-
Hello!
I noticed that the maximum eposides can be controlled by MAX_EPISODES during training, and EVAL_INTERVAL determines the evaluation intervals; however, the evaluation process seems to determi…
-
It seems need 1.1M.tsv In scripts/pretraining/1B_pt.sh.
the format should like: # line format: source, path, total_time, start_time, end_time, target
but the [UniFormerV2] provide # line format: pa…
-
Hi Nikhil,
Thank you for sharing the code! However, I can't find any information about downloading the data. Could you share more details of that?
Thanks,
Changan
-
Hello, I wondered if there was a way to train the model and new data. I am sorry if I missed the documentation somewhere.
-
Hi, thank you for your great job!
I want to finetune ControlNeXt-SVD-v2 on my own dataset. And I have some problems with data preprocessing.
---
First is `guide_path` in [meta_info.json](http…
-
If I try to download the training data in R, I get the following:
`
download(training = TRUE)
`
Error in vers[i]
-
### GPT-3 data mix
* Datasets are not sampled in proportion to their size
* Datasets we view as higher-quality are sampled more frequently
* WebText2, Book1, Wikipedia datasets are sampl…
-
Hi, I want to continue pre-training based on the pre-trained model in the first stage, and then SFT,and I have some questions.
1, where is the weight of the pre-trained model in the first stage?
2, …
-
In the paper, it is mentioned that one could generate the training data using captions from images. Would it be possible to provide the training data used to train the models. I'd like to try and impr…
-
Implement a method in pzserver lib to create training sets with mock data from the parquet files stored on Lustre.
```[tasklist]
### Tasks
- [x] default method = random selection
- [ ] https:/…