Ijazul Haq

icon indicating a website link www.ijaz.me icon indicating an email [email protected]

icon company nlpashto icon location Guangzhou, China icon location PhD Student at Shanghai Jiao Tong University, Shanghai China. Working on AI, NLP and LLMs.

Results 3 issues of Ijazul Haq

[Question]: How the StackedEmbeddings function actually works?

3 comment

### Question How the StackedEmbeddings function actually works? Is it concatinating the two word embeddings, i.e. using torch.cat([emb1, emb2]). I want to concatinate BytePairEmbeddings with TransformerWordEmbeddings, so i'm doing like...

question

[Bug]: corpus.make_label_dictionary generate too many tags

2 comment

### Describe the bug I have only 38 tags in my POS corpus but corpus.make_label_dictionary return a dictionary of 400 tags. ### To Reproduce ```python columns = {0: 'text', 1:...

bug
Awaiting Response

TypeError: an integer is required (got type NoneType)

1 comment

**I want to pretrain BERT on custom corpus but got this error,** **Source Code** from simpletransformers.language_modeling import LanguageModelingModel, LanguageModelingArgs from sklearn.model_selection import train_test_split import pandas as pd import tokenizers orpus_path...

stale