from llama_index.llms.bedrock import Bedrock

llm = Bedrock(model="meta.llama2-13b-chat-v1", profile_name="machineuser1")

embed_model = HuggingFaceEmbedding(model_name="BAAI/bge-small-en-v1.5")

service_context = ServiceContext.from_defaults(
    llm = llm,
    embed_model = embed_model,
    chunk_size=256,
)

`
tru = Tru()
tru.reset_database()

llm = Ollama(model="wizard-vicuna-uncensored",base_url="http://192.168.1.232:11435")

response = completion(
    model="ollama/wizard-vicuna-uncensored", 
    messages=[{ "content": "respond in 20 words. who are you?","role": "user"}], 
    api_base="http://192.168.1.232:11435"
)

LiteLLM.set_verbose=True

litellm_provider = LiteLLM(model_engine="ollama/wizard-vicuna-uncensored", endpoint="http://192.168.1.232:11435")

grounded = Groundedness(groundedness_provider=litellm_provider)

f_groundedness = (
    Feedback(grounded.groundedness_measure_with_cot_reasons, name = "Groundedness")
    .on(Select.RecordCalls.retrieve.rets.collect())
    .on_output()
    .aggregate(grounded.grounded_statements_aggregator)
)

f_qa_relevance = (
    Feedback(litellm_provider.relevance_with_cot_reasons, name = "Answer Relevance")
    .on(Select.RecordCalls.retrieve.args.query)
    .on_output()
)
f_context_relevance = (
    Feedback(litellm_provider.qs_relevance_with_cot_reasons, name = "Context Relevance")
    .on(Select.RecordCalls.retrieve.args.query)
    .on(Select.RecordCalls.retrieve.rets.collect())
    .aggregate(np.mean)
)

from trulens_eval import TruLlama
from trulens_eval import FeedbackMode

tru_recorder = TruLlama(
    query_engine3,
    app_id="App_1",
    feedbacks=[
        f_qa_relevance,
        f_context_relevance,
        f_groundedness
    ]
)

for question in eval_questions:
    with tru_recorder as recording:
        print(question)
        query_engine3.query(question)