phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-28 09:11:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

fix: make cprint write to stderr

Browse source
This commit is contained in:
Raghotham Murthy 2025-05-24 21:46:40 -07:00
parent c290999c63
commit 8658109454
11 changed files with 81 additions and 44 deletions
Download patch file Download diff file Expand all files Collapse all files

9
llama_stack/providers/inline/inference/meta_reference/inference.py
View file

@ -6,6 +6,7 @@
import asyncio
import os
import sys
from collections.abc import AsyncGenerator
from pydantic import BaseModel
@ -455,9 +456,9 @@ class MetaReferenceInferenceImpl(
first = token_results[0]
if not first.finished and not first.ignore_token:
if os.environ.get("LLAMA_MODELS_DEBUG", "0") in ("1", "2"):
cprint(first.text, "cyan", end="")
cprint(first.text, color="cyan", end="", file=sys.stderr)
if os.environ.get("LLAMA_MODELS_DEBUG", "0") == "2":
cprint(f"<{first.token}>", "magenta", end="")
cprint(f"<{first.token}>", color="magenta", end="", file=sys.stderr)
for result in token_results:
idx = result.batch_idx
@ -519,9 +520,9 @@ class MetaReferenceInferenceImpl(
for token_results in self.generator.chat_completion([request]):
token_result = token_results[0]
if os.environ.get("LLAMA_MODELS_DEBUG", "0") == "1":
cprint(token_result.text, "cyan", end="")
cprint(token_result.text, color="cyan", end="", file=sys.stderr)
if os.environ.get("LLAMA_MODELS_DEBUG", "0") == "2":
cprint(f"<{token_result.token}>", "magenta", end="")
cprint(f"<{token_result.token}>", color="magenta", end="", file=sys.stderr)
if token_result.token == tokenizer.eot_id:
stop_reason = StopReason.end_of_turn