bump version to 0.1.14

.github/workflows/openai.yml

@@ -57,9 +57,9 @@ jobs:
         run: |
           pip install -e .[teachable]
       - name: Install packages for RetrieveChat with QDrant when needed
-        if: matrix.python-version == '3.9'
+        if: matrix.python-version == '3.11'
         run: |
-          pip install qdrant_client[fastembed]
+          pip install -e .[retrievechat] qdrant_client[fastembed]
       - name: Coverage
         if: matrix.python-version == '3.9'
         env:
@@ -80,6 +80,7 @@ jobs:
           OAI_CONFIG_LIST: ${{ secrets.OAI_CONFIG_LIST }}
         run: |
           pip install nbconvert nbformat ipykernel
+          coverage run -a -m pytest test/agentchat/test_qdrant_retrievechat.py
           coverage run -a -m pytest test/test_with_openai.py
           coverage run -a -m pytest test/test_notebook.py
           coverage xml

autogen/retrieve_utils.py

@@ -1,9 +1,8 @@
-from typing import List, Union, Dict, Tuple, Callable
+from typing import List, Union, Callable
 import os
 import requests
 from urllib.parse import urlparse
 import glob
-import tiktoken
 import chromadb
 
 if chromadb.__version__ < "0.4.15":

autogen/version.py

@@ -1 +1 @@
-__version__ = "0.1.13"
+__version__ = "0.1.14"

setup.py

@@ -57,7 +57,7 @@
         ],
         "blendsearch": ["flaml[blendsearch]"],
         "mathchat": ["sympy", "pydantic==1.10.9", "wolframalpha"],
-        "retrievechat": ["chromadb", "tiktoken", "sentence_transformers", "pypdf"],
+        "retrievechat": ["chromadb", "tiktoken", "sentence_transformers", "pypdf", "ipython"],
         "teachable": ["chromadb"],
     },
     classifiers=[

test/agentchat/test_retrievechat.py

@@ -10,7 +10,6 @@
     from autogen.agentchat.contrib.retrieve_user_proxy_agent import (
         RetrieveUserProxyAgent,
     )
-    from autogen.retrieve_utils import create_vector_db_from_dir, query_vector_db
     import chromadb
     from chromadb.utils import embedding_functions as ef
 
@@ -61,6 +60,7 @@ def test_retrievechat():
             "model": config_list[0]["model"],
             "client": chromadb.PersistentClient(path="/tmp/chromadb"),
             "embedding_function": sentence_transformer_ef,
+            "get_or_create": True,
         },
     )
 
@@ -72,26 +72,5 @@ def test_retrievechat():
     print(conversations)
 
 
-@pytest.mark.skipif(
-    sys.platform in ["darwin", "win32"] or skip_test,
-    reason="do not run on MacOS or windows",
-)
-def test_retrieve_utils():
-    client = chromadb.PersistentClient(path="/tmp/chromadb")
-    create_vector_db_from_dir(dir_path="./website/docs", client=client, collection_name="autogen-docs")
-    results = query_vector_db(
-        query_texts=[
-            "How can I use AutoGen UserProxyAgent and AssistantAgent to do code generation?",
-        ],
-        n_results=4,
-        client=client,
-        collection_name="autogen-docs",
-        search_string="AutoGen",
-    )
-    print(results["ids"][0])
-    assert len(results["ids"][0]) == 4
-
-
 if __name__ == "__main__":
     test_retrievechat()
-    test_retrieve_utils()

test/oai/test_completion.py

@@ -231,7 +231,7 @@ def test_humaneval(num_samples=1):
         raise_on_ratelimit_or_timeout=False,
     )
     # assert response == -1
-    config_list = autogen.config_list_openai_aoai(KEY_LOC, exclude="aoai")
+    config_list = autogen.config_list_openai_aoai(KEY_LOC)
     # a minimal tuning example
     config, _ = autogen.Completion.tune(
         data=tune_data,
@@ -376,11 +376,11 @@ def test_math(num_samples=-1):
     ]
 
     autogen.Completion.set_cache(seed)
-    config_list = autogen.config_list_openai_aoai(KEY_LOC)[:2]
+    config_list = autogen.config_list_openai_aoai(KEY_LOC)
     vanilla_config = {
-        "model": "text-davinci-003",
+        "model": "text-ada-001",
         "temperature": 1,
-        "max_tokens": 2048,
+        "max_tokens": 1024,
         "n": 1,
         "prompt": prompts[0],
         "stop": "###",
@@ -451,5 +451,5 @@ def my_average(results):
     # test_chatcompletion()
     # test_multi_model()
     # test_nocontext()
-    test_humaneval(1)
-    # test_math(1)
+    # test_humaneval(1)
+    test_math(1)

test/test_retrieve_utils.py

@@ -7,19 +7,15 @@
     extract_text_from_pdf,
     split_files_to_chunks,
     get_files_from_dir,
-    get_file_from_url,
     is_url,
     create_vector_db_from_dir,
     query_vector_db,
-    TEXT_FORMATS,
 )
 from autogen.token_count_utils import count_token
 
 import os
-import sys
 import pytest
 import chromadb
-import tiktoken
 
 
 test_dir = os.path.join(os.path.dirname(__file__), "test_files")
@@ -157,13 +153,29 @@ def custom_text_split_function(text):
             client=client,
             collection_name="mytestcollection",
             custom_text_split_function=custom_text_split_function,
+            get_or_create=True,
         )
         results = query_vector_db(["autogen"], client=client, collection_name="mytestcollection", n_results=1)
         assert (
             results.get("documents")[0][0]
             == "AutoGen is an advanced tool designed to assist developers in harnessing the capabilities\nof Large Language Models (LLMs) for various applications. The primary purpose o"
         )
 
+    def test_retrieve_utils(self):
+        client = chromadb.PersistentClient(path="/tmp/chromadb")
+        create_vector_db_from_dir(dir_path="./website/docs", client=client, collection_name="autogen-docs")
+        results = query_vector_db(
+            query_texts=[
+                "How can I use AutoGen UserProxyAgent and AssistantAgent to do code generation?",
+            ],
+            n_results=4,
+            client=client,
+            collection_name="autogen-docs",
+            search_string="AutoGen",
+        )
+        print(results["ids"][0])
+        assert len(results["ids"][0]) == 4
+
 
 if __name__ == "__main__":
     pytest.main()