Hi Boss, how are you ? I just finished that and I noticied that the code in github need to be changed to the one you have in the video. I found out a problem because to model you have used is not adapted to a large amount of data. I tried a txt file with around 55000 lines and the model could not execute the code. I tried with around 100 lignes and I works.
Am sure the data exceeds the context window of the model. Kindly reduce the data or use another method to build your chatbot. Follow this tutorial here:ruclips.net/video/iaBBkBZQpa4/видео.htmlsi=cvjM8fNwl0mdv03N. I didn’t change the code all I did is that I added comments
I just checked it up....very good ......But another question, haven't gotten time to watch it fully though but, what were you using to store the vector chunks?
Hi Boss, how are you ? I just finished that and I noticied that the code in github need to be changed to the one you have in the video.
I found out a problem because to model you have used is not adapted to a large amount of data. I tried a txt file with around 55000 lines and the model could not execute the code.
I tried with around 100 lignes and I works.
Am sure the data exceeds the context window of the model. Kindly reduce the data or use another method to build your chatbot. Follow this tutorial here:ruclips.net/video/iaBBkBZQpa4/видео.htmlsi=cvjM8fNwl0mdv03N. I didn’t change the code all I did is that I added comments
@@tech_watt thanks dear sir
What about extracting data from PDF and breaking them as chunks or vector forms and using langchain and deepseek to build a chatbot
I did the same thing in this tutorial here:ruclips.net/video/iaBBkBZQpa4/видео.htmlsi=VQiGRjSBwZPrnbwg
I just checked it up....very good ......But another question, haven't gotten time to watch it fully though but, what were you using to store the vector chunks?
@cascatechssolutions6151 vector database. Faiss or pinecone
In the video you used which of them?
@@cascatechssolutions6151 Time to ask Deepseek