Adjust directory structure for multiple splits and configurations

.gitattributes

@@ -56,4 +56,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.webp filter=lfs diff=lfs merge=lfs -text
 *.json filter=lfs diff=lfs merge=lfs -text
 *.gz filter=lfs diff=lfs merge=lfs -text
-stats/stats.json filter= diff= merge= text
+# Stats
+**/stats/stats.json filter= diff= merge= text

.gitignore

@@ -2,6 +2,6 @@
 out
 original
 # Generated data
-pack/gen
+resources/gen
 *.pyc
 scripts

dataset.yaml

@@ -27,7 +27,7 @@ _output_steps:
       type: FastChatOutput
       sync: main
       config:
-        path: "out/bluemoon.json"
+        path: "out/data/full/train/bluemoon.json"
         indent: 2
         #includeMessageId: true
 
@@ -153,10 +153,10 @@ _steps:
   - step: &spelling
       type: ExactReplace
       description: "Spelling mistakes and typos."
-      pack: [ 'pack/spelling.csv' ]
+      pack: [ 'resources/spelling.csv' ]
       config:
         caseSensitive: false
-        packedReplacements: [ 'pack/spelling.csv' ]
+        packedReplacements: [ 'resources/spelling.csv' ]
 
   - step: &empty
       type: FullMatch
@@ -182,19 +182,19 @@ _steps:
   - step: &edge_cases
       type: ExactReplace
       description: "Remove a few edge cases."
-      pack: [ 'pack/edge_cases.csv' ]
+      pack: [ 'resources/edge_cases.csv' ]
       config:
-        packedReplacements: [ 'pack/edge_cases.csv' ]
+        packedReplacements: [ 'resources/edge_cases.csv' ]
 
   - step: &stats_output
       type: StatsOutput
       description: "Output statistics"
       sync: main
       config:
-        path: 'stats/stats.json'
-        vocabPath: 'stats/vocabulary.json.gz'
+        path: 'out/data/full/train/stats/stats.json'
+        vocabPath: 'out/data/full/train/stats/vocabulary.json.gz'
         gZipVocabulary: true
-        conversationsPath: 'stats/conversations.json.gz'
+        conversationsPath: 'out/data/full/train/stats/conversations.json.gz'
         gZipConversations: true
 _passes:
   - pass: &clean
@@ -211,7 +211,7 @@ _passes:
                *bluemoon_fastchat, *stats_output
       ]
       artifacts:
-        - file: 'out/bluemoon.json'
+        - file: 'out/data/full/train/bluemoon.json'
           sha512: 70ddecc11f8b280a5d641ce502e91a2392aa9dc12e0704617edb1599e4c235b6e8015987f109a09c83746f717cec715fcad04587cc066897583eaf2bd439616b
 
 passes: [ *clean ]

debug.yaml

@@ -56,10 +56,10 @@ _debug_steps:
 
   - step: &strip_usernames
       type: ExactReplace
-      pack: [ 'pack/gen/usernames.csv' ]
+      pack: [ 'resources/gen/usernames.csv' ]
       config:
         caseSensitive: true
-        packedReplacements: [ 'pack/gen/usernames.csv' ]
+        packedReplacements: [ 'resources/gen/usernames.csv' ]
 
 
 passes:
@@ -79,19 +79,19 @@ passes:
         - type: FileConcatenate
           sync: main
           config:
-            path: 'pack/gen/usernames_all.csv'
+            path: 'resources/gen/usernames_all.csv'
             files: [ 'out/usernames_*.csv' ]
             ignoreDuplicateLines: true
             eol: "\r\n"
             delete: true
       artifacts:
-        - file: 'pack/gen/usernames_all.csv'
+        - file: 'resources/gen/usernames_all.csv'
           sha512: ffcc8aa980f4d8b0cace9092280d07215337cc316cb3fa45742a5f5a735d21a31f581d27884a414df21ac99616d14c7ebec1ab68a5766c10faa4b1564ceb7c20
 
   # Count the occurrences of usernames in messages in the source data
   - pass: &count_usernames
       required:
-        - path: "pack/gen/usernames_all.csv"
+        - path: "resources/gen/usernames_all.csv"
           sha512: ffcc8aa980f4d8b0cace9092280d07215337cc316cb3fa45742a5f5a735d21a31f581d27884a414df21ac99616d14c7ebec1ab68a5766c10faa4b1564ceb7c20
         - path: "data/rentry/bluemoon/bluemoon-fanbased-roleplays-1x1.csv"
           sha512: f31d6bd278bc4211736d6aace3917cd0c1d0143bec9bf9d07054f5f9b32060e17399b6ea0935774b2271ac45a88309a60ef8eec8c4ac5283d1b353a255529cc5
@@ -99,10 +99,10 @@ passes:
         - *bluemoon
         - type: StatsCountOccurrences
           description: "Count occurrences of usernames."
-          pack: [ 'pack/gen/usernames_all.csv' ]
+          pack: [ 'resources/gen/usernames_all.csv' ]
           config:
             path: 'out/usernames_counted_%worker%.csv'
-            packedPatterns: [ 'pack/gen/usernames_all.csv' ]
+            packedPatterns: [ 'resources/gen/usernames_all.csv' ]
             countCol: 2
             regex: false
             min: 1
@@ -110,9 +110,9 @@ passes:
           description: "Merge username count csv files"
           sync: main
           config:
-            path: 'pack/gen/usernames.csv'
+            path: 'resources/gen/usernames.csv'
             files: [ 'out/usernames_counted_*.csv' ]
             cols: [ 2 ]
             min: 1
       artifacts:
-        - file: 'pack/gen/usernames.csv'
+        - file: 'resources/gen/usernames.csv'