🎯 Key Takeaways for quick navigation: 00:00 📋 *Introduction and Agenda* - Introduction to LangChain and the purpose of the video. - Agenda for the video, focusing on practical orientation and end-to-end projects. 02:46 🛠️ *Environment Setup and API Key* - Setting up a virtual environment (VENV) for the project. - Obtaining and handling the OpenAI API key. - Installing necessary libraries. 08:35 🧠 *Understanding Temperature in Language Models* - Explanation of the temperature parameter in language models. - How temperature affects model output creativity. 16:12 🤖 *Model Prediction and Output* - Demonstrating model prediction with a sample input. - Displaying the output of the language model. 18:30 🤝 *Using Hugging Face Models* - Installing the Hugging Face library. - Setting up the Hugging Face API token for accessing open-source models. - Exploring available Hugging Face models for text generation. 21:21 🧩 *Understanding LangChain and Hugging Face Integration* - Learn how to integrate LangChain and Hugging Face for language models. 23:28 📊 *Comparing OpenAI GPT-3.5 and Hugging Face LLM Models* - OpenAI GPT-3.5 provides more detailed responses compared to Hugging Face LLM models. 27:09 🧵 *Working with Prompt Templates in LangChain* - Using prompt templates to structure input for LLN models. 34:12 🔗 *Combining Multiple Chains with Simple Sequential Chain* - How to combine multiple chains using the Simple Sequential Chain approach. 01:07:22 🚀 *Setting Up Basic Application* - Creating a basic application. - Initializing a Streamlit app. - Setting page titles and headers. - Handling user input with a submit button. 01:09:55 🧩 *Capturing User Input and Calling OpenAI Model* - Capturing user input using a text field. - Sending user input to the OpenAI model. - Processing and returning the model's response. - Exploring how the OpenAI model can be customized. 01:13:19 🌐 *Deployment on Hugging Face Spaces* - Deploying the application on Hugging Face Spaces. - Adding secret keys for OpenAI API. - Uploading the application files. - Testing the deployed chatbot in a public space. Made with HARPA AI
Thanks Krish , these video are amazing & because of you it seems like its really possible to quickly learn & leverage open ai llm models . Keep up the good work !!! Cheers !!
it's a great tutorial. just one issue that video title says hugging face hub and video is mostly implemented using openai where we usually get rate limit error in openai api.
If anybody is encountering this error- {“error”:“Authorization header is correct, but the token seems invalid”} Do the following steps- 1. go to hugging face account → setting → Access Token 2. Edit Access Token Permissions 3. Under the Repositories check box Write access to contents/settings of all repos under your personal namespace. 4. Under the Inference check box Make calls to the serverless Inference API 5. Save and retry to run your code.(If still not working then delete this access token, create a new one will these permissions) Happy Learning!!😉
Hello Sir, The openAI api and google gemini api both are paid so how a student who wants to include this project in his portfolio bears the cost because they are expensive
hi Krish! I am getting AttributeError: module 'openai' has no attribute 'error'. while trying to run llm.predict(text). I don't know the issue. tried looking online but couldn't find a soultion. could it possibly a version related issue?
i am unable to implement / run part after sequential chain becoz i dont have access to openAI model any alternative for those models cant use opensource models from huggingface in ChatOpenAI please suggest alternet suggestion
I faced ERROR "AttributeError: module 'openai' has no attribute 'Completion' while running llm=OpenAI() after today's update in openai, might be this error is coming. I switched to "openai==v0.28.1" and it worked fine.
What LLM model do you use in the project? I mean is it chatgpt or llama? In the code we dont mention the type of LLM model? Is it gpt4 since we are using OPEN AI? @krish naik
Krish please tell me why I am getting the error from langchain.llms import google_palm import os from dotenv import load_dotenv load_dotenv() TypeError Traceback (most recent call last) Cell In[16], line 1 ----> 1 llm = google_palm(google_api_key=os.environ["GOOGLE_API_KEY"], temperature=0.1) TypeError: 'module' object is not callable
I have this use-case where there are different types of documents. I can parse documents using document loaders using langchain. But, there are images also in these documents. I want to store them as metadata and if answer generated from a context chunk it show the image also. Please help sir
I am facing Ratelimt error and found no usage on my OpenAi dashboarad, what may be the possible reason? Please help me if anyone know hot to solve this?
Thanks Krish , can you make a video on training lanchain and open ai api with a set of input and output documents so that with a set of new documents it is able to create an output document
if i install packages without versions in requirements.txt am getting following error: pip install -r requirements.txt ERROR: Could not find a version that satisfies the requirement lanchain==0.1.7 (from versions: none) ERROR: No matching distribution found for lanchain==0.1.7
Two things here: 1) If the current execution run after multiple successful response yields this error, then wait for a while before you run for next time. The model has a limit of at least 3 responses per request. 2) if you're facing this issue when your existing API key for the model has expired, create a new API key with new mobile and email address. It will help you create a new API key for the model. Hope this helps.
Because all the python libraries used in this project are compatible and tested with 3.9 versions, if you want you can update only issue is sometime syntax get change
Not everyone is providing how to finetune llms cause they donot have any idea how it works.They will , not now only in future after copy pasting someone ele's content.
If you don't probably stop using probably so insanely probably incorrectly in ur probable speech, I'll probably unsubscribe despite such probably good content.
Hello Krish I saw ur langchain video. It was nice & i want to implement it. But i m getting error while running api key. Erro code 429. That my credit is used. I have created new (1st time) in openAi, still i m getting this error. Can u help mein this
hi Krish, every time when i pass argument in final, getting below error. hipe this will not trouble: "Retrying langchain.chat_models.openai.ChatOpenAI.completion_with_retry.._completion_with_retry in 4.0 seconds as it raised ServiceUnavailableError: The server is overloaded ......."
i have made a google form for reasearch study on data science and i want to reach to as many as possible, to collect the study data. if anyone can help me to get in a group or would like to contribute please reply. i'm student so don't worry about the data integrity it's for study!
@@AIMLDLNLP-TECH real data, i need response from data science field from student to employee if you know any group i can give you my number, this project is for my study i want to do data cleaning and analysis on this
Everytime in all your videos you make a virtual env naming venv! I wonder how many venv you have? So for 100 projects if there are 100 venvs won't it be confusing? @KrishNaik
the venv's made here are with -p (path) switch, which tells you not a project specific to that folder. so to activate it you will have to provide the path to that environment. To access this environment globally you have to provide the entire path to that venv which in any case you will know where are going.
🎯 Key Takeaways for quick navigation:
00:00 📋 *Introduction and Agenda*
- Introduction to LangChain and the purpose of the video.
- Agenda for the video, focusing on practical orientation and end-to-end projects.
02:46 🛠️ *Environment Setup and API Key*
- Setting up a virtual environment (VENV) for the project.
- Obtaining and handling the OpenAI API key.
- Installing necessary libraries.
08:35 🧠 *Understanding Temperature in Language Models*
- Explanation of the temperature parameter in language models.
- How temperature affects model output creativity.
16:12 🤖 *Model Prediction and Output*
- Demonstrating model prediction with a sample input.
- Displaying the output of the language model.
18:30 🤝 *Using Hugging Face Models*
- Installing the Hugging Face library.
- Setting up the Hugging Face API token for accessing open-source models.
- Exploring available Hugging Face models for text generation.
21:21 🧩 *Understanding LangChain and Hugging Face Integration*
- Learn how to integrate LangChain and Hugging Face for language models.
23:28 📊 *Comparing OpenAI GPT-3.5 and Hugging Face LLM Models*
- OpenAI GPT-3.5 provides more detailed responses compared to Hugging Face LLM models.
27:09 🧵 *Working with Prompt Templates in LangChain*
- Using prompt templates to structure input for LLN models.
34:12 🔗 *Combining Multiple Chains with Simple Sequential Chain*
- How to combine multiple chains using the Simple Sequential Chain approach.
01:07:22 🚀 *Setting Up Basic Application*
- Creating a basic application.
- Initializing a Streamlit app.
- Setting page titles and headers.
- Handling user input with a submit button.
01:09:55 🧩 *Capturing User Input and Calling OpenAI Model*
- Capturing user input using a text field.
- Sending user input to the OpenAI model.
- Processing and returning the model's response.
- Exploring how the OpenAI model can be customized.
01:13:19 🌐 *Deployment on Hugging Face Spaces*
- Deploying the application on Hugging Face Spaces.
- Adding secret keys for OpenAI API.
- Uploading the application files.
- Testing the deployed chatbot in a public space.
Made with HARPA AI
That joke on 3cr package was very subtle :)
What a Blessing you are! GOD Bless you exponentially 🎉
Thanks Krish , these video are amazing & because of you it seems like its really possible to quickly learn & leverage open ai llm models . Keep up the good work !!! Cheers !!
I think practical and particular domain related videos will help more, this is very generic
Thank you sir for creating this type of projects with langchain
Super excited for this series
Are you a data scientist Aditya?
it is a great tutorial without doubt! Please upload more videos like this❤❤
👏🏼
learning by practice is the best thing ; THANK YOU SIR KRISH
Amazing tutorial.. thank you krish .. you are great.. love you ..
Thank you Krish sir for such an informative video!
Thanks very much needed resource.
Thanks Krish, Please upload more videos also.
it's a great tutorial. just one issue that video title says hugging face hub and video is mostly implemented using openai where we usually get rate limit error in openai api.
Great Sir❤
Please try to upload videos on Generative AI more frequently. 😢
What kind of videos you are looking for in Generative AI NLP or computer vision related Farhan ?
Great tutorial
It's my sincere request for you to make a detailed Computer Vision playlist
❤
I will probably like this video
Nice Explanation
Love it!!!!!!!!!!!
If anybody is encountering this error-
{“error”:“Authorization header is correct, but the token seems invalid”}
Do the following steps-
1. go to hugging face account → setting → Access Token
2. Edit Access Token Permissions
3. Under the Repositories check box Write access to contents/settings of all repos under your personal namespace.
4. Under the Inference check box Make calls to the serverless Inference API
5. Save and retry to run your code.(If still not working then delete this access token, create a new one will these permissions)
Happy Learning!!😉
Thank you sir
Please make a video about end to end design of a chatbot using gen ai for any website
You are correct! I was thinking the capital of Austrailia is Melbourne! LOL!
Thanks for the video
Legend
Hello Sir,
The openAI api and google gemini api both are paid so how a student who wants to include this project in his portfolio bears the cost because they are expensive
very informative video..can we implement embeddings when we use gpt-3.5-turbo model?
Hey, Krsih make some videos on AutoGen to, BTW awesome job.
Good explanation Sir...
Please, Can you elaborate how to do deployment using cli as well especially in huggingface?
hi Krish! I am getting AttributeError: module 'openai' has no attribute 'error'. while trying to run llm.predict(text). I don't know the issue. tried looking online but couldn't find a soultion. could it possibly a version related issue?
I am facing the same issue.. any solution to this? Please help
@@atharvpingle6937yes.. i changed my file's name from lanchain.ipynb to something else. That fixed the issue..
i am unable to implement / run part after sequential chain becoz i dont have access to openAI model
any alternative for those models
cant use opensource models from huggingface in ChatOpenAI
please suggest alternet suggestion
I faced ERROR "AttributeError: module 'openai' has no attribute 'Completion' while running llm=OpenAI()
after today's update in openai, might be this error is coming.
I switched to "openai==v0.28.1" and it worked fine.
yes, it's the latest openai version that's the issue
I have downgraded openai to v0.28.1, still I am facing the issue.
how u resolved it my mind is
fucked up@@vmajay007
I still get the same error. what is the langchain version you've tried?
at 1:17:10 how do we run the app after updating the API key?
Is necessary to buy credits to use it? In other video I got the message " insufficient quota" .
Did you get the solution?
hey krish, can u try to integrate sql in to your projects
Can you specify what do you need ? ETL or langchain with ur own data ??
@@AIMLDLNLP-TECH store and retrieve data in project pipeline
Why are people always begging for more? I always see these comments below krish's videos.
What LLM model do you use in the project? I mean is it chatgpt or llama? In the code we dont mention the type of LLM model? Is it gpt4 since we are using OPEN AI?
@krish naik
Sir I have a douby while using API keys , I get rate limit error
Installing IPI kernal required only 1 time for Jupitar Notebook as we are installing out of environment?
will it be safe to use llm when your main concern is privacy can i use it with my important data
I like how he keeps dangling the API key infront of us 😂
33:58 Error achhe hai😅
what kind of token I should create in the hugging face
Krish please tell me why I am getting the error
from langchain.llms import google_palm
import os
from dotenv import load_dotenv
load_dotenv()
TypeError Traceback (most recent call last)
Cell In[16], line 1
----> 1 llm = google_palm(google_api_key=os.environ["GOOGLE_API_KEY"], temperature=0.1)
TypeError: 'module' object is not callable
Hi Krish, Can you please share what's in the .env file.
I have this use-case where there are different types of documents. I can parse documents using document loaders using langchain. But, there are images also in these documents. I want to store them as metadata and if answer generated from a context chunk it show the image also. Please help sir
I am facing Ratelimt error and found no usage on my OpenAi dashboarad, what may be the possible reason? Please help me if anyone know hot to solve this?
why when I try pip install, it shows "Defaulting to user installation because normal site-packages is not writeable"
Make a video on how to use AutoGen
Do you need autogen research paper explanation or start to end llm project using autogen ?
Thanks Krish , can you make a video on training lanchain and open ai api with a set of input and output documents so that with a set of new documents it is able to create an output document
Langchain llm trained on your pdf document 📃 chatbot system do you need Satvik?
@@AIMLDLNLP-TECH yeah kind of we need to train llm on our pdf(s) so that it is able to create the output document , not a chatbot system definitely
It worked completely fine and ven openeed the streamlit webpage, but nothing is shown
Hi Sir, you have a video where you teach how to query a PDF. Is it possible to query, at the same .py my PDF and gpt-4, for example? Tks a lot
Temperature value is between 0 and 2.
can u please make a tutorial that uses Mistral AI 7B from huggingface to make a chatbot
I have tried everything i could but still getting this error:
ImportError: Can't connect to HTTPS URL because the SSL module is not available.
why there are so many ways to run llm im confuse with them like when to use when like there is predict, call, invoke, generate
if i install packages without versions in requirements.txt am getting following error:
pip install -r requirements.txt
ERROR: Could not find a version that satisfies the requirement lanchain==0.1.7 (from versions: none)
ERROR: No matching distribution found for lanchain==0.1.7
spelling mistake - langchai not lanchain
I am getting that I have exceeded my limit for open AI
Do we need to get paid subscription for OpenAI API ?
ia m also getting the same . kindly reply if u find a soln
hey any solution for this@@kanishkasharma353
Can try this in Google colab
why openAI cant we use LLAMA2 for RAG?
I am facing RateLimitError... How to fix this error
Two things here:
1) If the current execution run after multiple successful response yields this error, then wait for a while before you run for next time. The model has a limit of at least 3 responses per request.
2) if you're facing this issue when your existing API key for the model has expired, create a new API key with new mobile and email address. It will help you create a new API key for the model.
Hope this helps.
"Authorization header is correct, but the token seems invalid" ye error bar bar araha hain ..please koi batao kaise fix karun yesab ???
Hi Krish Sir.. I am getting AttributeError: module 'openai' has no attribute 'error' . Can you please help how to resolve and proceed further?
Make sure your python version is greater than 3.8. And also upgrade your open AI
how should i solve the rate limit errror
Why python==3.9 version is used in this project. There are new version available like 3.10, 3.11, 3.12
Because all the python libraries used in this project are compatible and tested with 3.9 versions, if you want you can update only issue is sometime syntax get change
Thanks @@AIMLDLNLP-TECH
Its quite outdated now. Most of the langchain packages have changed.
How do I learn I kind of need spoon feeding
Not everyone is providing how to finetune llms cause they donot have any idea how it works.They will , not now only in future after copy pasting someone ele's content.
If you don't probably stop using probably so insanely probably incorrectly in ur probable speech, I'll probably unsubscribe despite such probably good content.
Hi, Can we access the open Ai api key for free?
Guys is the open AI Api free?
Naah, It's not 🙃
Hello Krish
I saw ur langchain video.
It was nice & i want to implement it.
But i m getting error while running api key.
Erro code 429.
That my credit is used.
I have created new (1st time) in openAi, still i m getting this error.
Can u help mein this
hi Krish, every time when i pass argument in final, getting below error. hipe this will not trouble:
"Retrying langchain.chat_models.openai.ChatOpenAI.completion_with_retry.._completion_with_retry in 4.0 seconds as it raised ServiceUnavailableError: The server is overloaded ......."
I am frustrated get job for as fresher I should I can learn
Communication Skills and English.
is open ai key free?
yes
evey youtuber make video on openAi why
half & half
i have made a google form for reasearch study on data science and i want to reach to as many as possible, to collect the study data.
if anyone can help me to get in a group or would like to contribute please reply.
i'm student so don't worry about the data integrity it's for study!
Do you need real data or random data
@@AIMLDLNLP-TECH real data, i need response from data science field from student to employee if you know any group i can give you my number,
this project is for my study i want to do data cleaning and analysis on this
I'm interested
Rate Limit error agya 😢
Who wants your API KEY dude? C’mon.. and btw you can always create a new one. Stop behaving like a kid about the API KEY
Bhai phle to tu probably probably bolna km kr, boht irritate krta h
I thought it was Sydney
Everytime in all your videos you make a virtual env naming venv! I wonder how many venv you have? So for 100 projects if there are 100 venvs won't it be confusing? @KrishNaik
the venv's made here are with -p (path) switch, which tells you not a project specific to that folder. so to activate it you will have to provide the path to that environment. To access this environment globally you have to provide the entire path to that venv which in any case you will know where are going.
I am getting ratelimit error why?