104 lines
3.3 KiB
Python
104 lines
3.3 KiB
Python
from pprint import pformat
|
|
from typing import Any, Dict, Generic, List, Optional, TypeVar
|
|
|
|
from pydantic import Extra
|
|
|
|
from .hashable_base_model import HashableBaseModel
|
|
from .model import Model
|
|
|
|
T = TypeVar("T")
|
|
|
|
|
|
class Trace(Generic[T], HashableBaseModel):
|
|
"""Universal structure for storing prediction traces and training data.
|
|
|
|
Attributes:
|
|
trace_id: UUID4 identifier for uniquely referring to a trace.
|
|
created: Timestamp of its (original) construction.
|
|
original_execution_time_ms: Wall-time elapsed while its generating
|
|
TracingContext was alive.
|
|
logged_values: Values persisted through using `@parameter` or `log_metric()`.
|
|
models: Marks left by each encountered `@use_model` decorated function.
|
|
exception: Exception description if any was encountered.
|
|
output: Return value of the function wrapped by GreatAI.
|
|
feedback: Feedback obtained using the REST API of `add_ground_truth`.
|
|
tags: Tags used for filtering traces. Contains the name of the original
|
|
function, value of `ENVIRONMENT`, its split if has any, and either
|
|
`ground_truth` or `online` depending on the origin of the Trace.
|
|
"""
|
|
|
|
trace_id: str
|
|
created: str
|
|
original_execution_time_ms: float
|
|
logged_values: Dict[str, Any]
|
|
models: List[Model]
|
|
exception: Optional[str]
|
|
output: Optional[T]
|
|
feedback: Any = None
|
|
tags: List[str]
|
|
|
|
class Config:
|
|
extra = Extra.ignore
|
|
|
|
@property
|
|
def input(self) -> Any:
|
|
return (
|
|
self.logged_values["input"]
|
|
if list(self.logged_values.keys()) == ["input"]
|
|
else self.logged_values
|
|
)
|
|
|
|
@property
|
|
def models_flat(self) -> str:
|
|
return ", ".join(f"{m.key}:{m.version}" for m in self.models)
|
|
|
|
@property
|
|
def output_flat(self) -> str:
|
|
return pformat(self.output, indent=2, compact=True)
|
|
|
|
@property
|
|
def exception_flat(self) -> str:
|
|
return (
|
|
"null"
|
|
if self.exception is None
|
|
else pformat(self.exception, indent=2, compact=True)
|
|
)
|
|
|
|
@property
|
|
def feedback_flat(self) -> str:
|
|
return (
|
|
"null"
|
|
if self.feedback is None
|
|
else pformat(self.feedback, indent=2, compact=True)
|
|
)
|
|
|
|
@property
|
|
def tags_flat(self) -> str:
|
|
return ",\n".join(self.tags)
|
|
|
|
def to_flat_dict(self, include_original: bool = True) -> Dict[str, Any]:
|
|
return {
|
|
**(
|
|
self.dict()
|
|
if include_original
|
|
else {
|
|
"trace_id": self.trace_id,
|
|
"created": self.created,
|
|
"original_execution_time_ms": self.original_execution_time_ms,
|
|
}
|
|
),
|
|
**{
|
|
k: pformat(v, indent=2, compact=True)
|
|
for k, v in self.logged_values.items()
|
|
},
|
|
"models_flat": self.models_flat,
|
|
"exception_flat": self.exception_flat,
|
|
"output_flat": self.output_flat,
|
|
"feedback_flat": self.feedback_flat,
|
|
"tags_flat": self.tags_flat,
|
|
}
|
|
|
|
def __repr__(self) -> str:
|
|
return f"""Trace[{type(self.output).__name__}]({
|
|
pformat(self.dict(), indent=2, compact=True).replace('{ ', '{', 1)
|
|
})"""
|