Understanding BERT Embeddings and Tokenization | NLP | HuggingFace| Data Science | Machine Learning

[@sidharth-singh]

Quite helpful. I watched it after reading BERT paper and it makes more sense now.

[@kevon217]

great step-by-step. easy to follow and helpful!

[@RohanPaul-AI]

Great to know you liked @kevon

[@punithandharani]

Easy to understand. Thank you so much! 🙏

[@RohanPaul-AI]

Glad it was helpful!

[@Lagzilla1]

Great video but i have one question. What you have shown in this video is strictly related to "BertTokenizer" from the transformers library? I've heard that you said a lot about word embeddings and now im confused what is the purpose of SentenceTransformer('all-MiniLM-L6-v2') then? Why not just use BertTokenizer?

[@starsmaker9964]

The video was very useful for me, thank you!

[@RohanPaul-AI]

Great to know you liked @starsmaker

[@RishiGarg-ld5hx]

To implement the BERT model, do I need to run it on the google colab where I have access to their GPU or my local windows is sufficient for it?

[@RohanPaul-AI]

Always better to run in a machine with GPU. In windows without GPU you may be able to run, but will be terribly slow.

[@CppExpedition]

Exactly what i needed Thx! :D

[@RohanPaul-AI]

Great to hear you liked 🙂

[@muhdbaasit8326]

Greate video. Please Explain regarding how tokenTypeIds gets generated. its required for for the tabular data.

[@RohanPaul-AI]

Thanks @muhdbaasit8326 - And Good question, will try to cover that very soon. Stay tuned.

[@8003066717]

is there sources available, actually i want to get the BERT embedding and pass it to BiLSTM ?? i want to learn about ?

[@RohanPaul-AI]

Generally first learning resources on BERT is HuggingFace official doc. Then for many implementations examples you can search in Kaggle and Github

[@Devamandarangel]

Arrasou, obrigada pela video aula.

[@RohanPaul-AI]

Pleasure and glad to know

[@marouahamdi4293]

very helpful ! thank you

[@MuhammadJaalouk]

Thank you!