WIP training Batching

Project_Model/Libs/BPE/Classes/TokeNanoCore.py

@@ -6,6 +6,7 @@ from ..Classes import NanoSocratesSpecial
 
 from ..Utils import special_regex_maker
 from ..Enums import TokenType
+from ..Enums import SpecialToken
 
 
 class TokeNanoCore:
@@ -44,6 +45,27 @@ class TokeNanoCore:
                 output.extend(self.__bpe_encoder.encode(piece))
 
         return output
+    
+
+
+    def encode_incomplete_string(self, corpus: str) -> list[int]:
+        """
+        Encode string which don't end with a special token
+        """
+        corpus = corpus + SpecialToken.CORPUS_END.value
+        output: list[int] = []
+        for piece, token_type in self.__splitter.split_text(corpus):
+
+            if token_type == TokenType.SPECIAL:
+                output.extend(self.__special_encoder.encode(piece))
+
+            # slow but clear
+            if token_type == TokenType.BPE:
+                output.extend(self.__bpe_encoder.encode(piece))
+
+        return output[:-1]
+
+
 
     def decode(self, corpus: list[int]) -> str:
         output_str = ""

Project_Model/Libs/Batch/Classes/Batcher.py

@@ -1,10 +1,12 @@
 import pandas as pd
-
-from BPE import TokeNanoCore as Tokenizer
+from pathlib import Path
+import Project_Model.Libs.BPE as BPE
+#from BPE import TokeNanoCore as Tokenizer
+from Scripts.Libs.CleaningPipeline.special_token import SpecialToken
 
 class Batcher:
 
-    def __init__(self, dataset_path: str, batch_size:int, tokenizer: Tokenizer) -> None:
+    def __init__(self, dataset_path: str, batch_size:int, tokenizer: BPE.TokeNanoCore) -> None:
         # ABSTRACT, TRIPLE 
         # tasks: 
         #   rdf2text: X: TRIPLE, Y: ABSTRACT 
@@ -21,24 +23,50 @@ class Batcher:
             # each batch get 4 transformation for the 4 tasks and then shuffled
             # now a batch is ["Abstract"], ["Triples"]
             # tokenize the strings:
-            tokenized_batch 
-            tokenized_batch[["Abstract","Triples"]] = batch[["Abstract","Triples"]].map(lambda t: self._tokenizer.encode(t))
+            # batch = batch.drop(columns=['MovieID'])
+            tokenized_batch = pd.DataFrame()
+            # bho = batch.map(lambda x: self._tokenizer.encode(x))
+            tokenized_batch[["Abstract","RDFs"]] = batch[["Abstract","RDFs"]].map(
+                lambda t: self._tokenizer.encode_incomplete_string(t))
+            
             # ??? i hope this works, later will be tested
-            rdf2_txt_batch = self.__rdf2txt_transformation(tokenized_batch)
+            rdf2txt_batch = self.__rdf2txt_transformation(tokenized_batch)
+            txt2rdf_batch = self.__txt2rdf_transformation(tokenized_batch)
+
+            output = pd.concat([rdf2txt_batch,txt2rdf_batch],ignore_index=True)
+            output.sample(frac=1).reset_index(drop=True)
+            yield output
             
 
-
+    def __random_subset_rdfs(self, batch: pd.DataFrame):
+        batch["RDFs"] = batch["RDFs"].map(
+            lambda x: x.split(SpecialToken.START_TRIPLE.value)[1:]
+        )
 
     def __rdf2txt_transformation(self, batch: pd.DataFrame):
         # rename ["Triples"] as ["X"]
         # rename ["Abstract"] as ["Y"]
         # return just them
-        batch = batch.rename(columns={"Triples": "X", "Abstract": "Y"})
+        batch = batch.rename(columns={"RDFs": "X", "Abstract": "Y"})
         return batch[["X", "Y"]] #.sample(frac=1).reset_index(drop=True)
 
 
     def __txt2rdf_transformation(self, batch: pd.DataFrame):
-        batch = batch.rename(columns={ "Abstract": "X","Triples": "Y"})
+        batch = batch.rename(columns={ "Abstract": "X","RDFs": "Y"})
         return batch[["X", "Y"]]# .sample(frac=1).reset_index(drop=True)
     
-    def __masking()
+    #def __masking()
+        
+
+
+DATASET_PATH = "Assets/Dataset/Tmp/rdf_text.csv"
+VOCABULARY_path = "Assets/Dataset/Tmp/trimmed.json"
+VOCABULARY = BPE.load_nanos_vocabulary(Path(VOCABULARY_path))
+SPECIAL_TOKEN_LIST = [token.value for token in SpecialToken]
+TOKENANO = BPE.TokeNanoCore(VOCABULARY, SPECIAL_TOKEN_LIST)
+
+prova = "<ABS>Cactus Flower is a 1969 American screwball comedy film directed by Gene Saks, and starring Walter Matthau, Ingrid Bergman and Goldie Hawn, who won an Academy Award for her performance.The screenplay was adapted by I. A. L. Diamond from the 1965 Broadway play of the same title written by Abe Burrows, which, in turn, is based on the French play Fleur de cactus by Pierre Barillet and Jean-Pierre Gredy. Cactus Flower was the ninth highest-grossing film of 1969."
+print(TOKENANO.encode(prova))
+batcher = Batcher(DATASET_PATH,3,TOKENANO)
+for batch in batcher.get_batch():
+    print(batch)