Nguyen Nguyen Anh
Nguyen Nguyen Anh
Can you please share how step by step you train your custom tensorflow object detection model until tflite/quantized (or not quantized / FLOAT) to use in your apps ? Thanks...
Hi, I am following your code and it looks good, will know the result sometime end week :). However, can you show me how do I get features extracted from...
Can someone please help how to receive rtmp as examples but then save into multi bitrate and accessible as VOD afterward ? ffmpeg -re -i rtmp://localhost:1935/$app/$name -filter_complex '[0:v]split=3[v1][v2][v3]; [v1]copy[v1out]; [v2]scale=w=1280:h=720[v2out];...
Hi, I have been pre-training and fine-tuning Vietnamese language on LlaMA-7B, so far so good, we just do not yet have enough computational capacity to move on. We extended the...
Hi, Can you show us how to train with COCO-style custom dataset that I have created from scratch with labeling tool. I have train.jso, test.json and image directory ready but...
I usually extend vocab to make the model closer to Vietnames language. The code is below. However, it seems that the tokenizer of LLaMA-3 is no longer work with SentencePiece....
Something is WRONG. The decoding of PreTrainedTokenizerFast (which LLaMA-3 are using) decode weird output once you add that token to the vocab using .add_tokens(word) function. I use standard tokenizer from...
A big question: should I re-pretrain after extending vocab with LLaMA-3 pretrained weight or finetuned weight ? It took forever and costly to pretrain, hence I would be eager to...
Hi, I found that this repo is focusing ONLY on fine-tuning (with LoRA) for Chinese language. However, LLaMA was trained mostly on English-corpus, with about 30,000 vocab size which is...
Is T5 (mLongT5, FlanT5 etc.) being developed with GGML ? We are looking towards this port for encoder-decoder LLM. Thanks in advanced for great works. Steve