423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-12-16 03:40:34 +08:00
Code Issues Packages Projects Releases Wiki Activity

add docstrings, refactor logger

Browse Source 
Former-commit-id: 54c6905937
This commit is contained in:
hiyouga
2024-09-08 00:56:56 +08:00
parent 857d5b9d0a
commit 7ccb86b215
30 changed files with 334 additions and 57 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
src/llamafactory/eval/evaluator.py
View File

@@ -39,7 +39,7 @@
import json
import os
from typing import Any, Dict, List, Optional
from typing import TYPE_CHECKING, Any, Dict, List, Optional
import numpy as np
import torch
@@ -54,6 +54,10 @@ from ..model import load_model, load_tokenizer
from .template import get_eval_template
if TYPE_CHECKING:
from numpy.typing import NDArray
class Evaluator:
def __init__(self, args: Optional[Dict[str, Any]] = None) -> None:
self.model_args, self.data_args, self.eval_args, finetuning_args = get_eval_args(args)
@@ -65,7 +69,7 @@ class Evaluator:
self.choice_inputs = [self.tokenizer.encode(ch, add_special_tokens=False)[-1] for ch in CHOICES]
@torch.inference_mode()
def batch_inference(self, batch_input: Dict[str, torch.Tensor]) -> List[str]:
def batch_inference(self, batch_input: Dict[str, "torch.Tensor"]) -> List[str]:
logits = self.model(**batch_input).logits
lengths = torch.sum(batch_input["attention_mask"], dim=-1)
word_probs = torch.stack([logits[i, lengths[i] - 1] for i in range(len(lengths))], dim=0)
@@ -132,7 +136,7 @@ class Evaluator:
pbar.close()
self._save_results(category_corrects, results)
def _save_results(self, category_corrects: Dict[str, np.ndarray], results: Dict[str, Dict[int, str]]) -> None:
def _save_results(self, category_corrects: Dict[str, "NDArray"], results: Dict[str, Dict[int, str]]) -> None:
score_info = "\n".join(
[
"{:>15}: {:.2f}".format(category_name, 100 * np.mean(category_correct))