add

squall.py

@@ -155,7 +155,18 @@ class Squall(datasets.GeneratorBasedBuilder):
         squall_full = filepath["squall"]
         dev_ids = filepath["dev-" + self.config.name]
         test = filepath["wtq-test"]
-
+        
+        # transform the original squall data structure (list of things) to dict 
+        def transform(sample, sample_key, keys):
+            cols = {}
+            n_col = len(sample[sample_key])
+            for k in range(len(keys)):
+                tmp = []
+                for j in range(n_col):
+                    tmp.append(sample[sample_key][j][k])
+                cols[keys[k]] = tmp
+            return cols
+        
         if split_key != 'test':
             with open(squall_full, encoding="utf-8") as f:
                 squall_full_data = json.load(f)
@@ -236,16 +247,6 @@ class Squall(datasets.GeneratorBasedBuilder):
                 instance["numbers"] = numbers
                 instance["has_number"] = has_number
 
-            def transform(sample, sample_key, keys):
-                cols = {}
-                n_col = len(sample[sample_key])
-                for k in range(len(keys)):
-                    tmp = []
-                    for j in range(n_col):
-                        tmp.append(sample[sample_key][j][k])
-                    cols[keys[k]] = tmp
-                return cols
-
             with open(dev_ids) as f:
                 dev_ids = json.load(f)
             if split_key == "train":
@@ -254,7 +255,6 @@ class Squall(datasets.GeneratorBasedBuilder):
                 set = [x for x in squall_full_data if x["tbl"] in dev_ids]
             idx = 0
             for sample in set:
-
                 # transform columns
                 keys = ["raw_header", "tokenized_header", "column_suffixes", "column_dtype", "example"]
                 cols = transform(sample, "columns", keys)
@@ -267,7 +267,6 @@ class Squall(datasets.GeneratorBasedBuilder):
                 # transform ralign
                 keys = ["aligned_sql_token_type", "aligned_sql_token_info"]
                 raligns = transform(sample, "nl_ralign", keys)
-
                 yield idx, {
                     "nt": sample["nt"],
                     "tbl": sample["tbl"],
@@ -289,11 +288,9 @@ class Squall(datasets.GeneratorBasedBuilder):
                 test_data = json.load(f)
             idx = 0
             for sample in test_data:
-                
                 # transform columns
                 keys = ["raw_header", "tokenized_header", "column_suffixes", "column_dtype", "example"]
                 cols = transform(sample, "columns", keys)
-
                 yield idx, {
                     "nt": sample["nt"],
                     "tbl": sample["tbl"],