Python: #6501 Anthropic Chat Completion

[andrewldesousa]

Motivation and Context

1. Why is this changed required? Support Anthropic models with Python (Python: Add Anthropic AI Connector #6501)
2. What problem does it solve? ^^
3. What scenario does it contribute to? Multi model use cases and applications that would like to use Anthropic models
4. If it fixes an open issue, please link to the issue here: Python: Add Anthropic AI Connector #6501

Description

The changes made follow the designs followed by the Mistral AI and Open AI connectors.

I added the following Classes :

• AnthropicChatPromptExcecutionSettings
• AnthropicChatCompletion
• AnthropicSettings

To test the changes with the tests please add ANTHROPIC_API_KEY and
ANTHROPIC_CHAT_MODEL_ID as Enviorment Variable

What is implemented:

• Integrate Anthropic ChatCompletion Models without Streaming
• Integrate Anthropic Chat Completion Models with Streaming
• Unit tests and integration tests
• Extended Testing including Unit Testing & More Integration Tests

Contribution Checklist

• The code builds clean without any errors or warnings
• The PR follows the SK Contribution Guidelines and the pre-submission formatting script raises no violations
• All unit tests pass, and I have added new tests where possible
• I didn't break anyone 😄

[eavanvalkenburg]

Couple of small things, but overall this looks great!

[moonbox3]

This is looking great, @andrewldesousa. Thank you for working on it!

[andrewldesousa]

Some additional info on streaming with the Anthropic SDK. Here is an example showing the streaming events. I make sure to stream all the content and still stream the metadata about usage, message author role, and stop reason.

async with client.messages.stream(
        max_tokens=1024,
        messages=[
            {
                "role": "user",
                "content": "What is semantic kernel??",
            }
        ],
        model="claude-3-opus-20240229",
    ) as stream:
        async for text in stream:
            print(text)
        print()

Output:

RawMessageStartEvent(message=Message(id='msg_01Mqrmvgr8SKRMAWkAgzaUGe', content=[], model='claude-3-opus-20240229', role='assistant', stop_reason=None, stop_sequence=None, type='message', usage=Usage(input_tokens=11, output_tokens=3)), type='message_start')
RawContentBlockStartEvent(content_block=TextBlock(text='', type='text'), index=0, type='content_block_start')
RawContentBlockDeltaEvent(delta=TextDelta(text='Hello there!', type='text_delta'), index=0, type='content_block_delta')
TextEvent(type='text', text='Hello there!', snapshot='Hello there!')
ContentBlockStopEvent(index=0, type='content_block_stop', content_block=TextBlock(text='Hello there!', type='text'))
RawMessageDeltaEvent(delta=Delta(stop_reason='end_turn', stop_sequence=None), type='message_delta', usage=MessageDeltaUsage(output_tokens=6))
MessageStopEvent(type='message_stop', message=Message(id='msg_01Mqrmvgr8SKRMAWkAgzaUGe', content=[TextBlock(text='Hello there!', type='text')], model='claude-3-opus-20240229', role='assistant', stop_reason='end_turn', stop_sequence=None, type='message', usage=Usage(input_tokens=11, output_tokens=6)))

To learn more about the streaming events for Anthropic:
https://docs.anthropic.com/en/api/messages-streaming#event-types

cc: @TaoChenOSU @moonbox3 @eavanvalkenburg

[markwallace-microsoft]

Python Test Coverage Report

File	Stmts	Miss	Cover	Missing
semantic_kernel
kernel.py	199	47	76%	150, 161, 165, 315–318, 425, 439–482
semantic_kernel/agents/open_ai
open_ai_assistant_base.py	384	2	99%	300–301
semantic_kernel/connectors/ai/anthropic/services
anthropic_chat_completion.py	94	25	73%	171–181, 196–205, 222–236
semantic_kernel/connectors/ai/azure_ai_inference/services
azure_ai_inference_chat_completion.py	149	4	97%	169, 312, 416, 430
semantic_kernel/connectors/ai/embeddings
embedding_generator_base.py	8	1	88%	50
semantic_kernel/connectors/ai/google
shared_utils.py	39	2	95%	100, 122
semantic_kernel/connectors/ai/google/google_ai/services
google_ai_chat_completion.py	145	3	98%	149, 251, 287
utils.py	63	3	95%	139, 157–162
semantic_kernel/connectors/ai/google/vertex_ai/services
utils.py	65	3	95%	141, 159–164
vertex_ai_chat_completion.py	145	3	98%	143, 244, 280
semantic_kernel/connectors/ai/hugging_face/services
hf_text_completion.py	54	3	94%	97, 105, 136
hf_text_embedding.py	32	5	84%	79–83
semantic_kernel/connectors/memory/redis
redis_collection.py	157	2	99%	145, 313
utils.py	45	11	76%	145–146, 164, 166, 173–188
semantic_kernel/connectors/openapi_plugin
openapi_manager.py	58	2	97%	110–111
openapi_parser.py	88	1	99%	127
openapi_runner.py	105	2	98%	181–182
semantic_kernel/connectors/openapi_plugin/models
rest_api_operation.py	129	1	99%	242
semantic_kernel/contents
function_call_content.py	97	1	99%	201
semantic_kernel/data
vector_store_record_collection.py	249	19	92%	400, 460–464, 472–476, 516–519, 526–529
vector_store_record_utils.py	26	2	92%	50, 52
semantic_kernel/functions
kernel_function_decorator.py	98	1	99%	102
kernel_function_from_method.py	96	1	99%	153
kernel_function_from_prompt.py	154	7	95%	165–166, 180, 201, 219, 239, 322
kernel_plugin.py	187	2	99%	472, 475
semantic_kernel/planners
plan.py	234	45	81%	54, 163–165, 197, 214–227, 264, 269, 277–278, 288–291, 308, 313, 329, 332–337, 355, 360, 363, 365, 372, 386–388, 393–397
semantic_kernel/planners/function_calling_stepwise_planner
function_calling_stepwise_planner.py	116	4	97%	145, 189–190, 198
semantic_kernel/planners/sequential_planner
sequential_planner.py	64	6	91%	71, 75, 109, 125, 134–135
sequential_planner_extensions.py	50	9	82%	31–32, 56, 110–124
sequential_planner_parser.py	77	12	84%	66–74, 93, 117–120
semantic_kernel/schema
kernel_json_schema_builder.py	119	5	96%	50, 84, 206, 210–211
semantic_kernel/template_engine/blocks
code_block.py	77	1	99%	119
named_arg_block.py	43	1	98%	98
semantic_kernel/utils/telemetry
decorators.py	113	4	96%	232, 240, 243, 252
user_agent.py	16	2	88%	18–19
TOTAL	10003	242	98%

Python Unit Test Overview

Tests	Skipped	Failures	Errors	Time
2263	1 💤	0 ❌	0 🔥	50.355s ⏱️

[moonbox3]

A small comment otherwise LGTM.

[eavanvalkenburg]

couple of small notes, but looking good overall!

[eavanvalkenburg]

@andrewldesousa there is a merge conflict, let me know when you resolve that, then we can merge this in! nice work!

[andrewldesousa]

@andrewldesousa there is a merge conflict, let me know when you resolve that, then we can merge this in! nice work!

Thanks! Will look into function calling next

[andrewldesousa]

@moonbox3 @eavanvalkenburg can you guys rereview given I have made some changes since your last review?