Merge branch 'main' into falkordbrmcreation

.github/workflows/build_and_release.yml

@@ -84,15 +84,19 @@ jobs:
         run: python3 setup.py sdist bdist_wheel
       - name: Publish distribution 📦 to PyPI (dspy)
         uses: pypa/gh-action-pypi-publish@release/v1 # This requires a trusted publisher to be setup in pypi
-      # Publish to dspy-ai
+        with: 
+          attestations: false
+        # Publish to dspy-ai
       - name: Update version in setup.py (dspy-ai)
         run: sed -i '/#replace_package_version_marker/{n;s/version="[^"]*"/version="${{ needs.extract-tag.outputs.version }}"/;}' ./dspy/.internal_dspyai/setup.py
       - name: Update package name in setup.py  
         run: sed -i '/#replace_package_name_marker/{n;s/name="[^"]*"/name="dspy-ai"/;}' ./dspy/.internal_dspyai/setup.py
       - name: Update dspy dependency version in setup.py  
         run: |
-          sed -i '/#replace_dspy_version_marker/{n;s/dspy==[^"]*/dspy==${{ needs.extract-tag.outputs.version }}/;}' ./dspy/.internal_dspyai/setup.py  
+          sed -i '/#replace_dspy_version_marker/{n;s/dspy==[^"]*/dspy>=${{ needs.extract-tag.outputs.version }}/;}' ./dspy/.internal_dspyai/setup.py  
       - name: Build a binary wheel
         run: python3 ./dspy/.internal_dspyai/setup.py sdist bdist_wheel
       - name: Publish distribution 📦 to PyPI (dspy-ai)
-        uses: pypa/gh-action-pypi-publish@release/v1 # This requires a trusted publisher to be setup in pypi
+        uses: pypa/gh-action-pypi-publish@release/v1 # This requires a trusted publisher to be setup in pypi
+        with: 
+          attestations: false

dspy/.internal_dspyai/setup.py

@@ -19,5 +19,5 @@
     packages=find_packages(include=["dsp.*", "dspy.*", "dsp", "dspy"]),	
     python_requires=">=3.9",
     #replace_dspy_version_marker
-    install_requires=["dspy==2.5.3"]
+    install_requires=["dspy>=2.5.3"]
 )	

pyproject.toml

@@ -6,7 +6,7 @@ build-backend = "setuptools.build_meta"
 #replace_package_name_marker
 name="dspy"
 #replace_package_version_marker
-version="2.5.25"
+version="2.5.27"
 description = "DSPy"
 readme = "README.md"
 authors = [{ name = "Omar Khattab", email = "[email protected]" }]
@@ -133,7 +133,7 @@ llama-index = {version = "^0.10.30", optional = true}
 snowflake-snowpark-python = { version = "*",optional=true, python = ">=3.9,<3.12" }
 jinja2 = "^3.1.3"
 magicattr = "^0.1.6"
-litellm = "1.49.1"
+litellm = "1.51.0"
 diskcache = "^5.6.0"
 
 redis = "^5.1.1"

requirements.txt

@@ -4,7 +4,7 @@ diskcache
 httpx
 joblib~=1.3
 json-repair
-litellm<=1.49.1
+litellm==1.51.0
 magicattr~=0.1.6
 openai
 optuna

setup.py

@@ -12,7 +12,7 @@
     #replace_package_name_marker
     name="dspy",
     #replace_package_version_marker
-    version="2.5.25", 	
+    version="2.5.27", 	
     description="DSPy",	
     long_description=long_description,	
     long_description_content_type="text/markdown",	

tests/clients/test_lm.py

@@ -1,25 +1,145 @@
 from unittest import mock
 
-from dspy.clients.lm import LM
+import pytest
+# from litellm.router import RetryPolicy
 
+# from dspy.clients.lm import LM, _get_litellm_router
 
-def test_lm_chat_respects_max_retries():
-    lm = LM(model="openai/gpt4o", model_type="chat", max_retries=17)
 
-    with mock.patch("dspy.clients.lm.litellm.completion") as litellm_completion_api:
-        lm(messages=[{"content": "Hello, world!", "role": "user"}])
+# @pytest.mark.parametrize("keys_in_env_vars", [True, False])
+# def test_lm_chat_respects_max_retries(keys_in_env_vars, monkeypatch):
+#     model_name = "openai/gpt4o"
+#     num_retries = 17
+#     temperature = 0.5
+#     max_tokens = 100
+#     prompt = "Hello, world!"
+#     api_version = "2024-02-01"
+#     api_key = "apikey"
 
-    assert litellm_completion_api.call_count == 1
-    assert litellm_completion_api.call_args[1]["max_retries"] == 17
-    # assert litellm_completion_api.call_args[1]["retry_strategy"] == "exponential_backoff_retry"
+#     lm_kwargs = {
+#         "model": model_name,
+#         "model_type": "chat",
+#         "num_retries": num_retries,
+#         "temperature": temperature,
+#         "max_tokens": max_tokens,
+#     }
+#     if keys_in_env_vars:
+#         api_base = "http://testfromenv.com"
+#         monkeypatch.setenv("OPENAI_API_KEY", api_key)
+#         monkeypatch.setenv("OPENAI_API_BASE", api_base)
+#         monkeypatch.setenv("OPENAI_API_VERSION", api_version)
+#     else:
+#         api_base = "http://test.com"
+#         lm_kwargs["api_key"] = api_key
+#         lm_kwargs["api_base"] = api_base
+#         lm_kwargs["api_version"] = api_version
 
+#     lm = LM(**lm_kwargs)
 
-def test_lm_completions_respects_max_retries():
-    lm = LM(model="openai/gpt-3.5-turbo", model_type="completions", max_retries=17)
+#     MockRouter = mock.MagicMock()
+#     mock_completion = mock.MagicMock()
+#     MockRouter.completion = mock_completion
 
-    with mock.patch("dspy.clients.lm.litellm.text_completion") as litellm_completion_api:
-        lm(prompt="Hello, world!")
+#     with mock.patch("dspy.clients.lm.Router", return_value=MockRouter) as MockRouterConstructor:
+#         lm(prompt=prompt)
 
-    assert litellm_completion_api.call_count == 1
-    assert litellm_completion_api.call_args[1]["max_retries"] == 17
-    # assert litellm_completion_api.call_args[1]["retry_strategy"] == "exponential_backoff_retry"
+#         MockRouterConstructor.assert_called_once_with(
+#             model_list=[
+#                 {
+#                     "model_name": model_name,
+#                     "litellm_params": {
+#                         "model": model_name,
+#                         "api_key": api_key,
+#                         "api_base": api_base,
+#                         "api_version": api_version,
+#                     },
+#                 }
+#             ],
+#             retry_policy=RetryPolicy(
+#                 TimeoutErrorRetries=num_retries,
+#                 RateLimitErrorRetries=num_retries,
+#                 InternalServerErrorRetries=num_retries,
+#                 BadRequestErrorRetries=0,
+#                 AuthenticationErrorRetries=0,
+#                 ContentPolicyViolationErrorRetries=0,
+#             ),
+#         )
+#         mock_completion.assert_called_once_with(
+#             model=model_name,
+#             messages=[{"role": "user", "content": prompt}],
+#             temperature=temperature,
+#             max_tokens=max_tokens,
+#             cache=mock.ANY,
+#         )
+
+
+# @pytest.mark.parametrize("keys_in_env_vars", [True, False])
+# def test_lm_completions_respects_max_retries(keys_in_env_vars, monkeypatch):
+#     model_name = "azure/gpt-3.5-turbo"
+#     expected_model = "text-completion-openai/" + model_name.split("/")[-1]
+#     num_retries = 17
+#     temperature = 0.5
+#     max_tokens = 100
+#     prompt = "Hello, world!"
+#     api_version = "2024-02-01"
+#     api_key = "apikey"
+#     azure_ad_token = "adtoken"
+
+#     lm_kwargs = {
+#         "model": model_name,
+#         "model_type": "text",
+#         "num_retries": num_retries,
+#         "temperature": temperature,
+#         "max_tokens": max_tokens,
+#     }
+#     if keys_in_env_vars:
+#         api_base = "http://testfromenv.com"
+#         monkeypatch.setenv("AZURE_API_KEY", api_key)
+#         monkeypatch.setenv("AZURE_API_BASE", api_base)
+#         monkeypatch.setenv("AZURE_API_VERSION", api_version)
+#         monkeypatch.setenv("AZURE_AD_TOKEN", azure_ad_token)
+#     else:
+#         api_base = "http://test.com"
+#         lm_kwargs["api_key"] = api_key
+#         lm_kwargs["api_base"] = api_base
+#         lm_kwargs["api_version"] = api_version
+#         lm_kwargs["azure_ad_token"] = azure_ad_token
+
+#     lm = LM(**lm_kwargs)
+
+#     MockRouter = mock.MagicMock()
+#     mock_text_completion = mock.MagicMock()
+#     MockRouter.text_completion = mock_text_completion
+
+#     with mock.patch("dspy.clients.lm.Router", return_value=MockRouter) as MockRouterConstructor:
+#         lm(prompt=prompt)
+
+#         MockRouterConstructor.assert_called_once_with(
+#             model_list=[
+#                 {
+#                     "model_name": expected_model,
+#                     "litellm_params": {
+#                         "model": expected_model,
+#                         "api_key": api_key,
+#                         "api_base": api_base,
+#                         "api_version": api_version,
+#                         "azure_ad_token": azure_ad_token,
+#                     },
+#                 }
+#             ],
+#             retry_policy=RetryPolicy(
+#                 TimeoutErrorRetries=num_retries,
+#                 RateLimitErrorRetries=num_retries,
+#                 InternalServerErrorRetries=num_retries,
+#                 BadRequestErrorRetries=0,
+#                 AuthenticationErrorRetries=0,
+#                 ContentPolicyViolationErrorRetries=0,
+#             ),
+#         )
+#         mock_text_completion.assert_called_once_with(
+#             model=expected_model,
+#             prompt=prompt + "\n\nBEGIN RESPONSE:",
+#             temperature=temperature,
+#             max_tokens=max_tokens,
+#             cache=mock.ANY,
+#         )