[Python] RunnableSequence instead of LLMChain throws an error (updating from depreacated...

When I have this code first

from langchain_community.llms import HuggingFacePipeline
from transformers import AutoTokenizer
import transformers
import torch

model="meta-llama/Llama-2-7b-chat-hf"
tokenizer=AutoTokenizer.from_pretrained(model)

pipeline=transformers.pipeline(
"text-generation",
model=model,
tokenizer=tokenizer,
torch_dtype=torch.bfloat16,
trust_remote_code=True,
device_map="auto",
max_length=1000,
do_sample=True,
top_k=10,
num_return_sequences=1,
eos_token_id=tokenizer.eos_token_id
)

llm=HuggingFacePipeline(pipeline=pipeline, model_kwargs={'temperature':0})

from langchain.prompts import PromptTemplate
prompt_template=PromptTemplate(input_variables=["book_name"],
template="Provide me a concise summary of the book {book_name}")


and then I complete it with

from langchain.chains import LLMChain

chain = LLMChain(llm=llm, prompt=prompt_template, verbose=True)
response= chain.run("Alchemist")
print(response)


I get a response with the summary I wanted , but I get deprecation warnings. So following the warnings I try to replace the second part with

chain = prompt | llm
response = chain.invoke("The name of the rose")
print(response)


but I get the error

TypeError: Expected a Runnable, callable or dict.Instead got an unsupported type: <class 'str'>


What am I doing wrong?

I have used something similar but with llm being a HuggingFaceEndPoint and in that case it worked, so I suspect that it has to do with llm being a HuggingFacePipeline but can someone tell me how to correct the code?

Edit: I tried

chain2 = prompt | llm | StrOutputParser()
response2 = chain2.invoke({"bookname":"The name of the rose"})
# response2 = chain2.invoke("The name of the rose")
print(response2)


but the error persists

Continue reading...