-
### 현재 상황과 문제점
본 Task에서 데이터 증강은 Question과 Answer에 대한 증강이 필요하다.
Context와 Answer를 고정시키고 Question을 생성하는 Question Generation 방법으로 Question과 Answer Pair를 생성할 수 있다.
### 개선 제안 사항
1. 모델 학습을 통한 Question 생성…
-
### Question Validation
- [X] I have searched both the documentation and discord for an answer.
### Question
Hello there, I have an use case for my company of which I set up a naive RAG + reranker …
-
Hi,
I have compiled and installed CycloneDDS, and am having issue finding `dds_idlc` to generate the message. Is `idlc` different from `dds_idlc`, and where could `dds_idlc` be found in `CycloneDDS…
-
During the image generation process, what‘s in positions of is their embedding? which is independent of the last generated representations. If so, can I assume that the four representations were g…
-
Hello, author.
I use your script and tutorial to fine-tune on the concode dataset, but I get the same error every time. Do you know how to solve it?
```shell
# python run.py \
--do_train \
…
-
Hi there!
First of all, thanks for your awesome work!!!
Already asked it at the bitcrack repo.
But would it be possible to use the hex-privatey as Input , treat it as bytes, take one round of sha25…
-
### Self Checks
- [X] This is only for bug report, if you would like to ask a question, please head to [Discussions](https://github.com/langgenius/dify/discussions/categories/general).
- [X] I have s…
-
Hi,
This is a great work and thanks for sharing code.
In act_max.tvd.center_part.py, I think octaves hyperparams are used for denoise purpose. But I don't quite understand it. Could you please …
-
- [ ] Collect a bunch of example inputs and outputs from last summer's exploratory work
- **We have one or two examples already.**
- [ ] Wrangle these examples into something we can use to train (f…
-
根据文档中的命令，完成了40B的lora微调训练，微调训练命令如下：
`GPUS=8 PER_DEVICE_BATCH_SIZE=2 sh shell/internvl2.0/2nd_finetune/internvl2_40b_hermes2_yi_34b_dynamic_res_2nd_finetune_lora.sh`
针对微调后的模型，模型可以成功加载，加载命令如下：
`mode…