Activity
Mon
Wed
Fri
Sun
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
What is this?
Less
More

Memberships

Data Alchemy

37.5k members • Free

22 contributions to Data Alchemy
Rag-based LLM Chatbot - Problems
Hello all! I am using mistral model for building a chatbot assistant. I am facing problems with the accuracy of the model. Sometimes it doesn't it respond with the retrieved context from the documents and just gives info from outside rag, and it doesn't stick to the prompt guideline after 5 queries. It's really frustrating I've tried changing the prompt style and even tried prompt chaining but the quality of the response is very low. What's the solution for this? Or have you faced a similar problem?
0 likes • Dec '24
@Jimmy Jones For example, if I ask a random question, "give me the recipe of chocolate cake" it should respond saying sorry there is not relevant context in the documents (the ragbase) instead it gives the details about the recipe instead. Either that or it just responds with the info from the chunks retrieved. I tried lowering the temperature but it didn't really make a big difference.
0 likes • Dec '24
@Jimmy Jones yes it does.
Ask me anything!
I want to give back to this community. So if you have a question or anything you need help with (Mostly on the Developer part, since I'm an engineer) Feel free to reach out! So what makes me capable of helping you? I have been an engineer for the past 5 years, and have done many complicated tasks, especially within Machine learning, here is my latest projects: 1️⃣ Reinforcement Learning Agent – Developed an RL-based system to optimize water consumption and reduce costs for a water supply company, achieving significant improvements in resource management. 💧💡 2️⃣ LSTM-based Neural Network for Exoskeleton – Designed an advanced LSTM model for controlling exoskeletons with multiple degrees of freedom, enhancing precision and movement capabilities. 🤖💪 If you’re facing a tough challenge, feel free to reach out. I’m here to help!
0 likes • Dec '24
I am using mistral model for building a chatbot assistant. I am facing problems with the accuracy of the model. Sometimes it doesn't it respond with the retrieved context from the documents and just gives info from outside rag, and it doesn't stick to the prompt guideline after 5 queries. It's really frustrating I've tried changing the prompt style and even tried prompt chaining but the quality of the response is very low. What's the solution for this? Or have you faced a similar problem?
1 like • Dec '24
@Louis Ildal I did try prompt chaining, it was a bit better but I faced the similar problems again. I'll try combing the different techniques you mentioned. Thanks a lot!
Suggestions and guidance on building rag based chatbot
I need to build a rag based document retrieval llm, with the following features. I am having difficulty in coding it, because of the multiple dependencies and package versions, so if anyone has any reference code that I can use, it would be helpful. Features: - It must give information strictly from the documents only and not from the internet - It must give sources such as page number and document at the end of each response - It must be able to accept text, audio and images as input - It should very accurate in it's response giving information from the document chunks - The documents knowledge base should support PDF, audio and video formats as well. - This one I find very tricky, the bot must probe the user to explore more topics based on question, or probe the user if the query is very vague or hard to understand. - Other features like chat history, search through chats etc must also be there. I would appreciate the support!
0 likes • Nov '24
@Marcio Pacheco thanks! will check it out.
Help
Hi guys! I’ve completed the Kaggle first tutorial on programming and I feel like all the info went right over my head. Anyone feel the same way? Or have constructive learning tips?
1 like • Aug '24
If you're new to programming I would suggest you to understand the basic concepts first. So that you can know the meaning of each line of code that you do. You can also use ChatGPT if you are stuck, ask it to explain the reason, explanation and provide similar examples. Once you are familiar with the concepts, things will start clicking!
0 likes • Aug '24
@Stephanie Wuethrich if you want to really go deep into the conceptual level I would suggest to follow along with a textbook, there are some pretty beginner friendly books on python programming. Or enroll into free online courses as well. Just keep the learning going! Are you in computer science field or?
How to confuse machine learning 🤣
Someone posted this image on LinkedIn. I found it hilarious.
How to confuse machine learning 🤣
1 like • Aug '24
😂
1-10 of 22
Rishita Umasankaran
4
57points to level up
@rishita-umasankaran-6754
AI & ML Enthusiast

Active 260d ago
Joined Aug 12, 2024
INFJ
India
Powered by