Merge branch 'master' into dockerfile-improvement

camel-ai · Jul 23, 2024 · 1dc2c62 · 1dc2c62
2 parents 9b44bb8 + fd944b7
commit 1dc2c62
Show file tree

Hide file tree

Showing 47 changed files with 1,861 additions and 121 deletions.
diff --git a/.github/workflows/build_package.yml b/.github/workflows/build_package.yml
@@ -35,4 +35,8 @@ jobs:
           OPENWEATHERMAP_API_KEY: "${{ secrets.OPENWEATHERMAP_API_KEY }}"
           ANTHROPIC_API_KEY: "${{ secrets.ANTHROPIC_API_KEY }}"
           COHERE_API_KEY: "${{ secrets.COHERE_API_KEY }}"
+          AZURE_OPENAI_API_KEY: ${{ secrets.AZURE_OPENAI_API_KEY }}"
+          AZURE_API_VERSION: ${{ secrets.AZURE_API_VERSION }}"
+          AZURE_DEPLOYMENT_NAME: ${{ secrets.AZURE_DEPLOYMENT_NAME }}"
+          AZURE_OPENAI_ENDPOINT: ${{ secrets.AZURE_OPENAI_ENDPOINT }}"
         run: pytest --fast-test-mode ./test
diff --git a/.github/workflows/pytest_package.yml b/.github/workflows/pytest_package.yml
@@ -33,6 +33,10 @@ jobs:
         NVIDIA_API_KEY: "${{ secrets.NVIDIA_API_KEY }}"
         ZHIPUAI_API_BASE_URL: "${{ secrets.ZHIPUAI_API_BASE_URL }}"
         ZHIPUAI_API_KEY: "${{ secrets.ZHIPUAI_API_KEY }}"
+        AZURE_OPENAI_API_KEY: "to-be-filled"
+        AZURE_API_VERSION: "to-be-filled"
+        AZURE_DEPLOYMENT_NAME: "to-be-filled"
+        AZURE_OPENAI_ENDPOINT: "https://camel.openai.azure.com/"
       run: poetry run pytest --fast-test-mode test/
 
   pytest_package_llm_test:
@@ -55,6 +59,10 @@ jobs:
         NVIDIA_API_KEY: "${{ secrets.NVIDIA_API_KEY }}"
         ZHIPUAI_API_BASE_URL: "${{ secrets.ZHIPUAI_API_BASE_URL }}"
         ZHIPUAI_API_KEY: "${{ secrets.ZHIPUAI_API_KEY }}"
+        AZURE_OPENAI_API_KEY: "to-be-filled"
+        AZURE_API_VERSION: "to-be-filled"
+        AZURE_DEPLOYMENT_NAME: "to-be-filled"
+        AZURE_OPENAI_ENDPOINT: "https://camel.openai.azure.com/"
       run: poetry run pytest --llm-test-only test/
 
   pytest_package_very_slow_test:
@@ -77,4 +85,8 @@ jobs:
         NVIDIA_API_KEY: "${{ secrets.NVIDIA_API_KEY }}"
         ZHIPUAI_API_BASE_URL: "${{ secrets.ZHIPUAI_API_BASE_URL }}"
         ZHIPUAI_API_KEY: "${{ secrets.ZHIPUAI_API_KEY }}"
+        AZURE_OPENAI_API_KEY: ${{ secrets.AZURE_OPENAI_API_KEY }}"
+        AZURE_API_VERSION: ${{ secrets.AZURE_API_VERSION }}"
+        AZURE_DEPLOYMENT_NAME: ${{ secrets.AZURE_DEPLOYMENT_NAME }}"
+        AZURE_OPENAI_ENDPOINT: ${{ secrets.AZURE_OPENAI_ENDPOINT }}"
       run: poetry run pytest --very-slow-test-only test/
diff --git a/README.md b/README.md
@@ -236,6 +236,41 @@ Please note that the environment variable is session-specific. If you open a new
     print(assistant_response.msg.content)
     ```
 
+## Use Open-Source Models as Backends (ex. using vLLM to set Phi-3 locally)
+- [Install vLLM](https://docs.vllm.ai/en/latest/getting_started/installation.html)
+- After setting up vLLM, start an OpenAI compatible server for example by
+    ```bash
+    python -m vllm.entrypoints.openai.api_server --model microsoft/Phi-3-mini-4k-instruct --api-key vllm --dtype bfloat16
+    ```
+- Create and run following script (more details please refer to this [example](https://github.com/camel-ai/camel/blob/master/examples/models/vllm_model_example.py))
+    ```python
+    from camel.agents import ChatAgent
+    from camel.messages import BaseMessage
+    from camel.models import ModelFactory
+    from camel.types import ModelPlatformType
+
+    vllm_model = ModelFactory.create(
+        model_platform=ModelPlatformType.VLLM,
+        model_type="microsoft/Phi-3-mini-4k-instruct",
+        url="http://localhost:8000/v1",
+        model_config_dict={"temperature": 0.0},
+        api_key="vllm",
+    )
+
+    assistant_sys_msg = BaseMessage.make_assistant_message(
+        role_name="Assistant",
+        content="You are a helpful assistant.",
+    )
+    agent = ChatAgent(assistant_sys_msg, model=vllm_model, token_limit=4096)
+
+    user_msg = BaseMessage.make_user_message(
+        role_name="User",
+        content="Say hi to CAMEL AI",
+    )
+    assistant_response = agent.step(user_msg)
+    print(assistant_response.msg.content)
+    ```
+
 ## Data (Hosted on Hugging Face)
 | Dataset        | Chat format                                                                                         | Instruction format                                                                                               | Chat format (translated)                                                                   |
 |----------------|-----------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------|

diff --git a/camel/agents/chat_agent.py b/camel/agents/chat_agent.py
@@ -127,7 +127,7 @@ def __init__(
             if model is not None
             else ModelFactory.create(
                 model_platform=ModelPlatformType.OPENAI,
-                model_type=ModelType.GPT_3_5_TURBO,
+                model_type=ModelType.GPT_4O_MINI,
                 model_config_dict=ChatGPTConfig().__dict__,
                 api_key=self._api_key,
             )

diff --git a/camel/agents/critic_agent.py b/camel/agents/critic_agent.py
@@ -33,7 +33,7 @@ class CriticAgent(ChatAgent):
             agent.
         model (BaseModelBackend, optional): The model backend to use for
             generating responses. (default: :obj:`OpenAIModel` with
-            `GPT_3_5_TURBO`)
+            `GPT_4O_MINI`)
         message_window_size (int, optional): The maximum number of previous
             messages to include in the context window. If `None`, no windowing
             is performed. (default: :obj:`6`)

diff --git a/camel/agents/deductive_reasoner_agent.py b/camel/agents/deductive_reasoner_agent.py
@@ -35,7 +35,7 @@ class DeductiveReasonerAgent(ChatAgent):
     Args:
         model (BaseModelBackend, optional): The model backend to use for
             generating responses. (default: :obj:`OpenAIModel` with
-            `GPT_3_5_TURBO`)
+            `GPT_4O_MINI`)
     """
 
     def __init__(

diff --git a/camel/agents/embodied_agent.py b/camel/agents/embodied_agent.py
@@ -35,7 +35,7 @@ class EmbodiedAgent(ChatAgent):
         system_message (BaseMessage): The system message for the chat agent.
         model (BaseModelBackend, optional): The model backend to use for
             generating responses. (default: :obj:`OpenAIModel` with
-            `GPT_3_5_TURBO`)
+            `GPT_4O_MINI`)
         message_window_size (int, optional): The maximum number of previous
             messages to include in the context window. If `None`, no windowing
             is performed. (default: :obj:`None`)

diff --git a/camel/agents/knowledge_graph_agent.py b/camel/agents/knowledge_graph_agent.py
@@ -115,7 +115,7 @@ def __init__(
         Args:
         model (BaseModelBackend, optional): The model backend to use for
             generating responses. (default: :obj:`OpenAIModel` with
-            `GPT_3_5_TURBO`)
+            `GPT_4O_MINI`)
         """
         system_message = BaseMessage(
             role_name="Graphify",

diff --git a/camel/agents/role_assignment_agent.py b/camel/agents/role_assignment_agent.py
@@ -27,7 +27,7 @@ class RoleAssignmentAgent(ChatAgent):
     Args:
         model (BaseModelBackend, optional): The model backend to use for
             generating responses. (default: :obj:`OpenAIModel` with
-            `GPT_3_5_TURBO`)
+            `GPT_4O_MINI`)
 
     Attributes:
         role_assignment_prompt (TextPrompt): A prompt for the agent to generate

diff --git a/camel/agents/search_agent.py b/camel/agents/search_agent.py
@@ -26,10 +26,9 @@ class SearchAgent(ChatAgent):
     relevance of an answer.
 
     Args:
-        model_type (ModelType, optional): The type of model to use for the
-            agent. (default: :obj:`ModelType.GPT_3_5_TURBO`)
-        model_config (Any, optional): The configuration for the model.
-            (default: :obj:`None`)
+        model (BaseModelBackend, optional): The model backend to use for
+            generating responses. (default: :obj:`OpenAIModel` with
+            `GPT_4O_MINI`)
     """
 
     def __init__(

diff --git a/camel/agents/task_agent.py b/camel/agents/task_agent.py
@@ -32,7 +32,7 @@ class TaskSpecifyAgent(ChatAgent):
     Args:
         model (BaseModelBackend, optional): The model backend to use for
             generating responses. (default: :obj:`OpenAIModel` with
-            `GPT_3_5_TURBO`)
+            `GPT_4O_MINI`)
         task_type (TaskType, optional): The type of task for which to generate
             a prompt. (default: :obj:`TaskType.AI_SOCIETY`)
         task_specify_prompt (Union[str, TextPrompt], optional): The prompt for
@@ -126,7 +126,7 @@ class TaskPlannerAgent(ChatAgent):
     Args:
         model (BaseModelBackend, optional): The model backend to use for
             generating responses. (default: :obj:`OpenAIModel` with
-            `GPT_3_5_TURBO`)
+            `GPT_4O_MINI`)
         output_language (str, optional): The language to be output by the
             agent. (default: :obj:`None`)
     """
@@ -201,7 +201,7 @@ class TaskCreationAgent(ChatAgent):
             perform the task.
         model (BaseModelBackend, optional): The LLM backend to use for
             generating responses. (default: :obj:`OpenAIModel` with
-            `GPT_3_5_TURBO`)
+            `GPT_4O_MINI`)
         output_language (str, optional): The language to be output by the
             agent. (default: :obj:`None`)
         message_window_size (int, optional): The maximum number of previous
@@ -312,7 +312,7 @@ class TaskPrioritizationAgent(ChatAgent):
             perform the task.
         model (BaseModelBackend, optional): The LLM backend to use for
             generating responses. (default: :obj:`OpenAIModel` with
-            `GPT_3_5_TURBO`)
+            `GPT_4O_MINI`)
         output_language (str, optional): The language to be output by the
             agent. (default: :obj:`None`)
         message_window_size (int, optional): The maximum number of previous

diff --git a/camel/configs/__init__.py b/camel/configs/__init__.py
@@ -24,6 +24,7 @@
     ChatGPTConfig,
     OpenSourceConfig,
 )
+from .vllm_config import VLLM_API_PARAMS, VLLMConfig
 from .zhipuai_config import ZHIPUAI_API_PARAMS, ZhipuAIConfig
 
 __all__ = [
@@ -41,4 +42,6 @@
     'ZHIPUAI_API_PARAMS',
     'GeminiConfig',
     'Gemini_API_PARAMS',
+    'VLLMConfig',
+    'VLLM_API_PARAMS',
 ]
diff --git a/camel/configs/vllm_config.py b/camel/configs/vllm_config.py
@@ -0,0 +1,103 @@
+# =========== Copyright 2023 @ CAMEL-AI.org. All Rights Reserved. ===========
+# Licensed under the Apache License, Version 2.0 (the “License”);
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an “AS IS” BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# =========== Copyright 2023 @ CAMEL-AI.org. All Rights Reserved. ===========
+from __future__ import annotations
+
+from dataclasses import asdict, dataclass, field
+from typing import Sequence
+
+from openai._types import NOT_GIVEN, NotGiven
+
+from camel.configs.base_config import BaseConfig
+
+
+# flake8: noqa: E501
+@dataclass(frozen=True)
+class VLLMConfig(BaseConfig):
+    r"""Defines the parameters for generating chat completions using the
+    OpenAI API.
+
+    Reference: https://docs.vllm.ai/en/latest/serving/openai_compatible_server.html
+
+    Args:
+        temperature (float, optional): Sampling temperature to use, between
+            :obj:`0` and :obj:`2`. Higher values make the output more random,
+            while lower values make it more focused and deterministic.
+            (default: :obj:`0.2`)
+        top_p (float, optional): An alternative to sampling with temperature,
+            called nucleus sampling, where the model considers the results of
+            the tokens with top_p probability mass. So :obj:`0.1` means only
+            the tokens comprising the top 10% probability mass are considered.
+            (default: :obj:`1.0`)
+        n (int, optional): How many chat completion choices to generate for
+            each input message. (default: :obj:`1`)
+        response_format (object, optional): An object specifying the format
+            that the model must output. Compatible with GPT-4 Turbo and all
+            GPT-3.5 Turbo models newer than gpt-3.5-turbo-1106. Setting to
+            {"type": "json_object"} enables JSON mode, which guarantees the
+            message the model generates is valid JSON. Important: when using
+            JSON mode, you must also instruct the model to produce JSON
+            yourself via a system or user message. Without this, the model
+            may generate an unending stream of whitespace until the generation
+            reaches the token limit, resulting in a long-running and seemingly
+            "stuck" request. Also note that the message content may be
+            partially cut off if finish_reason="length", which indicates the
+            generation exceeded max_tokens or the conversation exceeded the
+            max context length.
+        stream (bool, optional): If True, partial message deltas will be sent
+            as data-only server-sent events as they become available.
+            (default: :obj:`False`)
+        stop (str or list, optional): Up to :obj:`4` sequences where the API
+            will stop generating further tokens. (default: :obj:`None`)
+        max_tokens (int, optional): The maximum number of tokens to generate
+            in the chat completion. The total length of input tokens and
+            generated tokens is limited by the model's context length.
+            (default: :obj:`None`)
+        presence_penalty (float, optional): Number between :obj:`-2.0` and
+            :obj:`2.0`. Positive values penalize new tokens based on whether
+            they appear in the text so far, increasing the model's likelihood
+            to talk about new topics. See more information about frequency and
+            presence penalties. (default: :obj:`0.0`)
+        frequency_penalty (float, optional): Number between :obj:`-2.0` and
+            :obj:`2.0`. Positive values penalize new tokens based on their
+            existing frequency in the text so far, decreasing the model's
+            likelihood to repeat the same line verbatim. See more information
+            about frequency and presence penalties. (default: :obj:`0.0`)
+        logit_bias (dict, optional): Modify the likelihood of specified tokens
+            appearing in the completion. Accepts a json object that maps tokens
+            (specified by their token ID in the tokenizer) to an associated
+            bias value from :obj:`-100` to :obj:`100`. Mathematically, the bias
+            is added to the logits generated by the model prior to sampling.
+            The exact effect will vary per model, but values between:obj:` -1`
+            and :obj:`1` should decrease or increase likelihood of selection;
+            values like :obj:`-100` or :obj:`100` should result in a ban or
+            exclusive selection of the relevant token. (default: :obj:`{}`)
+        user (str, optional): A unique identifier representing your end-user,
+            which can help OpenAI to monitor and detect abuse.
+            (default: :obj:`""`)
+    """
+
+    temperature: float = 0.2  # openai default: 1.0
+    top_p: float = 1.0
+    n: int = 1
+    stream: bool = False
+    stop: str | Sequence[str] | NotGiven = NOT_GIVEN
+    max_tokens: int | NotGiven = NOT_GIVEN
+    presence_penalty: float = 0.0
+    response_format: dict | NotGiven = NOT_GIVEN
+    frequency_penalty: float = 0.0
+    logit_bias: dict = field(default_factory=dict)
+    user: str = ""
+
+
+VLLM_API_PARAMS = {param for param in asdict(VLLMConfig()).keys()}
diff --git a/camel/interpreters/__init__.py b/camel/interpreters/__init__.py
@@ -16,6 +16,7 @@
 from .docker_interpreter import DockerInterpreter
 from .internal_python_interpreter import InternalPythonInterpreter
 from .interpreter_error import InterpreterError
+from .ipython_interpreter import JupyterKernelInterpreter
 from .subprocess_interpreter import SubprocessInterpreter
 
 __all__ = [
@@ -24,4 +25,5 @@
     'InternalPythonInterpreter',
     'SubprocessInterpreter',
     'DockerInterpreter',
+    'JupyterKernelInterpreter',
 ]