Merge branch 'main' into main

.github/workflows/contrib-openai.yml

@@ -41,19 +41,23 @@ jobs:
           pip install -e .
           python -c "import autogen"
           pip install coverage pytest-asyncio
+      - name: Install PostgreSQL
+        run: |
+          sudo apt install postgresql -y
+      - name: Start PostgreSQL service
+        run: sudo service postgresql start
       - name: Install packages for test when needed
         run: |
           pip install docker
-          pip install qdrant_client[fastembed]
-          pip install -e .[retrievechat]
+          pip install -e .[retrievechat-qdrant,retrievechat-pgvector]
       - name: Coverage
         env:
           OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
           AZURE_OPENAI_API_KEY: ${{ secrets.AZURE_OPENAI_API_KEY }}
           AZURE_OPENAI_API_BASE: ${{ secrets.AZURE_OPENAI_API_BASE }}
           OAI_CONFIG_LIST: ${{ secrets.OAI_CONFIG_LIST }}
         run: |
-          coverage run -a -m pytest test/agentchat/contrib/test_retrievechat.py::test_retrievechat test/agentchat/contrib/test_qdrant_retrievechat.py::test_retrievechat
+          coverage run -a -m pytest test/agentchat/contrib/test_retrievechat.py::test_retrievechat test/agentchat/contrib/test_qdrant_retrievechat.py::test_retrievechat test/agentchat/contrib/test_pgvector_retrievechat.py::test_retrievechat
           coverage xml
       - name: Upload coverage to Codecov
         uses: codecov/codecov-action@v3

.github/workflows/contrib-tests.yml

@@ -42,16 +42,20 @@ jobs:
       - name: Install qdrant_client when python-version is 3.10
         if: matrix.python-version == '3.10'
         run: |
-          pip install qdrant_client[fastembed]
+          pip install .[retrievechat-qdrant]
       - name: Install unstructured when python-version is 3.9 and on linux
-        if: matrix.python-version == '3.9' && matrix.os == 'ubuntu-latest'
         run: |
           sudo apt-get update
           sudo apt-get install -y tesseract-ocr poppler-utils
           pip install unstructured[all-docs]==0.13.0
-      - name: Install packages and dependencies for RetrieveChat
+      - name: Install and Start PostgreSQL
+        runs-on: ubuntu-latest
         run: |
-          pip install -e .[retrievechat]
+          sudo apt install postgresql -y
+          sudo service postgresql start
+      - name: Install packages and dependencies for PGVector
+        run: |
+          pip install -e .[retrievechat-pgvector]
       - name: Set AUTOGEN_USE_DOCKER based on OS
         shell: bash
         run: |

.github/workflows/dotnet-build.yml

@@ -22,13 +22,15 @@ permissions:
 
 jobs:
   build:
-    name: Build
+    name: Dotnet Build
     runs-on: ubuntu-latest
     defaults:
       run:
         working-directory: dotnet
     steps:
     - uses: actions/checkout@v4
+      with:
+        lfs: true
     - name: Setup .NET
       uses: actions/setup-dotnet@v4
       with:
@@ -54,6 +56,8 @@ jobs:
     needs: build
     steps:
     - uses: actions/checkout@v4
+      with:
+        lfs: true
     - name: Setup .NET
       uses: actions/setup-dotnet@v4
       with:

.github/workflows/dotnet-release.yml

@@ -27,6 +27,8 @@ jobs:
         working-directory: dotnet
     steps:
     - uses: actions/checkout@v4
+      with:
+        lfs: true
     - name: Setup .NET
       uses: actions/setup-dotnet@v4
       with:
@@ -66,4 +68,4 @@ jobs:
         $version = $metaInfoContent | Select-String -Pattern "<VersionPrefix>(.*)</VersionPrefix>" | ForEach-Object { $_.Matches.Groups[1].Value }
         git tag -a "$version" -m "AutoGen.Net release $version"
         git push origin --tags
-      shell: pwsh
+      shell: pwsh

.github/workflows/lfs-check.yml

@@ -0,0 +1,15 @@
+name: "Git LFS Check"
+
+on: pull_request
+permissions: {}
+jobs:
+  lfs-check:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          lfs: true
+      - name: Check Git LFS files for consistency
+        run: |
+          git lfs fsck

.gitignore

@@ -188,3 +188,5 @@ local_cache
 
 notebook/result.png
 samples/apps/autogen-studio/autogenstudio/models/test/
+
+notebook/coding

.pre-commit-config.yaml

@@ -43,6 +43,7 @@ repos:
               website/static/img/ag.svg |
               website/yarn.lock |
               website/docs/tutorial/code-executors.ipynb |
+              website/docs/topics/code-execution/custom-executor.ipynb |
               website/docs/topics/non-openai-models/cloud-gemini.ipynb |
               notebook/.*
             )$

OAI_CONFIG_LIST_sample

@@ -5,7 +5,8 @@
 [
     {
         "model": "gpt-4",
-        "api_key": "<your OpenAI API key here>"
+        "api_key": "<your OpenAI API key here>",
+        "tags": ["gpt-4", "tool"]
     },
     {
         "model": "<your Azure OpenAI deployment name>",

README.md

@@ -14,7 +14,7 @@
     <img src="https://github.com/microsoft/autogen/blob/main/website/static/img/flaml.svg"  width=200>
     <br>
 </p> -->
-:fire: Mar 26, 2024: Andrew Ng gave a shoutout to AutoGen in [What's next for AI agentic workflows](https://youtu.be/sal78ACtGTc?si=JduUzN_1kDnMq0vF) at Sequoia Capital's AI Ascent.
+:fire: Apr 17, 2024: Andrew Ng cited AutoGen in [The Batch newsletter](https://www.deeplearning.ai/the-batch/issue-245/) and [What's next for AI agentic workflows](https://youtu.be/sal78ACtGTc?si=JduUzN_1kDnMq0vF) at Sequoia Capital's AI Ascent (Mar 26).
 
 :fire: Mar 3, 2024: What's new in AutoGen? 📰[Blog](https://microsoft.github.io/autogen/blog/2024/03/03/AutoGen-Update); 📺[Youtube](https://www.youtube.com/watch?v=j_mtwQiaLGU).
 

autogen/agentchat/chat.py

@@ -171,6 +171,9 @@ def initiate_chats(chat_queue: List[Dict[str, Any]]) -> List[ChatResult]:
             - `"carryover"` - It can be used to specify the carryover information to be passed
                to this chat. If provided, we will combine this carryover with the "message" content when
                generating the initial chat message in `generate_init_message`.
+            - `"finished_chat_indexes_to_exclude_from_carryover"` - It can be used by specifying a list of indexes of the finished_chats list,
+               from which to exclude the summaries for carryover. If 'finished_chat_indexes_to_exclude_from_carryover' is not provided or an empty list,
+               then summary from all the finished chats will be taken.
     Returns:
         (list): a list of ChatResult objects corresponding to the finished chats in the chat_queue.
     """
@@ -182,9 +185,16 @@ def initiate_chats(chat_queue: List[Dict[str, Any]]) -> List[ChatResult]:
     while current_chat_queue:
         chat_info = current_chat_queue.pop(0)
         _chat_carryover = chat_info.get("carryover", [])
+        finished_chat_indexes_to_exclude_from_carryover = chat_info.get(
+            "finished_chat_indexes_to_exclude_from_carryover", []
+        )
+
         if isinstance(_chat_carryover, str):
             _chat_carryover = [_chat_carryover]
-        chat_info["carryover"] = _chat_carryover + [r.summary for r in finished_chats]
+        chat_info["carryover"] = _chat_carryover + [
+            r.summary for i, r in enumerate(finished_chats) if i not in finished_chat_indexes_to_exclude_from_carryover
+        ]
+
         __post_carryover_processing(chat_info)
         sender = chat_info["sender"]
         chat_res = sender.initiate_chat(**chat_info)

autogen/agentchat/contrib/capabilities/teachability.py

@@ -86,7 +86,7 @@ def prepopulate_db(self):
         """Adds a few arbitrary memos to the DB."""
         self.memo_store.prepopulate()
 
-    def process_last_received_message(self, text):
+    def process_last_received_message(self, text: Union[Dict, str]):
         """
         Appends any relevant memos to the message text, and stores any apparent teachings in new memos.
         Uses TextAnalyzerAgent to make decisions about memo storage and retrieval.
@@ -103,7 +103,7 @@ def process_last_received_message(self, text):
         # Return the (possibly) expanded message text.
         return expanded_text
 
-    def _consider_memo_storage(self, comment):
+    def _consider_memo_storage(self, comment: Union[Dict, str]):
         """Decides whether to store something from one user comment in the DB."""
         memo_added = False
 
@@ -161,7 +161,7 @@ def _consider_memo_storage(self, comment):
             # Yes. Save them to disk.
             self.memo_store._save_memos()
 
-    def _consider_memo_retrieval(self, comment):
+    def _consider_memo_retrieval(self, comment: Union[Dict, str]):
         """Decides whether to retrieve memos from the DB, and add them to the chat context."""
 
         # First, use the comment directly as the lookup key.
@@ -195,7 +195,7 @@ def _consider_memo_retrieval(self, comment):
         # Append the memos to the text of the last message.
         return comment + self._concatenate_memo_texts(memo_list)
 
-    def _retrieve_relevant_memos(self, input_text):
+    def _retrieve_relevant_memos(self, input_text: str) -> list:
         """Returns semantically related memos from the DB."""
         memo_list = self.memo_store.get_related_memos(
             input_text, n_results=self.max_num_retrievals, threshold=self.recall_threshold
@@ -213,7 +213,7 @@ def _retrieve_relevant_memos(self, input_text):
         memo_list = [memo[1] for memo in memo_list]
         return memo_list
 
-    def _concatenate_memo_texts(self, memo_list):
+    def _concatenate_memo_texts(self, memo_list: list) -> str:
         """Concatenates the memo texts into a single string for inclusion in the chat context."""
         memo_texts = ""
         if len(memo_list) > 0:
@@ -225,7 +225,7 @@ def _concatenate_memo_texts(self, memo_list):
             memo_texts = memo_texts + "\n" + info
         return memo_texts
 
-    def _analyze(self, text_to_analyze, analysis_instructions):
+    def _analyze(self, text_to_analyze: Union[Dict, str], analysis_instructions: Union[Dict, str]):
         """Asks TextAnalyzerAgent to analyze the given text according to specific instructions."""
         self.analyzer.reset()  # Clear the analyzer's list of messages.
         self.teachable_agent.send(
@@ -246,10 +246,16 @@ class MemoStore:
     Vector embeddings are currently supplied by Chroma's default Sentence Transformers.
     """
 
-    def __init__(self, verbosity, reset, path_to_db_dir):
+    def __init__(
+        self,
+        verbosity: Optional[int] = 0,
+        reset: Optional[bool] = False,
+        path_to_db_dir: Optional[str] = "./tmp/teachable_agent_db",
+    ):
         """
         Args:
             - verbosity (Optional, int): 1 to print memory operations, 0 to omit them. 3+ to print memo lists.
+            - reset (Optional, bool): True to clear the DB before starting. Default False.
             - path_to_db_dir (Optional, str): path to the directory where the DB is stored.
         """
         self.verbosity = verbosity
@@ -304,7 +310,7 @@ def reset_db(self):
         self.uid_text_dict = {}
         self._save_memos()
 
-    def add_input_output_pair(self, input_text, output_text):
+    def add_input_output_pair(self, input_text: str, output_text: str):
         """Adds an input-output pair to the vector DB."""
         self.last_memo_id += 1
         self.vec_db.add(documents=[input_text], ids=[str(self.last_memo_id)])
@@ -321,7 +327,7 @@ def add_input_output_pair(self, input_text, output_text):
         if self.verbosity >= 3:
             self.list_memos()
 
-    def get_nearest_memo(self, query_text):
+    def get_nearest_memo(self, query_text: str):
         """Retrieves the nearest memo to the given query text."""
         results = self.vec_db.query(query_texts=[query_text], n_results=1)
         uid, input_text, distance = results["ids"][0][0], results["documents"][0][0], results["distances"][0][0]
@@ -338,7 +344,7 @@ def get_nearest_memo(self, query_text):
             )
         return input_text, output_text, distance
 
-    def get_related_memos(self, query_text, n_results, threshold):
+    def get_related_memos(self, query_text: str, n_results: int, threshold: Union[int, float]):
         """Retrieves memos that are related to the given query text within the specified distance threshold."""
         if n_results > len(self.uid_text_dict):
             n_results = len(self.uid_text_dict)