Functions — github.com/QwenLM/Qwen3Guard

CodeHub MCPcopy Create free account

hub / github.com/QwenLM/Qwen3Guard / functions

Functions15 in github.com/QwenLM/Qwen3Guard

⨍ Functions 15 ◇ Types & classes 2

↓ 2 callersMethodcalculate_metrics

Calculates precision, recall, and F1 for the 'Unsafe' category.

eval/eval_stream.py:148

↓ 1 callersMethod_find_last_user_content_index

(self, tokens_list)

eval/eval_stream.py:120

↓ 1 callersMethod_get_model_prompt_predictions

(self, model_inputs, split_idx, input_ids)

eval/eval_stream.py:90

↓ 1 callersMethod_get_model_response_predictions

(self, model_inputs, split_idx, input_ids)

eval/eval_stream.py:59

↓ 1 callersMethodconsecutive_unsafe

(self, pred_list: List[str])

eval/eval_stream.py:139

↓ 1 callersMethodevaluate_f1

(self, file_path: str, data_type: str)

eval/eval_stream.py:162

↓ 1 callersMethodevaluate_unsafe_latency

(self, file_path: str)

eval/eval_stream.py:193

↓ 1 callersFunctionload_input_data

Load input data from either a local JSONL file or a Hugging Face dataset, and filter to keep only items where type == "thinking+response".

eval/eval_gen.py:9

↓ 1 callersFunctionmain

eval/eval_stream.py:257

↓ 1 callersFunctionmain

(model_path, input_file, output_file)

eval/eval_gen.py:32

↓ 1 callersMethodprocess_conversation_prompt

(self, messages)

eval/eval_stream.py:48

↓ 1 callersMethodprocess_conversation_response

(self, messages, thinking=False)

eval/eval_stream.py:30

(self, model_path: str)

eval/eval_stream.py:15

(self, classification_strategy="consecutive_unsafe")

eval/eval_stream.py:132

Methodprocess_bins

eval/eval_stream.py:194