datamaestro-text 2026.1.1__py3-none-any.whl → 2026.2.3__py3-none-any.whl

datamaestro_text/config/gov/nist/trec/tipster.py

@@ -20,165 +20,271 @@ See also https://trec.nist.gov/data/docs_eng.html and https://trec.nist.gov/data
 from datamaestro_text.data.ir.trec import TipsterCollection
 from datamaestro.download.links import linkfolder
 from datamaestro.definitions import (
+    Dataset,
     dataset,
 )
 from datamaestro.context import DatafolderPath
 
 # Store meta-information
-TIPSTER = dataset(TipsterCollection, url="https://catalog.ldc.upenn.edu/LDC93T3A")
+TIPSTER = dataset(url="https://catalog.ldc.upenn.edu/LDC93T3A")
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk2/AP")])
 @TIPSTER
-def ap88(documents):
+class Ap88(Dataset):
     """Associated Press document collection (1988)"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk2/AP")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk1/AP")])
 @TIPSTER
-def ap89(documents):
+class Ap89(Dataset):
     """Associated Press document collection (1989)"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk1/AP")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk3/AP")])
 @TIPSTER
-def ap90(documents):
+class Ap90(Dataset):
     """Associated Press document collection (1990)"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk3/AP")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk1/DOE")])
 @TIPSTER
-def doe1(documents):
+class Doe1(Dataset):
     """Department of Energy documents"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk1/DOE")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)
 
 
 # --- Wall Street Journal (1987-92)
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk1/WSJ/1987")])
 @TIPSTER
-def wsj87(documents):
+class Wsj87(Dataset):
     """Wall Street Journal (1987)"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk1/WSJ/1987")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk1/WSJ/1988")])
 @TIPSTER
-def wsj88(documents):
+class Wsj88(Dataset):
     """Wall Street Journal (1988)"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk1/WSJ/1988")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk1/WSJ/1989")])
 @TIPSTER
-def wsj89(documents):
+class Wsj89(Dataset):
     """Wall Street Journal (1989)"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk1/WSJ/1989")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk2/WSJ/1990")])
 @TIPSTER
-def wsj90(documents):
+class Wsj90(Dataset):
     """Wall Street Journal (1990)"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk2/WSJ/1990")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk2/WSJ/1991")])
 @TIPSTER
-def wsj91(documents):
+class Wsj91(Dataset):
     """Wall Street Journal (1991)"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk2/WSJ/1991")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk2/WSJ/1992")])
 @TIPSTER
-def wsj92(documents):
+class Wsj92(Dataset):
     """Wall Street Journal (1992)"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk2/WSJ/1992")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)
 
 
 # --- Federal Register (1988-89)
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk2/FR")])
 @TIPSTER
-def fr88(documents):
+class Fr88(Dataset):
     """Federal Register (1988)"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk2/FR")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk1/FR")])
 @TIPSTER
-def fr89(documents):
+class Fr89(Dataset):
     """Federal Register (1989)"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk1/FR")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk4/FR94")])
 @TIPSTER
-def fr94(documents):
+class Fr94(Dataset):
     """Federal Register (1994)"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk4/FR94")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)
 
 
 # # ZIFF (1988-92)
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk1/ZIFF")])
 @TIPSTER
-def ziff1(documents):
+class Ziff1(Dataset):
     """Information from the Computer Select disks (1989-90)"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk1/ZIFF")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk2/ZIFF")])
 @TIPSTER
-def ziff2(documents):
+class Ziff2(Dataset):
     """Information from the Computer Select disks (1989-90)"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk2/ZIFF")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk3/ZIFF")])
 @TIPSTER
-def ziff3(documents):
+class Ziff3(Dataset):
     """Information from the Computer Select disks (1990-91)"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk3/ZIFF")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk3/SJM")])
 @TIPSTER
-def sjm1(documents):
+class Sjm1(Dataset):
     """San Jose Mercury News (1991)"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk3/SJM")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk4/CR")])
 @TIPSTER
-def cr1(documents):
+class Cr1(Dataset):
     """TODO"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk4/CR")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk4/FT")])
 @TIPSTER
-def ft1(documents):
+class Ft1(Dataset):
     """Financial Times"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk4/FT")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk5/FBIS")])
 @TIPSTER
-def fbis1(documents):
+class Fbis1(Dataset):
     """Foreign Broadcast Information Service (1996)"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk5/FBIS")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)
 
 
-@linkfolder("documents", [DatafolderPath("gov.nist.trec.tipster", "Disk5/LATIMES")])
 @TIPSTER
-def la8990(documents):
+class La8990(Dataset):
     """Los Angeles Times (1989-90)"""
-    return {"path": documents}
+
+    DOCUMENTS = linkfolder(
+        "documents", [DatafolderPath("gov.nist.trec.tipster", "Disk5/LATIMES")]
+    )
+
+    def config(self) -> TipsterCollection:
+        return TipsterCollection.C(path=self.DOCUMENTS.path)

datamaestro_text/config/io/github/thunlp/fewrel.py

@@ -2,28 +2,17 @@
 
 from datamaestro.data import File
 from datamaestro.data.ml import Supervised
-from datamaestro.definitions import datatasks, datatags, dataset
-from datamaestro.download.single import filedownloader
+from datamaestro.definitions import Dataset, datatasks, datatags, dataset
+from datamaestro.download.single import FileDownloader
 from datamaestro.utils import HashCheck
 
 
 @datatags("unsupervised")
 @datatasks("information extraction")
-@filedownloader(
-    "train.json",
-    "https://github.com/thunlp/FewRel/raw/master/data/train_wiki.json",
-    checker=HashCheck("5e663e9c3f1bfbdb2de72696e9504fd7"),
-)
-@filedownloader(
-    "validation.json",
-    "https://github.com/thunlp/FewRel/raw/master/data/val_wiki.json",
-    checker=HashCheck("3f25573428c0332cb64b367a275ab0c7"),
-)
 @dataset(
-    Supervised,
     url="https://thunlp.github.io/1/fewrel1.html",
 )
-def v1(train, validation):
+class V1(Dataset):
     """FewRel 1.0 - a Few-shot Relation classification dataset
 
     FewRel is a Few-shot Relation classification dataset, which features 70, 000 natural
@@ -32,4 +21,20 @@ def v1(train, validation):
     Only the train and validation dataset are available. The test set is hidden
     for the leaderboard.
     """
-    return {"train": File.C(path=train), "validation": File.C(path=validation)}
+
+    TRAIN = FileDownloader(
+        "train.json",
+        "https://github.com/thunlp/FewRel/raw/master/data/train_wiki.json",
+        checker=HashCheck("5e663e9c3f1bfbdb2de72696e9504fd7"),
+    )
+    VALIDATION = FileDownloader(
+        "validation.json",
+        "https://github.com/thunlp/FewRel/raw/master/data/val_wiki.json",
+        checker=HashCheck("3f25573428c0332cb64b367a275ab0c7"),
+    )
+
+    def config(self) -> Supervised:
+        return Supervised.C(
+            train=File.C(path=self.TRAIN.path),
+            validation=File.C(path=self.VALIDATION.path),
+        )

datamaestro_text/config/io/metamind/research/wikitext.py

@@ -1,14 +1,24 @@
-from datamaestro.data import Base, File
+from datamaestro.data import File
 from datamaestro.definitions import (
+    Dataset,
     datatasks,
     datatags,
     dataset,
     metadataset,
 )
-from datamaestro.download.archive import zipdownloader
+from datamaestro.download.archive import ZipDownloader
 from datamaestro_text.data.text import TrainingText
 
 
+def _wikitext(data, type):
+    """Helper to build a TrainingText from data path and type."""
+    return TrainingText.C(
+        train=File.C(path=data / ("wiki.train.%s" % type)),
+        validation=File.C(path=data / ("wiki.valid.%s" % type)),
+        test=File.C(path=data / ("wiki.test.%s" % type)),
+    )
+
+
 @datatags("text")
 @datatasks("language modeling")
 @metadataset(TrainingText)
@@ -29,44 +39,52 @@ def WikiText(data, type):
 
     https://blog.einstein.ai/the-wikitext-long-term-dependency-language-modeling-dataset/
     """
-    return {
-        "train": File.C(path=data / ("wiki.train.%s" % type)),
-        "validation": File.C(path=data / ("wiki.valid.%s" % type)),
-        "test": File.C(path=data / ("wiki.test.%s" % type)),
-    }
+    return _wikitext(data, type)
 
 
-@zipdownloader(
-    "data", "https://s3.amazonaws.com/research.metamind.io/wikitext/wikitext-2-v1.zip"
-)
-@dataset(WikiText, id="2.tokens")
-def wikitext_2_words(data):
+@dataset(WikiText, id=".2.tokens")
+class Wikitext2Words(Dataset):
     """The small wikitext corpus, already tokenized"""
-    return WikiText(data, "tokens")
 
+    DATA = ZipDownloader(
+        "data",
+        "https://s3.amazonaws.com/research.metamind.io/wikitext/wikitext-2-v1.zip",
+    )
 
-@zipdownloader(
-    "data",
-    "https://s3.amazonaws.com/research.metamind.io/wikitext/wikitext-2-raw-v1.zip",
-)
-@dataset(WikiText, id="2.raw")
-def wikitext_2_raw(data):
+    def config(self) -> TrainingText:
+        return _wikitext(self.DATA.path, "tokens")
+
+
+@dataset(WikiText, id=".2.raw")
+class Wikitext2Raw(Dataset):
     """The small wikitext corpus (raw data)"""
-    return WikiText(data, "raw")
 
+    DATA = ZipDownloader(
+        "data",
+        "https://s3.amazonaws.com/research.metamind.io/wikitext/wikitext-2-raw-v1.zip",
+    )
 
-@zipdownloader(
-    "data", "https://s3.amazonaws.com/research.metamind.io/wikitext/wikitext-103-v1.zip"
-)
-@dataset(WikiText, id="103.tokens")
-def wikitext_103_words(data):
-    return WikiText(data, "tokens")
+    def config(self) -> TrainingText:
+        return _wikitext(self.DATA.path, "raw")
 
 
-@zipdownloader(
-    "data",
-    "https://s3.amazonaws.com/research.metamind.io/wikitext/wikitext-103-raw-v1.zip",
-)
-@dataset(WikiText, id="103.raw")
-def wikitext_103_raw(data):
-    return WikiText(data, "raw")
+@dataset(WikiText, id=".103.tokens")
+class Wikitext103Words(Dataset):
+    DATA = ZipDownloader(
+        "data",
+        "https://s3.amazonaws.com/research.metamind.io/wikitext/wikitext-103-v1.zip",
+    )
+
+    def config(self) -> TrainingText:
+        return _wikitext(self.DATA.path, "tokens")
+
+
+@dataset(WikiText, id=".103.raw")
+class Wikitext103Raw(Dataset):
+    DATA = ZipDownloader(
+        "data",
+        "https://s3.amazonaws.com/research.metamind.io/wikitext/wikitext-103-raw-v1.zip",
+    )
+
+    def config(self) -> TrainingText:
+        return _wikitext(self.DATA.path, "raw")

datamaestro_text/config/org/grouplens/movielens.py

@@ -1,53 +1,44 @@
-# tasks:
-#   - Recommendation
-#   - Collaborative Filtering
-
-# download:
-#   handler: /archive:Zip
-#   url: http://files.grouplens.org/datasets/movielens/ml-20m.zip
-#   size: 190M
-#   checksum: cd245b17a1ae2cc31bb14903e1204af3
-# ...
-# ---
-# id: tmdb
-# description: TMDB (The Movie database) download for MovieLens movies
-# download:
-#   handler: tmdb:MovieLens
-
-
-from datamaestro.definitions import dataset
-from datamaestro.download.archive import zipdownloader
+from datamaestro.definitions import Dataset, dataset
+from datamaestro.download.archive import ZipDownloader
 import datamaestro.data.csv as csv
 from datamaestro_text.data.recommendation import Movielens
 
 
-@zipdownloader(
-    "ds", "http://files.grouplens.org/datasets/movielens/ml-latest-small.zip"
-)
 @dataset(url="https://grouplens.org/datasets/movielens/latest/", timestamp=True)
-def small(ds) -> Movielens:
+class Small(Dataset):
     """MovieLens (small dataset)
 
     100,000 ratings and 3,600 tag applications applied to 9,000 movies by 600 users (as of 9/2018)
     """
-    return {
-        "ratings": csv.Generic.C(path=ds / "ratings.csv", names_row=0),
-        "links": csv.Generic.C(path=ds / "links.csv", names_row=0),
-        "movies": csv.Generic.C(path=ds / "movies.csv", names_row=0),
-        "tags": csv.Generic.C(path=ds / "tags.csv", names_row=0),
-    }
+
+    DS = ZipDownloader(
+        "ds", "http://files.grouplens.org/datasets/movielens/ml-latest-small.zip"
+    )
+
+    def config(self) -> Movielens:
+        return Movielens.C(
+            ratings=csv.Generic.C(path=self.DS.path / "ratings.csv", names_row=0),
+            links=csv.Generic.C(path=self.DS.path / "links.csv", names_row=0),
+            movies=csv.Generic.C(path=self.DS.path / "movies.csv", names_row=0),
+            tags=csv.Generic.C(path=self.DS.path / "tags.csv", names_row=0),
+        )
 
 
-@zipdownloader("ds", "http://files.grouplens.org/datasets/movielens/ml-latest.zip")
 @dataset(url="https://grouplens.org/datasets/movielens/latest/", timestamp=True)
-def full(ds) -> Movielens:
+class Full(Dataset):
     """MovieLens (full dataset)
 
     27,000,000 ratings and 1,100,000 tag applications applied to 58,000 movies by 280,000 users (as of 9/2018)
     """
-    return {
-        "ratings": csv.Generic.C(path=ds / "ratings.csv", names_row=0),
-        "links": csv.Generic.C(path=ds / "links.csv", names_row=0),
-        "movies": csv.Generic.C(path=ds / "movies.csv", names_row=0),
-        "tags": csv.Generic.C(path=ds / "tags.csv", names_row=0),
-    }
+
+    DS = ZipDownloader(
+        "ds", "http://files.grouplens.org/datasets/movielens/ml-latest.zip"
+    )
+
+    def config(self) -> Movielens:
+        return Movielens.C(
+            ratings=csv.Generic.C(path=self.DS.path / "ratings.csv", names_row=0),
+            links=csv.Generic.C(path=self.DS.path / "links.csv", names_row=0),
+            movies=csv.Generic.C(path=self.DS.path / "movies.csv", names_row=0),
+            tags=csv.Generic.C(path=self.DS.path / "tags.csv", names_row=0),
+        )

datamaestro_text/config/org/universaldependencies/french.py

@@ -1,7 +1,8 @@
-from datamaestro.definitions import dataset
+from datamaestro.definitions import Dataset, dataset
 from datamaestro.data.ml import Supervised
 from datamaestro_text.data.tagging import CoNLL_U
-from datamaestro.download.archive import zipdownloader
+from datamaestro.download.archive import ZipDownloader
+
 
 # --- gsd
 
@@ -22,22 +23,26 @@ from datamaestro.download.archive import zipdownloader
 #     path: fr_gsd-ud-test.conllu
 
 
-@zipdownloader(
-    "ds", "https://codeload.github.com/UniversalDependencies/UD_French-GSD/zip/master"
-)
 @dataset(url="https://github.com/UniversalDependencies/UD_French-GSD")
-def gsd(ds) -> Supervised:
+class Gsd(Dataset):
     """French GSD
 
     The UD_French-GSD was converted in 2015 from the content head version of the
     universal dependency treebank v2.0 (https://github.com/ryanmcd/uni-dep-tb). It
     is updated since 2015 independently from the previous source.
     """
-    return {
-        "train": CoNLL_U.C(path=ds / "fr_gsd-ud-train.conllu"),
-        "test": CoNLL_U.C(path=ds / "fr_gsd-ud-dev.conllu"),
-        "validation": CoNLL_U.C(path=ds / "fr_gsd-ud-test.conllu"),
-    }
+
+    DS = ZipDownloader(
+        "ds",
+        "https://codeload.github.com/UniversalDependencies/UD_French-GSD/zip/master",
+    )
+
+    def config(self) -> Supervised:
+        return Supervised.C(
+            train=CoNLL_U.C(path=self.DS.path / "fr_gsd-ud-train.conllu"),
+            test=CoNLL_U.C(path=self.DS.path / "fr_gsd-ud-dev.conllu"),
+            validation=CoNLL_U.C(path=self.DS.path / "fr_gsd-ud-test.conllu"),
+        )
 
 
 # --- partut

datamaestro_text/data/conversation/__init__.py

@@ -1,8 +1,8 @@
 from .base import (
-    AnswerEntry,
-    ConversationDataset,
-    ConversationHistory,
-    ConversationHistoryItem,
-    DecontextualizedItem,
-    EntryType,
+    AnswerEntry as AnswerEntry,
+    ConversationDataset as ConversationDataset,
+    ConversationHistory as ConversationHistory,
+    ConversationHistoryItem as ConversationHistoryItem,
+    DecontextualizedItem as DecontextualizedItem,
+    EntryType as EntryType,
 )

datamaestro_text/data/conversation/base.py

@@ -2,13 +2,13 @@ from abc import ABC, abstractmethod
 from enum import Enum
 from datamaestro_text.data.ir.base import IDItem, SimpleTextItem
 from experimaestro import Param
-from typing import Dict, Generic, Iterator, List, Optional, Sequence, Tuple
+from typing import Dict, Iterator, List, Optional, Sequence, Tuple
 from attr import define
 from datamaestro.record import record_type
 from datamaestro.data import Base
 from datamaestro.record import Record, Item
 from datamaestro_text.data.ir import TopicRecord, Topics
-from datamaestro_text.utils.iter import FactoryIterable, LazyList, RangeView
+from datamaestro_text.utils.iter import FactoryIterable, LazyList
 
 # ---- Basic types