Optimize code to use iter_files instead of globs

cantemist.py

@@ -22,12 +22,14 @@ mapped by clinical experts to a controlled terminology. Every tumor morphology
 mention is linked to an eCIE-O code (the Spanish equivalent of ICD-O).
 """
 
-import os
+import csv
+import os.path
+from collections import defaultdict
+from itertools import chain
 from pathlib import Path
 from typing import Dict, List, Tuple
 
 import datasets
-import pandas as pd
 
 from .bigbiohub import kb_features
 from .bigbiohub import text_features
@@ -236,113 +238,116 @@ class CantemistDataset(datasets.GeneratorBasedBuilder):
                 name=datasets.Split.TRAIN,
                 gen_kwargs={
                     "filepaths": {
-                        "task1": Path(
-                            os.path.join(data_dir, "train-set/cantemist-ner")
+                        "task1": dl_manager.iter_files(
+                            os.path.join(data_dir, "train-set", "cantemist-ner")
                         ),
-                        "task2": Path(
-                            os.path.join(data_dir, "train-set/cantemist-norm")
-                        ),
-                        "task3": Path(
-                            os.path.join(data_dir, "train-set/cantemist-coding")
+                        "task2": dl_manager.iter_files(
+                            os.path.join(data_dir, "train-set", "cantemist-norm")
                         ),
+                        "task3": [
+                            os.path.join(
+                                data_dir,
+                                "train-set",
+                                "cantemist-coding",
+                                "train-coding.tsv",
+                            )
+                        ],
                     },
-                    "split": "train",
                 },
             ),
             datasets.SplitGenerator(
                 name=datasets.Split.TEST,
                 gen_kwargs={
                     "filepaths": {
-                        "task1": Path(os.path.join(data_dir, "test-set/cantemist-ner")),
-                        "task2": Path(
-                            os.path.join(data_dir, "test-set/cantemist-norm")
+                        "task1": dl_manager.iter_files(
+                            os.path.join(data_dir, "test-set", "cantemist-ner")
                         ),
-                        "task3": Path(
-                            os.path.join(data_dir, "test-set/cantemist-coding")
+                        "task2": dl_manager.iter_files(
+                            os.path.join(data_dir, "test-set", "cantemist-norm")
                         ),
+                        "task3": [
+                            os.path.join(
+                                data_dir,
+                                "test-set",
+                                "cantemist-coding",
+                                "test-coding.tsv",
+                            )
+                        ],
                     },
-                    "split": "test",
                 },
             ),
             datasets.SplitGenerator(
                 name=datasets.Split.VALIDATION,
                 gen_kwargs={
                     "filepaths": {
-                        "task1_set1": Path(
-                            os.path.join(data_dir, "dev-set1/cantemist-ner")
-                        ),
-                        "task1_set2": Path(
-                            os.path.join(data_dir, "dev-set2/cantemist-ner")
-                        ),
-                        "task2_set1": Path(
-                            os.path.join(data_dir, "dev-set1/cantemist-norm")
-                        ),
-                        "task2_set2": Path(
-                            os.path.join(data_dir, "dev-set2/cantemist-norm")
-                        ),
-                        "task3_set1": Path(
-                            os.path.join(data_dir, "dev-set1/cantemist-coding")
+                        "task1": chain(
+                            dl_manager.iter_files(
+                                os.path.join(data_dir, "dev-set1", "cantemist-ner")
+                            ),
+                            dl_manager.iter_files(
+                                os.path.join(data_dir, "dev-set2", "cantemist-ner")
+                            ),
                         ),
-                        "task3_set2": Path(
-                            os.path.join(data_dir, "dev-set2/cantemist-coding")
+                        "task2": chain(
+                            dl_manager.iter_files(
+                                os.path.join(data_dir, "dev-set1", "cantemist-norm")
+                            ),
+                            dl_manager.iter_files(
+                                os.path.join(data_dir, "dev-set2", "cantemist-norm")
+                            ),
                         ),
+                        "task3": [
+                            os.path.join(
+                                data_dir,
+                                "dev-set1",
+                                "cantemist-coding",
+                                "dev1-coding.tsv",
+                            ),
+                            os.path.join(
+                                data_dir,
+                                "dev-set2",
+                                "cantemist-coding",
+                                "dev2-coding.tsv",
+                            ),
+                        ],
                     },
-                    "split": "dev",
                 },
             ),
         ]
 
-    def _generate_examples(self, filepaths, split: str) -> Tuple[int, Dict]:
+    def _generate_examples(self, filepaths) -> Tuple[int, Dict]:
         """
         This method handles input defined in _split_generators to yield (key, example) tuples from the dataset.
         Method parameters are unpacked from `gen_kwargs` as given in `_split_generators`.
         """
 
-        if split != "dev":
-            txt_files_task1 = list(filepaths["task1"].glob("*txt"))
-            txt_files_task2 = list(filepaths["task2"].glob("*txt"))
-            tsv_file_task3 = Path(
-                os.path.join(filepaths["task3"], f"{split}-coding.tsv")
-            )
-            task3_df = pd.read_csv(tsv_file_task3, sep="\t", header=None)
-        else:
-            txt_files_task1, txt_files_task2, dfs = [], [], []
-            for i in range(1, 3):
-                txt_files_task1 += list(filepaths[f"task1_set{i}"].glob("*txt"))
-                txt_files_task2 += list(filepaths[f"task2_set{i}"].glob("*txt"))
-                tsv_file_task3 = Path(
-                    os.path.join(filepaths[f"task3_set{i}"], f"{split}{i}-coding.tsv")
-                )
-                df = pd.read_csv(tsv_file_task3, sep="\t", header=0)
-                dfs.append(df)
-            task3_df = pd.concat(dfs)
-
         if self.config.schema == "source" or self.config.schema == "bigbio_text":
-            task3_dict = {}
-            for idx, row in task3_df.iterrows():
-                file, code = row[0], row[1]
-                if file not in task3_dict:
-                    task3_dict[file] = [code]
-                else:
-                    task3_dict[file] += [code]
+            task3_dict = defaultdict(list)
+            for file_path in filepaths["task3"]:
+                with open(file_path, newline="", encoding="utf-8") as f:
+                    reader = csv.DictReader(f, delimiter="\t")
+                    for row in reader:
+                        task3_dict[row["file"]].append(row["code"])
 
         if self.config.schema == "source":
-            for guid, txt_file in enumerate(txt_files_task2):
-                example = parse_brat_file(txt_file, parse_notes=True)
-                if example["document_id"] in task3_dict:
-                    example["labels"] = task3_dict[example["document_id"]]
-                else:
-                    example[
-                        "labels"
-                    ] = (
-                        []
-                    )  # few cases where subtrack 3 has no codes for the current document
+            for guid, file_path in enumerate(filepaths["task2"]):
+                if os.path.splitext(file_path)[-1] != ".txt":
+                    continue
+                example = parse_brat_file(
+                    Path(file_path), annotation_file_suffixes=[".ann"], parse_notes=True
+                )
+                # consider few cases where subtrack 3 has no codes for the current document
+                example["labels"] = task3_dict.get(example["document_id"], [])
                 example["id"] = str(guid)
                 yield guid, example
 
         elif self.config.schema == "bigbio_kb":
-            for guid, txt_file in enumerate(txt_files_task2):
-                parsed_brat = parse_brat_file(txt_file, parse_notes=True)
+            for guid, file_path in enumerate(filepaths["task2"]):
+                if os.path.splitext(file_path)[-1] != ".txt":
+                    continue
+                parsed_brat = parse_brat_file(
+                    Path(file_path), annotation_file_suffixes=[".ann"], parse_notes=True
+                )
                 example = brat_parse_to_bigbio_kb(parsed_brat)
                 example["id"] = str(guid)
                 for i in range(0, len(example["entities"])):
@@ -354,14 +359,16 @@ class CantemistDataset(datasets.GeneratorBasedBuilder):
                 yield guid, example
 
         elif self.config.schema == "bigbio_text":
-            for guid, txt_file in enumerate(txt_files_task1):
-                parsed_brat = parse_brat_file(txt_file, parse_notes=False)
-                if parsed_brat["document_id"] in task3_dict:
-                    labels = task3_dict[parsed_brat["document_id"]]
-                else:
-                    labels = (
-                        []
-                    )  # few cases where subtrack 3 has no codes for the current document
+            for guid, file_path in enumerate(filepaths["task1"]):
+                if os.path.splitext(file_path)[-1] != ".txt":
+                    continue
+                parsed_brat = parse_brat_file(
+                    Path(file_path),
+                    annotation_file_suffixes=[".ann"],
+                    parse_notes=False,
+                )
+                # consider few cases where subtrack 3 has no codes for the current document
+                labels = task3_dict.get(parsed_brat["document_id"], [])
                 example = {
                     "id": str(guid),
                     "document_id": parsed_brat["document_id"],