xrag.eval package

xrag.eval package#

Submodules#

xrag.eval.DeepEvalLocalModel module#

class xrag.eval.DeepEvalLocalModel.DeepEvalLocalModel(model, tokenizer)[source]#

Bases: DeepEvalBaseLLM

async a_generate(prompt)[source]#

Runs the model to output LLM response.

Return type:

str

Returns:

A string.

generate(prompt)[source]#

Runs the model to output LLM response.

Return type:

str

Returns:

A string.

get_model_name()[source]#
load_model()[source]#

Loads a model, that will be responsible for scoring.

Returns:

A model object

xrag.eval.EvalModelAgent module#

class xrag.eval.EvalModelAgent.EvalModelAgent(args)[source]#

Bases: object

xrag.eval.EvalModelAgent.qwen_completion_to_prompt(completion)[source]#

xrag.eval.evaluate_LLM module#

class xrag.eval.evaluate_LLM.EvaluationResult_LLM(metrics=None)[source]#

Bases: object

add(evaluate_result)[source]#
print_results()[source]#
print_results_to_path(path, config, sample_arr)[source]#
xrag.eval.evaluate_LLM.UptrainEvaluate(evalModelAgent, question, actual_response, retrieval_context, expected_answer, gold_context, checks, local_model='qwen:7b-chat-v1.5-q8_0')[source]#
xrag.eval.evaluate_LLM.evaluating(question, response, actual_response, retrieval_context, retrieval_ids, expected_answer, golden_context, golden_context_ids, metrics, evalModelAgent)[source]#
xrag.eval.evaluate_LLM.get_DeepEval_Metrices(evalModelAgent, model_name='DeepEval_retrieval_contextualPrecision')[source]#
xrag.eval.evaluate_LLM.get_llama_evaluator(evalModelAgent, model_name='Llama_retrieval_Faithfulness')[source]#
xrag.eval.evaluate_LLM.upTrain_evaluate_self(settings, data, checks, project_name=None, schema=None, metadata=None)[source]#

xrag.eval.evaluate_TGT module#

xrag.eval.evaluate_TGT.NLGEvaluate(questions, actual_responses, golden_contexts, golden_context_ids, metrics)[source]#

xrag.eval.evaluate_TRT module#

xrag.eval.evaluate_TRT.DCG(retrieved_ids, expected_ids)[source]#
xrag.eval.evaluate_TRT.Em(retrieved_ids, expected_ids)[source]#
class xrag.eval.evaluate_TRT.EvaluationResult_TRT(metrics=None)[source]#

Bases: object

add(evaluate_result)[source]#
print_results()[source]#
print_results_to_path(path, config, sample_arr)[source]#
xrag.eval.evaluate_TRT.F1(retrieved_ids, expected_ids)[source]#
xrag.eval.evaluate_TRT.Hit(retrieved_ids, expected_ids)[source]#
xrag.eval.evaluate_TRT.IDCG(retreived_ids, expected_ids)[source]#
xrag.eval.evaluate_TRT.MAP(retrieved_ids, expected_ids)[source]#
xrag.eval.evaluate_TRT.Mrr(retrieved_ids, expected_ids)[source]#
xrag.eval.evaluate_TRT.NDCG(retrieved_ids, expected_ids)[source]#
xrag.eval.evaluate_TRT.evaluating_TRT(retrieval_ids, golden_context_ids)[source]#

xrag.eval.evaluate_rag module#

xrag.eval.evaluate_rag.DCG(retrieved_ids, expected_ids)[source]#
xrag.eval.evaluate_rag.Em(retrieved_ids, expected_ids)[source]#
class xrag.eval.evaluate_rag.EvaluationResult(metrics=None)[source]#

Bases: object

add(evaluate_result)[source]#
get_results_str()[source]#
print_results()[source]#
print_results_to_path(path, config, sample_arr)[source]#
xrag.eval.evaluate_rag.F1(retrieved_ids, expected_ids)[source]#
xrag.eval.evaluate_rag.Hit(retrieved_ids, expected_ids)[source]#
xrag.eval.evaluate_rag.IDCG(retreived_ids, expected_ids)[source]#
xrag.eval.evaluate_rag.MAP(retrieved_ids, expected_ids)[source]#
xrag.eval.evaluate_rag.Mrr(retrieved_ids, expected_ids)[source]#
xrag.eval.evaluate_rag.NDCG(retrieved_ids, expected_ids)[source]#
xrag.eval.evaluate_rag.NLGEvaluate(questions, actual_responses, expect_answers, golden_context_ids, metrics)[source]#
xrag.eval.evaluate_rag.UptrainEvaluate(evalModelAgent, question, actual_response, retrieval_context, expected_answer, gold_context, checks, local_model='qwen:7b-chat-v1.5-q8_0')[source]#
xrag.eval.evaluate_rag.evaluating(question, response, actual_response, retrieval_context, retrieval_ids, expected_answer, golden_context, golden_context_ids, metrics, evalModelAgent)[source]#
xrag.eval.evaluate_rag.get_DeepEval_Metrices(evalModelAgent, model_name='DeepEval_retrieval_contextualPrecision')[source]#
xrag.eval.evaluate_rag.get_llama_evaluator(evalModelAgent, model_name='Llama_retrieval_Faithfulness')[source]#
xrag.eval.evaluate_rag.upTrain_evaluate_self(settings, data, checks, project_name=None, schema=None, metadata=None)[source]#

Module contents#