Update main.py

main.py

@@ -5,11 +5,22 @@ from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline,BitsAndBy
 from langchain_core.prompts import PromptTemplate
 
 quants = BitsAndBytesConfig(load_in_4bit=True)
-model_id = "mistralai/Mistral-7B-Instruct-v0.2"
-tokenizer = AutoTokenizer.from_pretrained(model_id,quantization_config=quants)
-model = AutoModelForCausalLM.from_pretrained(model_id,quantization_config=quants)
-pipe = pipeline("text-generation", model=model, tokenizer=tokenizer,max_new_tokens=1000)
-hf = HuggingFacePipeline(pipeline=pipe)
+# model_id = "mistralai/Mistral-7B-Instruct-v0.2"
+
+from langchain_community.llms.huggingface_pipeline import HuggingFacePipeline
+
+hf = HuggingFacePipeline.from_model_id(
+    model_id="mistralai/Mistral-7B-Instruct-v0.2",
+    task="text-generation",
+    pipeline_kwargs={"max_new_tokens": 10000},
+    model_kwargs={"quantization_config":quants}
+)
+
+
+# tokenizer = AutoTokenizer.from_pretrained(model_id,quantization_config=quants)
+# model = AutoModelForCausalLM.from_pretrained(model_id,quantization_config=quants)
+# pipe = pipeline("text-generation", model=model, tokenizer=tokenizer,max_new_tokens=1000)
+# hf = HuggingFacePipeline(pipeline=pipe)
 
 def generate_blog(role , words , topic):
     template = ''' You are an expert Blog generator , Given the Topic , the intended audience and the maximum number of words ,