#๐ trying to run llama3 on python
west halo
import time
def start_llama3():
process = subprocess.Popen(
['ollama', 'run', 'llama3'],
stdin=subprocess.PIPE,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
text=True
)
return process
# sending question to the llama 3 and reecievig answer
def send_to_llama3(process, user_input):
output, error = process.communicate(input=user_input)
if error:
print(f"Chyba: {error}")
return output
def main():
llama_process = start_llama3()
# Allow some time for the process to initialize
time.sleep(2)
while True:
user_input = input("enter input or use exit to exit: ")
if user_input.lower() == 'exit':
break
print("Please wait...")
llama_response = send_to_llama3(llama_process, user_input)
print("Odpovฤฤ LLaMA 3:", llama_response)
llama_process.terminate() # end llama3 process
if __name__ == "__main__":
main() ```PS C:\Users\Domino\Desktop\projekty\pyaiface> & 'c:\Users\Domino\AppData\Local\Microsoft\WindowsApps\python3.9.exe' 'c:\Users\Domino\debugpy\launcher' '63127' '--' 'c:\Users\Domino\Desktop\projekty\pyaiface\import subprocess.py'
enter input or use exit to exit: hi
Please wait...
Odpovฤฤ LLaMA 3: Hi! It's nice to meet you. Is there something I can help you with, or would you like to chat?
Traceback (most recent call last):
File "c:\Users\Domino\Desktop\projekty\pyaiface\import subprocess.py", line 41, in <module>
main()
File "c:\Users\Domino\Desktop\projekty\pyaiface\import subprocess.py", line 34, in main
llama_response = send_to_llama3(llama_process, user_input)
File "c:\Users\Domino\Desktop\projekty\pyaiface\import subprocess.py", line 16, in send_to_llama3
output, error = process.communicate(input=user_input)
File "C:\Program Files\WindowsApps\PythonSoftwareFoundation.Python.3.9_3.9.3568.0_x64__qbz5n2kfra8p0\lib\subprocess.py", line 1109,
in communicate
raise ValueError("Cannot send input after starting communication")
ValueError: Cannot send input after starting communication
PS C:\Users\Domino\Desktop\projekty\pyaiface>
surreal chasm
```import subprocess
import time
def start_llama3():
process = subprocess.P...
thats a weird way to use ollama. Why not use langchain?
Lots of tutorials on how to do this @west halo https://python.langchain.com/docs/tutorials/
But you have to swap out parts where it says OpenAI for Ollama. Like this https://python.langchain.com/api_reference/ollama/llms/langchain_ollama.llms.OllamaLLM.html#ollamallm
west halo
thanks i dont know why im using longchain becouse GPT told me to becouse i dont know much how to use python but the whole code looks like this https://paste.pythondiscord.com/QCYA
surreal chasm
thanks i dont know why im using longchain becouse GPT told me to becouse i dont ...
Well... I guess you have to properly learn Python first before you try to integrate to Ollama's llama3 model.
!resources below have lots of good beginner level material
robust sundialBOT
surreal chasm
Out of which, Byte of Python and Automate the Boring Stuff are usually recommended. But you do you.
west halo
ok thanks but do you know where can be problem in the code?
surreal chasm
ok thanks but do you know where can be problem in the code?
Yes. You're not using Langchain and you don't know Python. Learn Python, get the basics down, and then use Langchain.
west halo
ok thanks
surreal chasm
Don't jump to the end. Walk your way towards it properly
And lastly, don't use ChatGPT. As you can tell, it doesn't work.
west halo
ok thanks i will try to learn it
surreal chasm
ok thanks i will try to learn it
When you do get stuck learning it, please ask. All of us here will support you in your learning journey. ๐ All the best!
west halo
thanks
robust sundialBOT
Python help channel closed