Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[functionality] Implement completion() methods #168 #173

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
+120 −93
Open

[functionality] Implement completion() methods #168 #173

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions .gitignore
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,3 +13,4 @@ xcuserdata/
Package.resolved
*.pte
*.ipynb_checkpoints*
venv/
27 changes: 25 additions & 2 deletions llama_stack/providers/adapters/inference/bedrock/bedrock.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -95,8 +95,31 @@ async def completion(
sampling_params: Optional[SamplingParams] = SamplingParams(),
stream: Optional[bool] = False,
logprobs: Optional[LogProbConfig] = None,
) -> Union[CompletionResponse, CompletionResponseStreamChunk]:
raise NotImplementedError()
) -> AsyncGenerator[Union[CompletionResponse, CompletionResponseStreamChunk], None]:
# Convert the content to a single message
messages = [UserMessage(content=content)]

# Use the existing chat_completion method
async for response in self.chat_completion(
model=model,
messages=messages,
sampling_params=sampling_params,
stream=stream,
logprobs=logprobs,
):
if isinstance(response, ChatCompletionResponse):
yield CompletionResponse(
completion=response.completion_message.content[0],
logprobs=response.logprobs,
)
elif isinstance(response, ChatCompletionResponseStreamChunk):
yield CompletionResponseStreamChunk(
event=CompletionResponseEvent(
event_type=CompletionResponseEventType(response.event.event_type.value),
delta=response.event.delta,
stop_reason=response.event.stop_reason,
)
)

@staticmethod
def _bedrock_stop_reason_to_stop_reason(bedrock_stop_reason: str) -> StopReason:
Expand Down
47 changes: 44 additions & 3 deletions llama_stack/providers/adapters/inference/fireworks/fireworks.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.

from typing import AsyncGenerator
from typing import AsyncGenerator, Union

from fireworks.client import Fireworks

Expand Down Expand Up @@ -56,8 +56,49 @@ async def completion(
sampling_params: Optional[SamplingParams] = SamplingParams(),
stream: Optional[bool] = False,
logprobs: Optional[LogProbConfig] = None,
) -> AsyncGenerator:
raise NotImplementedError()
) -> AsyncGenerator[Union[CompletionResponse, CompletionResponseStreamChunk], None]:
fireworks_model = self.map_to_provider_model(model)
options = self.get_fireworks_chat_options(ChatCompletionRequest(
model=model,
messages=[],
sampling_params=sampling_params,
stream=stream,
logprobs=logprobs,
))

if not stream:
response = await self.client.completions.create(
model=fireworks_model,
prompt=content,
stream=False,
**options,
)
yield CompletionResponse(
completion=response.choices[0].text,
logprobs=None, # Fireworks doesn't provide logprobs
)
else:
async for chunk in self.client.completions.create(
model=fireworks_model,
prompt=content,
stream=True,
**options,
):
if chunk.choices[0].text:
yield CompletionResponseStreamChunk(
event=CompletionResponseEvent(
event_type=CompletionResponseEventType.progress,
delta=chunk.choices[0].text,
)
)

yield CompletionResponseStreamChunk(
event=CompletionResponseEvent(
event_type=CompletionResponseEventType.complete,
delta="",
stop_reason=StopReason.end_of_turn,
)
)

def _messages_to_fireworks_messages(self, messages: list[Message]) -> list:
fireworks_messages = []
Expand Down
2 changes: 1 addition & 1 deletion llama_stack/providers/adapters/inference/ollama/ollama.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -262,4 +262,4 @@ async def chat_completion(
delta="",
stop_reason=stop_reason,
)
)
)
52 changes: 48 additions & 4 deletions llama_stack/providers/adapters/inference/together/together.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.

from typing import AsyncGenerator
from typing import AsyncGenerator, Union

from llama_models.llama3.api.chat_format import ChatFormat

Expand Down Expand Up @@ -61,8 +61,52 @@ async def completion(
sampling_params: Optional[SamplingParams] = SamplingParams(),
stream: Optional[bool] = False,
logprobs: Optional[LogProbConfig] = None,
) -> AsyncGenerator:
raise NotImplementedError()
) -> AsyncGenerator[Union[CompletionResponse, CompletionResponseStreamChunk], None]:
together_api_key = self.config.api_key or self.get_request_provider_data().together_api_key
client = Together(api_key=together_api_key)

together_model = self.map_to_provider_model(model)
options = self.get_together_chat_options(ChatCompletionRequest(
model=model,
messages=[],
sampling_params=sampling_params,
stream=stream,
logprobs=logprobs,
))

if not stream:
response = client.completions.create(
model=together_model,
prompt=content,
stream=False,
**options,
)
yield CompletionResponse(
completion=response.choices[0].text,
logprobs=None, # Together doesn't provide logprobs
)
else:
for chunk in client.completions.create(
model=together_model,
prompt=content,
stream=True,
**options,
):
if chunk.choices[0].text:
yield CompletionResponseStreamChunk(
event=CompletionResponseEvent(
event_type=CompletionResponseEventType.progress,
delta=chunk.choices[0].text,
)
)

yield CompletionResponseStreamChunk(
event=CompletionResponseEvent(
event_type=CompletionResponseEventType.complete,
delta="",
stop_reason=StopReason.end_of_turn,
)
)

def _messages_to_together_messages(self, messages: list[Message]) -> list:
together_messages = []
Expand Down Expand Up @@ -262,4 +306,4 @@ async def chat_completion(
delta="",
stop_reason=stop_reason,
)
)
)
84 changes: 1 addition & 83 deletions llama_stack/providers/impls/meta_reference/inference/inference.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -127,86 +127,4 @@ async def chat_completion(
),
)
)
buffer = buffer[len("<|python_tag|>") :]
continue

if not request.stream:
if request.logprobs:
logprobs.append(token_result.logprob)

continue

if token_result.text == "<|eot_id|>":
stop_reason = StopReason.end_of_turn
text = ""
elif token_result.text == "<|eom_id|>":
stop_reason = StopReason.end_of_message
text = ""
else:
text = token_result.text

if ipython:
delta = ToolCallDelta(
content=text,
parse_status=ToolCallParseStatus.in_progress,
)
else:
delta = text

if stop_reason is None:
yield ChatCompletionResponseStreamChunk(
event=ChatCompletionResponseEvent(
event_type=ChatCompletionResponseEventType.progress,
delta=delta,
stop_reason=stop_reason,
)
)

if stop_reason is None:
stop_reason = StopReason.out_of_tokens

# TODO(ashwin): parse tool calls separately here and report errors?
# if someone breaks the iteration before coming here we are toast
message = self.generator.formatter.decode_assistant_message(
tokens, stop_reason
)
if request.stream:
parsed_tool_calls = len(message.tool_calls) > 0
if ipython and not parsed_tool_calls:
yield ChatCompletionResponseStreamChunk(
event=ChatCompletionResponseEvent(
event_type=ChatCompletionResponseEventType.progress,
delta=ToolCallDelta(
content="",
parse_status=ToolCallParseStatus.failure,
),
stop_reason=stop_reason,
)
)

for tool_call in message.tool_calls:
yield ChatCompletionResponseStreamChunk(
event=ChatCompletionResponseEvent(
event_type=ChatCompletionResponseEventType.progress,
delta=ToolCallDelta(
content=tool_call,
parse_status=ToolCallParseStatus.success,
),
stop_reason=stop_reason,
)
)

yield ChatCompletionResponseStreamChunk(
event=ChatCompletionResponseEvent(
event_type=ChatCompletionResponseEventType.complete,
delta="",
stop_reason=stop_reason,
)
)

# TODO(ashwin): what else do we need to send out here when everything finishes?
else:
yield ChatCompletionResponse(
completion_message=message,
logprobs=logprobs if request.logprobs else None,
)
buffer = buffer[len("
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

uh oh, this looks like an unintentional deletion?

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

uh oh, this looks like an unintentional deletion?

in my machine is not cut off i dont know why