Zabir Al Nazi Nabil

I can convert the model from onnx, can print the summary, but when I try to save the model with `model.save('k_model.h5')`, and then try to reload the model, I get...

[BUG] HTTP Error 429: Too Many Requests from linux server

4

**Describe the bug** HTTP Error 429: Too Many Requests ``` ytb = YouTube(vid) ytb.streams.filter(res="360p").first().download(output_path = f"{seg_id}_{ch}", filename = f"vid_{vid_idx}.mp4") ``` **To Reproduce** I'm on an ubuntu 18. The script was...

I was trying to train XLNet on a multilabel text classification dataset. I'm using an RTX 2060 to train, but the training seems extremely slow, I wanted to test only...

assert x.size()[1] == self.in_feats AssertionError

1

Hi, thanks for sharing this amazing repository. When using `aamsoftmax`, I am getting this error for all of the models. ``` Save path: exps/nist_old_bang_youtube_aamsoft_rawnet Embedding size is 256, encoder SAP....

How can use the pre-trained model for fine-tuning on a small dataset?

1

I suppose, the default `main.py` training will start training from scratch, how can use one of your pretrained weights and start training with that?

how to install what if in managed notebook of vertex ai?

3

It seems what-if is not available directly in managed notebooks? How can I install it? I have tried multiple times, but either I get permission error (don't have sudo), or...

running the generate.py on the base llama weights return garbage

5

I modified the codebase a little: ``` """code for zero shot instruction parsing""" import torch from peft import PeftModel import transformers import textwrap from transformers import AutoModel, AutoTokenizer, LlamaTokenizer, LlamaForCausalLM,...

Hi, I took the pytorch weight from your repository and tested on timit dataset. This is the result I got. ``` 100%|███████████████████████████████████████████████████████████████████████| 550000/550000 [00:20

Can you kindly share the pretrained weight in a public drive so it can be downloaded?

In the test list file, the entries are kept like following 1 filepath1 filepath2 0 filepath1 filepath3 How to generate this file?