CI yaml

tools/base_data.json

@@ -3,7 +3,7 @@
     "precision": 1.0,
     "recall": 1.0,
     "f1_score": 1.0,
-    "pdf间的平均编辑距离": 133.10256410256412,
+    "pdf间的平均编辑距离": 131.10256410256412,
     "pdf间的平均bleu": 0.28838311595434046,
     "分段准确率": 0.07220216606498195,
     "行内公式准确率": {

tools/ocr_badcase.py

@@ -843,8 +843,8 @@ def main(standard_file, test_file, zip_file, badcase_path, overall_path,base_dat
     save_results(result_dict, overall_report_dict,badcase_file,overall_file)
 
     result=compare_edit_distance(base_data_path, overall_report_dict)
-    
     print(result)
+    assert result == 1
 
 if __name__ == "__main__":
     parser = argparse.ArgumentParser(description="主函数，执行整个评估流程。")

tools/overall.json_2024-04-10_17-03-36.json

+{
+    "accuracy": 1.0,
+    "precision": 1.0,
+    "recall": 1.0,
+    "f1_score": 1.0,
+    "pdf间的平均编辑距离": 10.10256410256412,
+    "pdf间的平均bleu": 0.28838311595434046,
+    "分段准确率": 0.07220216606498195,
+    "行内公式准确率": {
+        "accuracy": 0.004835727492533068,
+        "precision": 0.008790072388831437,
+        "recall": 0.010634970284641852,
+        "f1_score": 0.009624911535739562
+    },
+    "行内公式编辑距离": 1.6176470588235294,
+    "行内公式bleu": 0.17154724654721457,
+    "行间公式准确率": {
+        "accuracy": 0.08490566037735849,
+        "precision": 0.1836734693877551,
+        "recall": 0.13636363636363635,
+        "f1_score": 0.1565217391304348
+    },
+    "行间公式编辑距离": 113.22222222222223,
+    "行间公式bleu": 0.2531053359913409,
+    "丢弃文本准确率": {
+        "accuracy": 0.00035398230088495576,
+        "precision": 0.0006389776357827476,
+        "recall": 0.0007930214115781126,
+        "f1_score": 0.0007077140835102619
+    },
+    "丢弃文本标签准确率": {
+        "rotate": {
+            "precision": 0.0,
+            "recall": 0.0,
+            "f1-score": 0.0,
+            "support": 63
+        },
+        "color_background_header_txt_block": {
+            "precision": 0.0,
+            "recall": 0.0,
+            "f1-score": 0.0,
+            "support": 41
+        },
+        "footnote": {
+            "precision": 1.0,
+            "recall": 0.009708737864077669,
+            "f1-score": 0.01923076923076923,
+            "support": 103
+        },
+        "header": {
+            "precision": 0.0,
+            "recall": 0.0,
+            "f1-score": 0.0,
+            "support": 4
+        },
+        "on-image": {
+            "precision": 0.0,
+            "recall": 0.0,
+            "f1-score": 0.0,
+            "support": 380
+        },
+        "on-table": {
+            "precision": 0.0,
+            "recall": 0.0,
+            "f1-score": 0.0,
+            "support": 665
+        },
+        "micro avg": {
+            "precision": 1.0,
+            "recall": 0.0007961783439490446,
+            "f1-score": 0.001591089896579157,
+            "support": 1256
+        }
+    },
+    "丢弃图片准确率": {
+        "accuracy": 0.0,
+        "precision": 0.0,
+        "recall": 0.0,
+        "f1_score": 0.0
+    },
+    "丢弃表格准确率": {
+        "accuracy": 0.0,
+        "precision": 0.0,
+        "recall": 0.0,
+        "f1_score": 0.0
+    }
+}
\ No newline at end of file

tools/overall.json_2024-04-10_17-06-39.json

+{
+    "accuracy": 1.0,
+    "precision": 1.0,
+    "recall": 1.0,
+    "f1_score": 1.0,
+    "pdf间的平均编辑距离": 133.10256410256412,
+    "pdf间的平均bleu": 0.28838311595434046,
+    "分段准确率": 0.07220216606498195,
+    "行内公式准确率": {
+        "accuracy": 0.004835727492533068,
+        "precision": 0.008790072388831437,
+        "recall": 0.010634970284641852,
+        "f1_score": 0.009624911535739562
+    },
+    "行内公式编辑距离": 1.6176470588235294,
+    "行内公式bleu": 0.17154724654721457,
+    "行间公式准确率": {
+        "accuracy": 0.08490566037735849,
+        "precision": 0.1836734693877551,
+        "recall": 0.13636363636363635,
+        "f1_score": 0.1565217391304348
+    },
+    "行间公式编辑距离": 113.22222222222223,
+    "行间公式bleu": 0.2531053359913409,
+    "丢弃文本准确率": {
+        "accuracy": 0.00035398230088495576,
+        "precision": 0.0006389776357827476,
+        "recall": 0.0007930214115781126,
+        "f1_score": 0.0007077140835102619
+    },
+    "丢弃文本标签准确率": {
+        "rotate": {
+            "precision": 0.0,
+            "recall": 0.0,
+            "f1-score": 0.0,
+            "support": 63
+        },
+        "footnote": {
+            "precision": 1.0,
+            "recall": 0.009708737864077669,
+            "f1-score": 0.01923076923076923,
+            "support": 103
+        },
+        "on-image": {
+            "precision": 0.0,
+            "recall": 0.0,
+            "f1-score": 0.0,
+            "support": 380
+        },
+        "color_background_header_txt_block": {
+            "precision": 0.0,
+            "recall": 0.0,
+            "f1-score": 0.0,
+            "support": 41
+        },
+        "on-table": {
+            "precision": 0.0,
+            "recall": 0.0,
+            "f1-score": 0.0,
+            "support": 665
+        },
+        "header": {
+            "precision": 0.0,
+            "recall": 0.0,
+            "f1-score": 0.0,
+            "support": 4
+        },
+        "micro avg": {
+            "precision": 1.0,
+            "recall": 0.0007961783439490446,
+            "f1-score": 0.001591089896579157,
+            "support": 1256
+        }
+    },
+    "丢弃图片准确率": {
+        "accuracy": 0.0,
+        "precision": 0.0,
+        "recall": 0.0,
+        "f1_score": 0.0
+    },
+    "丢弃表格准确率": {
+        "accuracy": 0.0,
+        "precision": 0.0,
+        "recall": 0.0,
+        "f1_score": 0.0
+    }
+}
\ No newline at end of file

tools/overall.json_2024-04-10_17-07-54.json

+{
+    "accuracy": 1.0,
+    "precision": 1.0,
+    "recall": 1.0,
+    "f1_score": 1.0,
+    "pdf间的平均编辑距离": 133.10256410256412,
+    "pdf间的平均bleu": 0.28838311595434046,
+    "分段准确率": 0.07220216606498195,
+    "行内公式准确率": {
+        "accuracy": 0.004835727492533068,
+        "precision": 0.008790072388831437,
+        "recall": 0.010634970284641852,
+        "f1_score": 0.009624911535739562
+    },
+    "行内公式编辑距离": 1.6176470588235294,
+    "行内公式bleu": 0.17154724654721457,
+    "行间公式准确率": {
+        "accuracy": 0.08490566037735849,
+        "precision": 0.1836734693877551,
+        "recall": 0.13636363636363635,
+        "f1_score": 0.1565217391304348
+    },
+    "行间公式编辑距离": 113.22222222222223,
+    "行间公式bleu": 0.2531053359913409,
+    "丢弃文本准确率": {
+        "accuracy": 0.00035398230088495576,
+        "precision": 0.0006389776357827476,
+        "recall": 0.0007930214115781126,
+        "f1_score": 0.0007077140835102619
+    },
+    "丢弃文本标签准确率": {
+        "header": {
+            "precision": 0.0,
+            "recall": 0.0,
+            "f1-score": 0.0,
+            "support": 4
+        },
+        "on-image": {
+            "precision": 0.0,
+            "recall": 0.0,
+            "f1-score": 0.0,
+            "support": 380
+        },
+        "rotate": {
+            "precision": 0.0,
+            "recall": 0.0,
+            "f1-score": 0.0,
+            "support": 63
+        },
+        "color_background_header_txt_block": {
+            "precision": 0.0,
+            "recall": 0.0,
+            "f1-score": 0.0,
+            "support": 41
+        },
+        "on-table": {
+            "precision": 0.0,
+            "recall": 0.0,
+            "f1-score": 0.0,
+            "support": 665
+        },
+        "footnote": {
+            "precision": 1.0,
+            "recall": 0.009708737864077669,
+            "f1-score": 0.01923076923076923,
+            "support": 103
+        },
+        "micro avg": {
+            "precision": 1.0,
+            "recall": 0.0007961783439490446,
+            "f1-score": 0.001591089896579157,
+            "support": 1256
+        }
+    },
+    "丢弃图片准确率": {
+        "accuracy": 0.0,
+        "precision": 0.0,
+        "recall": 0.0,
+        "f1_score": 0.0
+    },
+    "丢弃表格准确率": {
+        "accuracy": 0.0,
+        "precision": 0.0,
+        "recall": 0.0,
+        "f1_score": 0.0
+    }
+}
\ No newline at end of file