diff --git a/README.md b/README.md index d891ac332..586869b0a 100644 --- a/README.md +++ b/README.md @@ -340,7 +340,9 @@ dj-analyze --config configs/demo/analyzer.yaml dj-analyze --auto --dataset_path xx.jsonl [--auto_num 1000] ``` -- **Note:** Analyzer only compute stats of Filter ops. So extra Mapper or Deduplicator ops will be ignored in the analysis process. +- **Note:** Analyzer only compute stats for Filters that produce stats or other OPs that produce tags/categories in meta. So other OPs will be ignored in the analysis process. We use the following registries to decorate OPs: + - `NON_STATS_FILTERS`: decorate Filters that **DO NOT** produce any stats. + - `TAGGING_OPS`: decorate OPs that **DO** produce tags/categories in meta field. ### Data Visualization diff --git a/README_ZH.md b/README_ZH.md index 01633731b..42612964a 100644 --- a/README_ZH.md +++ b/README_ZH.md @@ -316,7 +316,9 @@ dj-analyze --config configs/demo/analyzer.yaml dj-analyze --auto --dataset_path xx.jsonl [--auto_num 1000] ``` -* **注意**:Analyzer 只计算 Filter 算子的状态,其他的算子(例如 Mapper 和 Deduplicator)会在分析过程中被忽略。 +* **注意**:Analyzer 只用于能在 stats 字段里产出统计信息的 Filter 算子和能在 meta 字段里产出 tags 或类别标签的其他算子。除此之外的其他的算子会在分析过程中被忽略。我们使用以下两种注册器来装饰相关的算子: + * `NON_STATS_FILTERS`:装饰那些**不能**产出任何统计信息的 Filter 算子。 + * `TAGGING_OPS`:装饰那些能在 meta 字段中产出 tags 或类别标签的算子。 ### 数据可视化