fix(ll4j): bypass some words

BakaBotTeam · Aug 25, 2024 · 4dd5d5c · 4dd5d5c
1 parent f98c5c8
commit 4dd5d5c
Show file tree

Hide file tree

Showing 2 changed files with 9 additions and 3 deletions.
diff --git a/src/main/kotlin/ltd/guimc/lgzbot/PluginMain.kt b/src/main/kotlin/ltd/guimc/lgzbot/PluginMain.kt
@@ -151,6 +151,7 @@ object PluginMain : KotlinPlugin(
  webHookService.start()
  logger.info("正在初始化 LL4J")
  LL4JUtils.init()
+ thread { LL4JUtils.downloadModel() }.start()
  logger.info("$name v$version 加载好了喵")
  }
 

diff --git a/src/main/kotlin/ltd/guimc/lgzbot/utils/LL4JUtils.kt b/src/main/kotlin/ltd/guimc/lgzbot/utils/LL4JUtils.kt
@@ -22,15 +22,20 @@ object LL4JUtils {
  }
 
  fun predict(string: String): Boolean =
- model.predictDebug(tokenizer.tokenize(0, string.replace("\n", "")).values).first == 1
+ model.predictDebug(
+ tokenizer.tokenize(
+ 0,
+ string.replace("\n", "").replace("live.bilibili.com", "")
+ ).values
+ ).first == 1
 
  fun predictDebug(string: String): Pair<Int, Double> =
- model.predictDebug(tokenizer.tokenize(0, string.replace("\n", "")).values)
+ model.predictDebug(tokenizer.tokenize(0, string.replace("\n", "").replace("live.bilibili.com", "")).values)
 
  fun predictAllResult(string: String): DoubleArray =
  model.predictAllResult(
  tokenizer.tokenize(
- 0, sbc2dbcCase(string.replace("\n", ""))
+ 0, sbc2dbcCase(string.replace("\n", "").replace("live.bilibili.com", ""))
  .lowercase()
  .removeInterference()
  .removeNonVisible()