From d9f30466f129fc87d2d333b0b9647bf1c4dae276 Mon Sep 17 00:00:00 2001
From: shataowei <taowei.sha@ymm56.com>
Date: Sat, 21 Dec 2019 16:37:26 +0800
Subject: [PATCH] fix bug

---
 .idea/workspace.xml                           | 213 ++++++++++++++----
 .../GBDT.md"                                  |   4 +-
 .../GloVe.md"                                 |   3 +
 3 files changed, 177 insertions(+), 43 deletions(-)

diff --git a/.idea/workspace.xml b/.idea/workspace.xml
index 33e9a27..847d288 100644
--- a/.idea/workspace.xml
+++ b/.idea/workspace.xml
@@ -2,8 +2,8 @@
 <project version="4">
   <component name="ChangeListManager">
     <list default="true" id="be5bf74e-ad1c-4197-8a1f-ac93566d8aa7" name="Default" comment="init">
-      <change type="MODIFICATION" beforePath="$PROJECT_DIR$/.idea/workspace.xml" afterPath="$PROJECT_DIR$/.idea/workspace.xml" />
-      <change type="MODIFICATION" beforePath="$PROJECT_DIR$/README.md" afterPath="$PROJECT_DIR$/README.md" />
+      <change type="MODIFICATION" beforePath="$PROJECT_DIR$/机器学习/集成学习/GBDT.md" afterPath="$PROJECT_DIR$/机器学习/集成学习/GBDT.md" />
+      <change type="MODIFICATION" beforePath="$PROJECT_DIR$/自然语言处理/GloVe.md" afterPath="$PROJECT_DIR$/自然语言处理/GloVe.md" />
     </list>
     <ignored path="Reflection_Summary.iws" />
     <ignored path=".idea/workspace.xml" />
@@ -24,11 +24,11 @@
   </component>
   <component name="FileEditorManager">
     <leaf SIDE_TABS_SIZE_LIMIT_KEY="300">
-      <file leaf-file-name="README.md" pinned="false" current-in-tab="true">
+      <file leaf-file-name="README.md" pinned="false" current-in-tab="false">
         <entry file="file://$PROJECT_DIR$/README.md">
           <provider selected="true" editor-type-id="split-provider[text-editor;markdown-preview-editor]">
             <state split_layout="SPLIT">
-              <first_editor relative-caret-position="280">
+              <first_editor relative-caret-position="3490">
                 <caret line="372" column="89" selection-start-line="372" selection-start-column="89" selection-end-line="372" selection-end-column="89" />
                 <folding />
               </first_editor>
@@ -48,6 +48,78 @@
           </provider>
         </entry>
       </file>
+      <file leaf-file-name="GBDT.md" pinned="false" current-in-tab="true">
+        <entry file="file://$PROJECT_DIR$/机器学习/集成学习/GBDT.md">
+          <provider selected="true" editor-type-id="split-provider[text-editor;markdown-preview-editor]">
+            <state split_layout="SPLIT">
+              <first_editor relative-caret-position="-800">
+                <caret line="21" column="0" selection-start-line="21" selection-start-column="0" selection-end-line="21" selection-end-column="0" />
+                <folding />
+              </first_editor>
+              <second_editor />
+            </state>
+          </provider>
+          <provider editor-type-id="split-provider[text-editor;MarkdownPreviewEditor]">
+            <state split_layout="SPLIT">
+              <first_editor relative-caret-position="0">
+                <caret line="0" column="0" selection-start-line="0" selection-start-column="0" selection-end-line="0" selection-end-column="0" />
+                <folding />
+              </first_editor>
+              <second_editor>
+                <js_state />
+              </second_editor>
+            </state>
+          </provider>
+        </entry>
+      </file>
+      <file leaf-file-name="GloVe.md" pinned="false" current-in-tab="false">
+        <entry file="file://$PROJECT_DIR$/自然语言处理/GloVe.md">
+          <provider selected="true" editor-type-id="split-provider[text-editor;markdown-preview-editor]">
+            <state split_layout="SPLIT">
+              <first_editor relative-caret-position="165">
+                <caret line="11" column="11" selection-start-line="11" selection-start-column="11" selection-end-line="11" selection-end-column="11" />
+                <folding />
+              </first_editor>
+              <second_editor />
+            </state>
+          </provider>
+          <provider editor-type-id="split-provider[text-editor;MarkdownPreviewEditor]">
+            <state split_layout="SPLIT">
+              <first_editor relative-caret-position="0">
+                <caret line="0" column="0" selection-start-line="0" selection-start-column="0" selection-end-line="0" selection-end-column="0" />
+                <folding />
+              </first_editor>
+              <second_editor>
+                <js_state />
+              </second_editor>
+            </state>
+          </provider>
+        </entry>
+      </file>
+      <file leaf-file-name="Word2Vec.md" pinned="false" current-in-tab="false">
+        <entry file="file://$PROJECT_DIR$/自然语言处理/Word2Vec.md">
+          <provider selected="true" editor-type-id="split-provider[text-editor;markdown-preview-editor]">
+            <state split_layout="SPLIT">
+              <first_editor relative-caret-position="495">
+                <caret line="33" column="18" selection-start-line="33" selection-start-column="2" selection-end-line="33" selection-end-column="19" />
+                <folding />
+              </first_editor>
+              <second_editor />
+            </state>
+          </provider>
+          <provider editor-type-id="split-provider[text-editor;MarkdownPreviewEditor]">
+            <state split_layout="SPLIT">
+              <first_editor relative-caret-position="0">
+                <caret line="0" column="0" selection-start-line="0" selection-start-column="0" selection-end-line="0" selection-end-column="0" />
+                <folding />
+              </first_editor>
+              <second_editor>
+                <js_state />
+              </second_editor>
+            </state>
+          </provider>
+        </entry>
+      </file>
       <file leaf-file-name="DIN.md" pinned="false" current-in-tab="false">
         <entry file="file://$PROJECT_DIR$/推荐/DIN.md">
           <provider selected="true" editor-type-id="split-provider[text-editor;markdown-preview-editor]">
@@ -114,7 +186,6 @@
         <option value="$PROJECT_DIR$/机器学习/集成学习/LightGBM.md" />
         <option value="$PROJECT_DIR$/自然语言处理/Bert.md" />
         <option value="$PROJECT_DIR$/机器学习/聚类/kmeans.md" />
-        <option value="$PROJECT_DIR$/自然语言处理/GloVe.md" />
         <option value="$PROJECT_DIR$/自然语言处理/CRF.md" />
         <option value="$PROJECT_DIR$/自然语言处理/LDA.md" />
         <option value="$PROJECT_DIR$/深度学习/Attention.md" />
@@ -129,6 +200,8 @@
         <option value="$PROJECT_DIR$/推荐/DIN.md" />
         <option value="$PROJECT_DIR$/数学/牛顿法/牛顿迭代法求平方根.md" />
         <option value="$PROJECT_DIR$/README.md" />
+        <option value="$PROJECT_DIR$/自然语言处理/GloVe.md" />
+        <option value="$PROJECT_DIR$/机器学习/集成学习/GBDT.md" />
       </list>
     </option>
   </component>
@@ -239,6 +312,38 @@
               <option name="myItemType" value="com.intellij.ide.projectView.impl.nodes.PsiDirectoryNode" />
             </PATH_ELEMENT>
           </PATH>
+          <PATH>
+            <PATH_ELEMENT>
+              <option name="myItemId" value="Reflection_Summary" />
+              <option name="myItemType" value="com.intellij.ide.projectView.impl.nodes.ProjectViewProjectNode" />
+            </PATH_ELEMENT>
+            <PATH_ELEMENT>
+              <option name="myItemId" value="Reflection_Summary" />
+              <option name="myItemType" value="com.intellij.ide.projectView.impl.nodes.PsiDirectoryNode" />
+            </PATH_ELEMENT>
+            <PATH_ELEMENT>
+              <option name="myItemId" value="机器学习" />
+              <option name="myItemType" value="com.intellij.ide.projectView.impl.nodes.PsiDirectoryNode" />
+            </PATH_ELEMENT>
+            <PATH_ELEMENT>
+              <option name="myItemId" value="集成学习" />
+              <option name="myItemType" value="com.intellij.ide.projectView.impl.nodes.PsiDirectoryNode" />
+            </PATH_ELEMENT>
+          </PATH>
+          <PATH>
+            <PATH_ELEMENT>
+              <option name="myItemId" value="Reflection_Summary" />
+              <option name="myItemType" value="com.intellij.ide.projectView.impl.nodes.ProjectViewProjectNode" />
+            </PATH_ELEMENT>
+            <PATH_ELEMENT>
+              <option name="myItemId" value="Reflection_Summary" />
+              <option name="myItemType" value="com.intellij.ide.projectView.impl.nodes.PsiDirectoryNode" />
+            </PATH_ELEMENT>
+            <PATH_ELEMENT>
+              <option name="myItemId" value="机器学习" />
+              <option name="myItemType" value="com.intellij.ide.projectView.impl.nodes.PsiDirectoryNode" />
+            </PATH_ELEMENT>
+          </PATH>
           <PATH>
             <PATH_ELEMENT>
               <option name="myItemId" value="Reflection_Summary" />
@@ -991,16 +1096,6 @@
         </state>
       </provider>
     </entry>
-    <entry file="file://$PROJECT_DIR$/自然语言处理/GloVe.md">
-      <provider selected="true" editor-type-id="split-provider[text-editor;markdown-preview-editor]">
-        <state split_layout="SPLIT">
-          <first_editor relative-caret-position="90">
-            <caret line="6" column="20" selection-start-line="6" selection-start-column="2" selection-end-line="6" selection-end-column="20" />
-          </first_editor>
-          <second_editor />
-        </state>
-      </provider>
-    </entry>
     <entry file="file://$PROJECT_DIR$/机器学习/聚类/kmeans.md">
       <provider selected="true" editor-type-id="split-provider[text-editor;markdown-preview-editor]">
         <state split_layout="SPLIT">
@@ -1051,16 +1146,6 @@
         </state>
       </provider>
     </entry>
-    <entry file="file://$PROJECT_DIR$/自然语言处理/Word2Vec.md">
-      <provider selected="true" editor-type-id="split-provider[text-editor;markdown-preview-editor]">
-        <state split_layout="SPLIT">
-          <first_editor relative-caret-position="667">
-            <caret line="33" column="18" selection-start-line="33" selection-start-column="2" selection-end-line="33" selection-end-column="19" />
-          </first_editor>
-          <second_editor />
-        </state>
-      </provider>
-    </entry>
     <entry file="file://$PROJECT_DIR$/深度学习/残差网络.md">
       <provider selected="true" editor-type-id="split-provider[text-editor;markdown-preview-editor]">
         <state split_layout="SPLIT">
@@ -1182,11 +1267,12 @@
         </state>
       </provider>
     </entry>
-    <entry file="file://$PROJECT_DIR$/机器学习/集成学习/GBDT.md">
+    <entry file="file://$PROJECT_DIR$/推荐/DIN.md">
       <provider selected="true" editor-type-id="split-provider[text-editor;markdown-preview-editor]">
         <state split_layout="SPLIT">
-          <first_editor relative-caret-position="255">
-            <caret line="17" column="20" selection-start-line="0" selection-start-column="0" selection-end-line="109" selection-end-column="51" />
+          <first_editor relative-caret-position="15">
+            <caret line="1" column="12" selection-start-line="1" selection-start-column="12" selection-end-line="1" selection-end-column="12" />
+            <folding />
           </first_editor>
           <second_editor />
         </state>
@@ -1195,6 +1281,7 @@
         <state split_layout="SPLIT">
           <first_editor relative-caret-position="0">
             <caret line="0" column="0" selection-start-line="0" selection-start-column="0" selection-end-line="0" selection-end-column="0" />
+            <folding />
           </first_editor>
           <second_editor>
             <js_state />
@@ -1202,12 +1289,11 @@
         </state>
       </provider>
     </entry>
-    <entry file="file://$PROJECT_DIR$/推荐/DIN.md">
+    <entry file="file://$PROJECT_DIR$/推荐/XDeepFM.md">
       <provider selected="true" editor-type-id="split-provider[text-editor;markdown-preview-editor]">
         <state split_layout="SPLIT">
-          <first_editor relative-caret-position="15">
-            <caret line="1" column="12" selection-start-line="1" selection-start-column="12" selection-end-line="1" selection-end-column="12" />
-            <folding />
+          <first_editor relative-caret-position="135">
+            <caret line="16" column="27" selection-start-line="16" selection-start-column="27" selection-end-line="16" selection-end-column="27" />
           </first_editor>
           <second_editor />
         </state>
@@ -1216,7 +1302,6 @@
         <state split_layout="SPLIT">
           <first_editor relative-caret-position="0">
             <caret line="0" column="0" selection-start-line="0" selection-start-column="0" selection-end-line="0" selection-end-column="0" />
-            <folding />
           </first_editor>
           <second_editor>
             <js_state />
@@ -1224,11 +1309,11 @@
         </state>
       </provider>
     </entry>
-    <entry file="file://$PROJECT_DIR$/推荐/XDeepFM.md">
+    <entry file="file://$PROJECT_DIR$/数学/牛顿法/牛顿迭代法求平方根.md">
       <provider selected="true" editor-type-id="split-provider[text-editor;markdown-preview-editor]">
         <state split_layout="SPLIT">
-          <first_editor relative-caret-position="135">
-            <caret line="16" column="27" selection-start-line="16" selection-start-column="27" selection-end-line="16" selection-end-column="27" />
+          <first_editor relative-caret-position="60">
+            <caret line="4" column="25" selection-start-line="4" selection-start-column="25" selection-end-line="4" selection-end-column="25" />
           </first_editor>
           <second_editor />
         </state>
@@ -1244,11 +1329,12 @@
         </state>
       </provider>
     </entry>
-    <entry file="file://$PROJECT_DIR$/数学/牛顿法/牛顿迭代法求平方根.md">
+    <entry file="file://$PROJECT_DIR$/自然语言处理/Bert.md">
       <provider selected="true" editor-type-id="split-provider[text-editor;markdown-preview-editor]">
         <state split_layout="SPLIT">
-          <first_editor relative-caret-position="60">
-            <caret line="4" column="25" selection-start-line="4" selection-start-column="25" selection-end-line="4" selection-end-column="25" />
+          <first_editor relative-caret-position="0">
+            <caret line="0" column="0" selection-start-line="0" selection-start-column="0" selection-end-line="0" selection-end-column="0" />
+            <folding />
           </first_editor>
           <second_editor />
         </state>
@@ -1257,6 +1343,7 @@
         <state split_layout="SPLIT">
           <first_editor relative-caret-position="0">
             <caret line="0" column="0" selection-start-line="0" selection-start-column="0" selection-end-line="0" selection-end-column="0" />
+            <folding />
           </first_editor>
           <second_editor>
             <js_state />
@@ -1264,13 +1351,35 @@
         </state>
       </provider>
     </entry>
-    <entry file="file://$PROJECT_DIR$/自然语言处理/Bert.md">
+    <entry file="file://$PROJECT_DIR$/自然语言处理/Word2Vec.md">
       <provider selected="true" editor-type-id="split-provider[text-editor;markdown-preview-editor]">
+        <state split_layout="SPLIT">
+          <first_editor relative-caret-position="495">
+            <caret line="33" column="18" selection-start-line="33" selection-start-column="2" selection-end-line="33" selection-end-column="19" />
+            <folding />
+          </first_editor>
+          <second_editor />
+        </state>
+      </provider>
+      <provider editor-type-id="split-provider[text-editor;MarkdownPreviewEditor]">
         <state split_layout="SPLIT">
           <first_editor relative-caret-position="0">
             <caret line="0" column="0" selection-start-line="0" selection-start-column="0" selection-end-line="0" selection-end-column="0" />
             <folding />
           </first_editor>
+          <second_editor>
+            <js_state />
+          </second_editor>
+        </state>
+      </provider>
+    </entry>
+    <entry file="file://$PROJECT_DIR$/自然语言处理/GloVe.md">
+      <provider selected="true" editor-type-id="split-provider[text-editor;markdown-preview-editor]">
+        <state split_layout="SPLIT">
+          <first_editor relative-caret-position="165">
+            <caret line="11" column="11" selection-start-line="11" selection-start-column="11" selection-end-line="11" selection-end-column="11" />
+            <folding />
+          </first_editor>
           <second_editor />
         </state>
       </provider>
@@ -1289,7 +1398,7 @@
     <entry file="file://$PROJECT_DIR$/README.md">
       <provider selected="true" editor-type-id="split-provider[text-editor;markdown-preview-editor]">
         <state split_layout="SPLIT">
-          <first_editor relative-caret-position="280">
+          <first_editor relative-caret-position="3490">
             <caret line="372" column="89" selection-start-line="372" selection-start-column="89" selection-end-line="372" selection-end-column="89" />
             <folding />
           </first_editor>
@@ -1308,5 +1417,27 @@
         </state>
       </provider>
     </entry>
+    <entry file="file://$PROJECT_DIR$/机器学习/集成学习/GBDT.md">
+      <provider selected="true" editor-type-id="split-provider[text-editor;markdown-preview-editor]">
+        <state split_layout="SPLIT">
+          <first_editor relative-caret-position="-800">
+            <caret line="21" column="0" selection-start-line="21" selection-start-column="0" selection-end-line="21" selection-end-column="0" />
+            <folding />
+          </first_editor>
+          <second_editor />
+        </state>
+      </provider>
+      <provider editor-type-id="split-provider[text-editor;MarkdownPreviewEditor]">
+        <state split_layout="SPLIT">
+          <first_editor relative-caret-position="0">
+            <caret line="0" column="0" selection-start-line="0" selection-start-column="0" selection-end-line="0" selection-end-column="0" />
+            <folding />
+          </first_editor>
+          <second_editor>
+            <js_state />
+          </second_editor>
+        </state>
+      </provider>
+    </entry>
   </component>
 </project>
\ No newline at end of file
diff --git "a/\346\234\272\345\231\250\345\255\246\344\271\240/\351\233\206\346\210\220\345\255\246\344\271\240/GBDT.md" "b/\346\234\272\345\231\250\345\255\246\344\271\240/\351\233\206\346\210\220\345\255\246\344\271\240/GBDT.md"
index 7c97d22..83be8b5 100644
--- "a/\346\234\272\345\231\250\345\255\246\344\271\240/\351\233\206\346\210\220\345\255\246\344\271\240/GBDT.md"
+++ "b/\346\234\272\345\231\250\345\255\246\344\271\240/\351\233\206\346\210\220\345\255\246\344\271\240/GBDT.md"
@@ -14,8 +14,8 @@
 adaboost，gbdt等等
 
 # gbdt和boostingtree的boosting分别体现在哪里？
-- boostingtree利用基模型学习器，拟合的是当前模型与标签值的残差
-- gbdt利用基模型学习器，拟合的是当前模型与标签值的残差的负梯度
+- boostingtree利用基模型学习器，拟合的是mse（回归）或者指数损失函数（分类）
+- gbdt利用基模型学习器，拟合的是当前模型与标签值的损失函数的负梯度
 
 # gbdt的中的tree是什么tree？有什么特征？
 Cart tree，但是都是回归树
diff --git "a/\350\207\252\347\204\266\350\257\255\350\250\200\345\244\204\347\220\206/GloVe.md" "b/\350\207\252\347\204\266\350\257\255\350\250\200\345\244\204\347\220\206/GloVe.md"
index 2c6f87a..9692151 100644
--- "a/\350\207\252\347\204\266\350\257\255\350\250\200\345\244\204\347\220\206/GloVe.md"
+++ "b/\350\207\252\347\204\266\350\257\255\350\250\200\345\244\204\347\220\206/GloVe.md"
@@ -8,3 +8,6 @@
 - GloVe算法本身使用了全局信息，自然内存费的也就多一些
     - 公现矩阵，NXN的，N为词袋量
 - W2V的工程实现结果相对来说支持的更多，比如most_similarty等功能
+
+# 如何处理未出现词？
+按照词性进行已知词替换，\[unknow-n],\[unknow-a],\[unknow-v]...，然后再进行训练。实际去用的时候，判断词性后直接使用对应的unknown-?向量替代