Pin llama-cpp-python version

deepset-ai · Aug 2, 2024 · 7e091da · 7e091da
1 parent dcd2e38
commit 7e091da
Show file tree

Hide file tree

Showing 4 changed files with 8 additions and 11 deletions.
diff --git a/integrations/anthropic/example/documentation_rag_with_claude.py b/integrations/anthropic/example/documentation_rag_with_claude.py
@@ -1,7 +1,7 @@
 # To run this example, you will need to set a `ANTHROPIC_API_KEY` environment variable.
 
 from haystack import Pipeline
-from haystack.components.builders import DynamicChatPromptBuilder
+from haystack.components.builders import ChatPromptBuilder
 from haystack.components.converters import HTMLToDocument
 from haystack.components.fetchers import LinkContentFetcher
 from haystack.components.generators.utils import print_streaming_chunk
@@ -18,7 +18,7 @@
 rag_pipeline = Pipeline()
 rag_pipeline.add_component("fetcher", LinkContentFetcher())
 rag_pipeline.add_component("converter", HTMLToDocument())
-rag_pipeline.add_component("prompt_builder", DynamicChatPromptBuilder(runtime_variables=["documents"]))
+rag_pipeline.add_component("prompt_builder", ChatPromptBuilder())
 rag_pipeline.add_component(
     "llm",
     AnthropicChatGenerator(
@@ -30,12 +30,12 @@
 
 rag_pipeline.connect("fetcher", "converter")
 rag_pipeline.connect("converter", "prompt_builder")
-rag_pipeline.connect("prompt_builder", "llm")
+rag_pipeline.connect("prompt_builder.prompt", "llm.messages")
 
 question = "What are the best practices in prompt engineering?"
 rag_pipeline.run(
     data={
         "fetcher": {"urls": ["https://docs.anthropic.com/claude/docs/prompt-engineering"]},
-        "prompt_builder": {"template_variables": {"query": question}, "prompt_source": messages},
+        "prompt_builder": {"template_variables": {"query": question}, "template": messages},
     }
 )
diff --git a/integrations/llama_cpp/pyproject.toml b/integrations/llama_cpp/pyproject.toml
@@ -26,7 +26,7 @@ classifiers = [
     "Programming Language :: Python :: Implementation :: CPython",
     "Programming Language :: Python :: Implementation :: PyPy",
 ]
-dependencies = ["haystack-ai", "llama-cpp-python"]
+dependencies = ["haystack-ai", "llama-cpp-python==0.2.79"]
 
 [project.urls]
 Documentation = "https://github.com/deepset-ai/haystack-core-integrations/tree/main/integrations/llama_cpp#readme"

diff --git a/integrations/llama_cpp/tests/test_chat_generator.py b/integrations/llama_cpp/tests/test_chat_generator.py
@@ -213,9 +213,7 @@ def test_run_rag_pipeline(self, generator):
             instance=InMemoryBM25Retriever(document_store=document_store, top_k=1),
             name="retriever",
         )
-        pipeline.add_component(
-            instance=ChatPromptBuilder(variables=["query", "documents"]), name="prompt_builder"
-        )
+        pipeline.add_component(instance=ChatPromptBuilder(variables=["query", "documents"]), name="prompt_builder")
         pipeline.add_component(instance=generator, name="llm")
         pipeline.connect("retriever.documents", "prompt_builder.documents")
         pipeline.connect("prompt_builder.prompt", "llm.messages")
@@ -412,7 +410,6 @@ def test_function_call_and_execute(self, generator):
             messages.append(function_message)
 
         second_response = generator.run(messages=messages)
-        print(second_response)
         assert "replies" in second_response
         assert len(second_response["replies"]) > 0
         assert any("San Francisco" in reply.content for reply in second_response["replies"])

diff --git a/integrations/mistral/examples/streaming_chat_with_rag.py b/integrations/mistral/examples/streaming_chat_with_rag.py
@@ -39,7 +39,7 @@
 
 text_embedder = MistralTextEmbedder()
 retriever = InMemoryEmbeddingRetriever(document_store=document_store)
-prompt_builder = ChatPromptBuilder(runtime_variables=["documents"])
+prompt_builder = ChatPromptBuilder(variables=["documents"])
 llm = MistralChatGenerator(streaming_callback=print_streaming_chunk)
 
 messages = [ChatMessage.from_user("Here are some the documents: {{documents}} \\n Answer: {{query}}")]
@@ -60,7 +60,7 @@
 result = rag_pipeline.run(
     {
         "text_embedder": {"text": question},
-        "prompt_builder": {"template_variables": {"query": question}, "prompt_source": messages},
+        "prompt_builder": {"template_variables": {"query": question}, "template": messages},
         "llm": {"generation_kwargs": {"max_tokens": 165}},
     }
 )