Update README.md

README.md

@@ -73,12 +73,16 @@ def score_fn(score):
 def preprocess(example, tokenizer, score_fn):
     # Get number of moves made in the game.
     max_ply = len(example['moves']) 
-    pick_random_move = random.randint(0, max_ply-1)
+    pick_random_move = random.randint(0, max_ply-2)
 
     # Get the FEN, move and score for our random choice.
     fen = example['fens'][pick_random_move]
-    move = example['moves'][pick_random_move]
-    score = example['scores'][pick_random_move]
+
+    # We add +1 to the index to select the next candidate move.
+    # Same with the score, which is the evaluation of this move.
+    # Please read the section about the data format clearly!
+    move = example['moves'][pick_random_move + 1]
+    score = example['scores'][pick_random_move + 1]
 
     # Transform data into the format of your choice.
     example['fens'] = tokenizer(fen)
@@ -132,8 +136,12 @@ def preprocess(example, tokenizer, score_fn):
 
     # Get the FEN, move and score for our random choice.
     fen = example['fens'][pick_random_move]
-    move = example['moves'][pick_random_move]
-    score = example['scores'][pick_random_move]
+
+    # We add +1 to the index to select the next candidate move.
+    # Same with the score, which is the evaluation of this move.
+    # Please read the section about the data format clearly!
+    move = example['moves'][pick_random_move + 1]
+    score = example['scores'][pick_random_move + 1]
 
     # Transform data into the format of your choice.
     example['fens'] = tokenizer(fen)
@@ -159,12 +167,9 @@ dataset = dataset.map(preprocess, fn_kwargs={'tokenizer': tokenizer,
 # PyTorch dataloader
 dataloader = DataLoader(dataset, batch_size=1, num_workers=1)
 
-
 n = 0
 for batch in dataloader:
-
     # do stuff
-
     print(batch)
     break