improve unit tests

.env.example

@@ -67,6 +67,9 @@ LLAMACPP_DOWNLOAD_LINK=https://huggingface.co/TheBloke/Llama-2-7B-Chat-GGML/reso
 # CHROMADB_INDEX_PATH="./data/chromadb_1024_512/"
 FAISS_INDEX_PATH="./data/faiss_1024_512/"
 
+CHAT_QUESTION="What's the capital city of Malaysia?"
+QA_QUESTION="What's deep learning?"
+
 QUESTIONS_FILE_PATH="./data/questions.txt"
 
 TOKENIZERS_PARALLELISM=true

Makefile

@@ -10,7 +10,7 @@ else
 endif
 
 test:
-	PYTORCH_MPS_HIGH_WATERMARK_RATIO=0.0 python test.py
+	PYTORCH_MPS_HIGH_WATERMARK_RATIO=0.0 python test.py $(TEST)
 
 chat:
 	python test.py chat

test.py

@@ -14,7 +14,7 @@ from app_modules.utils import get_device_types, print_llm_response
 
 
 class TestLLMLoader(unittest.TestCase):
-    question = "What's the capital city of Malaysia?"
+    question = os.environ.get("CHAT_QUESTION")
 
     def run_test_case(self, llm_model_type, query):
         n_threds = int(os.environ.get("NUMBER_OF_CPU_CORES") or "4")
@@ -52,7 +52,7 @@ class TestLLMLoader(unittest.TestCase):
 
 
 class TestChatChain(unittest.TestCase):
-    question = "What's the capital city of Malaysia?"
+    question = os.environ.get("CHAT_QUESTION")
 
     def run_test_case(self, llm_model_type, query):
         n_threds = int(os.environ.get("NUMBER_OF_CPU_CORES") or "4")
@@ -97,7 +97,7 @@ class TestChatChain(unittest.TestCase):
 
 class TestQAChain(unittest.TestCase):
     qa_chain: any
-    question = "What's deep learning?"
+    question = os.environ.get("QA_QUESTION")
 
     def run_test_case(self, llm_model_type, query):
         start = timer()