NousResearch/atropos
1
0
Fork 0
mirror of https://github.com/NousResearch/atropos.git synced 2026-04-19 12:57:58 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

[pre-commit.ci] auto fixes from pre-commit.com hooks

Browse source 
for more information, see https://pre-commit.ci
This commit is contained in:
pre-commit-ci[bot] 2025-12-25 09:49:29 +00:00
parent 9e9f1cd88e
commit 269fb71713
4 changed files with 32 additions and 19 deletions
Download patch file Download diff file Expand all files Collapse all files

39
environments/eval_environments/agieval_eval.py
View file

@ -557,7 +557,7 @@ class AGIEvalEnv(BaseEnv):
# Get model completion with retry logic
model_response = None
finish_reason = None
# Build completion kwargs - only include max_tokens if > 0
# (0 means "use model default", so we don't pass the parameter)
completion_kwargs = {
@ -568,19 +568,27 @@ class AGIEvalEnv(BaseEnv):
}
if self.config.eval_max_tokens > 0:
completion_kwargs["max_tokens"] = self.config.eval_max_tokens
for attempt in range(self.config.max_retries):
try:
if self.config.full_debug:
print(f" Making API request (attempt {attempt + 1}/{self.config.max_retries})...")
print(
f" Making API request (attempt {attempt + 1}/{self.config.max_retries})..."
)
try:
model_name = self.server.servers[0].config.model_name if hasattr(self.server, 'servers') else 'unknown'
model_name = (
self.server.servers[0].config.model_name
if hasattr(self.server, "servers")
else "unknown"
)
except Exception:
model_name = 'unknown'
model_name = "unknown"
print(f" Model: {model_name}")
print(f" Temperature: {self.config.eval_temperature}")
print(f" Max tokens: {self.config.eval_max_tokens if self.config.eval_max_tokens > 0 else 'model default'}")
print(
f" Max tokens: {self.config.eval_max_tokens if self.config.eval_max_tokens > 0 else 'model default'}"
)
completion = await self.server.chat_completion(**completion_kwargs)
if completion.choices and completion.choices[0].message.content:
@ -604,11 +612,16 @@ class AGIEvalEnv(BaseEnv):
actual_error = e
error_chain = []
while actual_error is not None:
error_chain.append(f"{type(actual_error).__name__}: {actual_error}")
error_chain.append(
f"{type(actual_error).__name__}: {actual_error}"
)
# Try to get the underlying cause
if hasattr(actual_error, '__cause__') and actual_error.__cause__ is not None:
if (
hasattr(actual_error, "__cause__")
and actual_error.__cause__ is not None
):
actual_error = actual_error.__cause__
elif hasattr(actual_error, 'last_attempt'):
elif hasattr(actual_error, "last_attempt"):
# tenacity RetryError stores the last attempt's exception
try:
actual_error = actual_error.last_attempt.exception()
@ -616,18 +629,18 @@ class AGIEvalEnv(BaseEnv):
break
else:
break
# Always log API errors to help diagnose issues
print(
f" API Error (attempt {attempt + 1}/{self.config.max_retries}): {type(e).__name__}: {e}"
)
# Print the full error chain for debugging
if len(error_chain) > 1:
print(" Error chain:")
for i, err in enumerate(error_chain):
print(f" {' ' * i}-> {err}")
if hasattr(e, "response"):
try:
resp_text = (

4
environments/eval_environments/gpqa_eval.py
View file

@ -479,7 +479,7 @@ class GPQAEvalEnv(BaseEnv):
# Get model completion with retry logic
model_response = None
finish_reason = None
# Build completion kwargs - only include max_tokens if > 0
# (0 means "use model default", so we don't pass the parameter)
completion_kwargs = {
@ -490,7 +490,7 @@ class GPQAEvalEnv(BaseEnv):
}
if self.config.eval_max_tokens > 0:
completion_kwargs["max_tokens"] = self.config.eval_max_tokens
for attempt in range(self.config.max_retries):
try:
completion = await self.server.chat_completion(**completion_kwargs)

4
environments/eval_environments/mmlu_eval.py
View file

@ -775,7 +775,7 @@ class MMLUEvalEnv(BaseEnv):
# Get model completion with retry logic
model_response = None
finish_reason = None
# Build completion kwargs - only include max_tokens if > 0
# (0 means "use model default", so we don't pass the parameter)
completion_kwargs = {
@ -786,7 +786,7 @@ class MMLUEvalEnv(BaseEnv):
}
if self.config.eval_max_tokens > 0:
completion_kwargs["max_tokens"] = self.config.eval_max_tokens
for attempt in range(self.config.max_retries):
try:
completion = await self.server.chat_completion(**completion_kwargs)

4
environments/eval_environments/mmlu_pro_eval.py
View file

@ -557,7 +557,7 @@ class MMLUProEvalEnv(BaseEnv):
# Get model completion with retry logic
model_response = None
finish_reason = None
# Build completion kwargs - only include max_tokens if > 0
# (0 means "use model default", so we don't pass the parameter)
completion_kwargs = {
@ -568,7 +568,7 @@ class MMLUProEvalEnv(BaseEnv):
}
if self.config.eval_max_tokens > 0:
completion_kwargs["max_tokens"] = self.config.eval_max_tokens
for attempt in range(self.config.max_retries):
try:
completion = await self.server.chat_completion(**completion_kwargs)