://github.com/Azure/data-api-b…