Merge pull request #122 from Undertone0809/feat-v1.10.0

test: optimize tests

Makefile

@@ -4,15 +4,13 @@ OS := $(shell python -c "import sys; print(sys.platform)")
 ifeq ($(OS),win32)
 	PYTHONPATH := $(shell python -c "import os; print(os.getcwd())")
     TEST_COMMAND := set PYTHONPATH=$(PYTHONPATH) && poetry run pytest -c pyproject.toml --cov-report=html --cov=promptulate ./tests/test_chat.py ./tests/output_formatter
+	TEST_PROD_COMMAND := set PYTHONPATH=$(PYTHONPATH) && poetry run pytest -c pyproject.toml --cov-report=html --cov=promptulate tests
 else
 	PYTHONPATH := `pwd`
     TEST_COMMAND := PYTHONPATH=$(PYTHONPATH) poetry run pytest -c pyproject.toml --cov-report=html --cov=promptulate ./tests/test_chat.py ./tests/output_formatter
+	TEST_PROD_COMMAND := PYTHONPATH=$(PYTHONPATH) poetry run pytest -c pyproject.toml --cov-report=html --cov=promptulate tests
 endif
 
-#* Poetry
-.PHONY: poetry-download
-poetry-download:
-	pip install poetry
 
 #* Installation
 .PHONY: install
@@ -33,14 +31,18 @@ polish-codestyle:
 .PHONY: formatting
 formatting: polish-codestyle
 
-
-
 #* Linting
 .PHONY: test
 test:
 	$(TEST_COMMAND)
 	poetry run coverage-badge -o docs/images/coverage.svg -f
 
+#* Linting
+.PHONY: test-prod
+test-prod:
+	$(TEST_PROD_COMMAND)
+	poetry run coverage-badge -o docs/images/coverage.svg -f
+
 .PHONY: check-codestyle
 check-codestyle:
 	poetry run isort --diff --check-only --settings-path pyproject.toml promptulate tests example

docs/_coverpage.md

@@ -20,7 +20,7 @@
 
 <div class="title">Promptulate</div>
 
-> All you need is an elegant LLM development framework.
+> All you need is an elegant LLM Agent development framework.
 
 
 <p align="center">

docs/get_started/quick_start.md

@@ -193,7 +193,10 @@ if __name__ == "__main__":
 - Python >= 3.8
 - make
 
-> make 不是必须的，但是利用 makefile 的能力轻松集成运行 test、lint 等模块。
+> 本项目使用 make 进行项目配套设施的构建，通过 makefile 的能力轻松集成运行 test、lint 等模块，请确保你的电脑已经安装了 make。
+> 
+> [how to install and use make in windows?](https://stackoverflow.com/questions/32127524/how-to-install-and-use-make-in-windows)
+
 
 运行以下命令：
 
@@ -208,12 +211,8 @@ pip install poetry
 make install
 ```
 
-如果你没有安装 make，也可以使用如下方式安装：
 
-```shell
-pip install poetry
-poetry install
-```
+本项目使用配备代码语法检查工具，如果你想提交 pr，则需要在 commit 之前运行 `make polish-codestyle` 进行代码规范格式化，并且运行 `make lint` 通过语法与单元测试的检查。
 
 ## 更多
 

example/llm/custom_conversation.py

@@ -0,0 +1,18 @@
+from promptulate.llms import ChatOpenAI
+from promptulate.schema import AssistantMessage, MessageSet, SystemMessage, UserMessage
+
+
+def main():
+    messages = MessageSet(
+        messages=[
+            SystemMessage(content="You are a helpful assitant"),
+            UserMessage(content="Hello?"),
+        ]
+    )
+
+    llm = ChatOpenAI()
+    answer: AssistantMessage = llm.predict(messages)
+    print(answer.content)
+
+
+main()

example/llm/llm_private_key.py

@@ -0,0 +1,12 @@
+"""This example will show how to use a specified key in OpenAI model."""
+from promptulate.llms import ChatOpenAI
+
+
+def main():
+    llm = ChatOpenAI()
+    llm.set_private_api_key("your key here")
+    print(llm("hello"))
+
+
+if __name__ == "__main__":
+    main()

promptulate/client/chat.py

@@ -158,7 +158,7 @@ def chat():
     if args.proxy_mode:
         set_proxy_mode(args.proxy_mode)
 
-    print_text(f"Hi there, here is promptulate chat terminal.", "pink")
+    print_text("Hi there, here is promptulate chat terminal.", "pink")
 
     terminal_mode = questionary.select(
         "Choose a chat terminal:",

promptulate/llms/erniebot/erniebot.py

@@ -7,13 +7,7 @@
 
 from promptulate.config import Config
 from promptulate.llms import BaseLLM
-from promptulate.schema import (
-    AssistantMessage,
-    BaseMessage,
-    LLMType,
-    MessageSet,
-    UserMessage,
-)
+from promptulate.schema import AssistantMessage, LLMType, MessageSet, UserMessage
 from promptulate.tips import LLMError
 from promptulate.utils import get_logger
 

promptulate/tools/arxiv/api_wrapper.py

@@ -59,7 +59,7 @@ class Config:
     def validate_environment(cls, values: Dict) -> Dict:
         """Validate that python package exists in environment."""
         try:
-            import arxiv
+            import arxiv  # noqa
         except ImportError:
             raise ValueError(
                 "Could not import arxiv python package. "
@@ -83,10 +83,13 @@ def _query(
         """
         if not keyword:
             keyword = ""
+
         if not id_list:
             id_list = []
+
         if not num_results:
             num_results = self.max_num_of_result
+
         if isinstance(id_list, str):
             id_list = [id_list]
 

promptulate/tools/base.py

@@ -69,7 +69,7 @@ class Tool(ABC):
     description: str
     """Tool description"""
 
-    def __init__(self, **kwargs):
+    def __init__(self, *args, **kwargs):
         self.check_params()
         if "hooks" in kwargs and kwargs["hooks"]:
             for hook in kwargs["hooks"]:

promptulate/tools/duckduckgo/api_wrapper.py

@@ -21,7 +21,7 @@ class Config:
     def validate_environment(cls, values: Dict) -> Dict:
         """Validate that python package exists in environment."""
         try:
-            from duckduckgo_search import DDGS
+            from duckduckgo_search import DDGS  # noqa
         except ImportError:
             raise ValueError(
                 "Could not import duckduckgo-search python package. "

promptulate/tools/human_feedback/tools.py

@@ -4,8 +4,9 @@
 from promptulate.utils.color_print import print_text
 
 
-def _print_func(content) -> None:
-    print_text(f"[Agent ask] {content}", "blue")
+def _print_func(llm_question: str) -> None:
+    """Default way to show llm question when llm using HumanFeedBackTool."""
+    print_text(f"[Agent ask] {llm_question}", "blue")
 
 
 class HumanFeedBackTool(Tool):
@@ -14,22 +15,22 @@ class HumanFeedBackTool(Tool):
     name: str = "human_feedback"
     description: str = (
         "Human feedback tools are used to collect human feedback information."
-        "Please only use this tool in situations where relevant contextual information is lacking or reasoning cannot "
-        "continue."
-        "Please enter the content you wish for human feedback and interaction, but do not ask for knowledge or let "
-        "humans reason. "
+        "Please only use this tool in situations where relevant contextual information"
+        "is lacking or reasoning cannot continue. Please enter the content you wish for "
+        "human feedback and interaction, but do not ask for knowledge or let humans reason."
     )
 
     def __init__(
         self,
-        prompt_func: Callable[[str], None] = _print_func,
+        output_func: Callable[[str], None] = _print_func,
         input_func: Callable = input,
+        *args,
         **kwargs,
     ):
-        super().__init__(**kwargs)
-        self.prompt_func = prompt_func
+        super().__init__(*args, **kwargs)
+        self.output_func = output_func
         self.input_func = input_func
 
     def _run(self, content: str, *args, **kwargs) -> str:
-        self.prompt_func(content)
+        self.output_func(content)
         return self.input_func()

promptulate/tools/iot_swith_mqtt/tools.py

@@ -41,7 +41,7 @@ def __init__(
 
     def _run(self, question: str, *args, **kwargs) -> str:
         try:
-            import paho.mqtt.client as mqtt
+            import paho.mqtt.client as mqtt  # noqa
         except ImportError:
             raise ImportError(
                 "Could not import paho python package. "

promptulate/tools/paper/tools.py

@@ -141,7 +141,7 @@ def get_advice():
                 if paper_info:
                     paper_info = paper_info[0]
 
-            except NetWorkError as e:
+            except NetWorkError:
                 paper_info = self.arxiv_apiwrapper.query(
                     keyword=query, num_results=1, specified_fields=["title", "summary"]
                 )

promptulate/tools/semantic_scholar/api_wrapper.py

@@ -90,7 +90,7 @@ def get_detail():
             self.current_result = response.json()["matches"]
 
             if len(self.current_result) == 0:
-                logger.debug(f"[pne] semantic scholar return none")
+                logger.debug("[pne] semantic scholar return none")
                 return []
 
             for item in self.current_result:

promptulate/utils/logger.py

@@ -21,7 +21,7 @@
 import logging
 import os
 
-from promptulate import utils
+from promptulate.utils.core_utils import get_default_storage_path
 
 logger = logging.getLogger(__name__)
 
@@ -31,7 +31,7 @@ def get_logger():
 
 
 def get_default_log_path():
-    return utils.get_default_storage_path("log")
+    return get_default_storage_path("log")
 
 
 def get_log_name() -> str:

pyproject.toml

@@ -33,6 +33,7 @@ pytest-cov = "^3.0.0"
 coverage = "^6.1.2"
 pre-commit = "^3.5.0"
 coverage-badge = "^1.1.0"
+langchain = "^0.0.324"
 
 [tool.poetry.scripts]
 pne-chat = "promptulate.client.chat:main"

tests/framework/test_conversation.py

@@ -1,46 +1,46 @@
-from unittest import TestCase
+# from unittest import TestCase
 
-from promptulate.frameworks.conversation import Conversation
-from promptulate.memory import FileChatMemory
-from promptulate.utils.logger import enable_log, get_logger
+# from promptulate.frameworks.conversation import Conversation
+# from promptulate.memory import FileChatMemory
+# from promptulate.utils.logger import enable_log, get_logger
 
-enable_log()
-logger = get_logger()
+# enable_log()
+# logger = get_logger()
 
 
-class TestConversation(TestCase):
-    def test_predict(self):
-        conversation = Conversation()
-        result = conversation.run("什么是大语言模型")
-        self.assertIsNotNone(result)
-        self.assertTrue("大语言模型" in result)
+# class TestConversation(TestCase):
+#     def test_predict(self):
+#         conversation = Conversation()
+#         result = conversation.run("什么是大语言模型")
+#         self.assertIsNotNone(result)
+#         self.assertTrue("大语言模型" in result)
 
-    def test_predict_with_stop(self):
-        conversation = Conversation()
-        prompt = """
-        Please strictly output the following content.
-        ```
-        [start] This is a test [end]
-        ```
-        """
-        result = conversation.run(prompt, stop=["test"])
-        self.assertTrue("test [end]" not in result)
-        self.assertIsNotNone(result)
+#     def test_predict_with_stop(self):
+#         conversation = Conversation()
+#         prompt = """
+#         Please strictly output the following content.
+#         ```
+#         [start] This is a test [end]
+#         ```
+#         """
+#         result = conversation.run(prompt, stop=["test"])
+#         self.assertTrue("test [end]" not in result)
+#         self.assertIsNotNone(result)
 
-    def test_memory_with_buffer(self):
-        conversation = Conversation()
-        prompt = """给我想5个公司的名字"""
-        conversation.run(prompt)
-        conversation_id = conversation.conversation_id
-        new_conversation = Conversation(conversation_id=conversation_id)
-        new_conversation.predict("再给我五个")
+#     def test_memory_with_buffer(self):
+#         conversation = Conversation()
+#         prompt = """give me 5 company names"""
+#         conversation.run(prompt)
+#         conversation_id = conversation.conversation_id
+#         new_conversation = Conversation(conversation_id=conversation_id)
+#         new_conversation.predict("give me 5 more")
 
-    def test_memory_with_file(self):
-        conversation = Conversation(memory=FileChatMemory())
-        prompt = """给我想5个公司的名字"""
-        conversation.run(prompt)
-        conversation_id = conversation.conversation_id
-        new_conversation = Conversation(
-            conversation_id=conversation_id, memory=FileChatMemory()
-        )
-        new_conversation.predict("再给我五个")
+#     def test_memory_with_file(self):
+#         conversation = Conversation(memory=FileChatMemory())
+#         prompt = """give me 5 company names"""
+#         conversation.run(prompt)
+#         conversation_id = conversation.conversation_id
+#         new_conversation = Conversation(
+#             conversation_id=conversation_id, memory=FileChatMemory()
+#         )
+#         new_conversation.predict("give me 5 more")

tests/hook/test_agent.py

@@ -34,7 +34,7 @@ def handle_result(*args, **kwargs):
             self.assertIsNotNone(result)
             print(f"<instance> result: {result}")
 
-        hooks = [handle_result, handle_start, handle_result]
+        hooks = [handle_create, handle_start, handle_result]
         tools = [DuckDuckGoTool(), Calculator()]
         agent = ToolAgent(tools=tools, hooks=hooks)
         agent.run("What is promptulate?")
@@ -71,8 +71,9 @@ def handle_result(*args, **kwargs):
             self.assertIsNotNone(result)
             print(f"<component> result: {result}")
 
-        tool = DuckDuckGoTool()
-        tool.run("What is promptulate?")
+        tools = [DuckDuckGoTool(), Calculator()]
+        agent = ToolAgent(tools=tools)
+        agent.run("What is promptulate?")
 
         self.assertTrue(create_flag)
         self.assertTrue(start_flag)

tests/hook/test_tool.py

@@ -33,7 +33,7 @@ def handle_result(*args, **kwargs):
             self.assertIsNotNone(result)
             print(f"<instance> result: {result}")
 
-        hooks = [handle_result, handle_start, handle_result]
+        hooks = [handle_create, handle_start, handle_result]
         tool = DuckDuckGoTool(hooks=hooks)
         tool.run("What is LLM?")