datachain 0.2.9__py3-none-any.whl → 0.2.10__py3-none-any.whl

{datachain-0.2.9.dist-info → datachain-0.2.10.dist-info}/RECORD

@@ -1,14 +1,14 @@
-datachain/__init__.py,sha256=WTZQycUOpP1b-Ry_Qje5HH0EE14ptne-ZiQQ5070UMA,798
+datachain/__init__.py,sha256=L5IlHOD4AaHkV7P5dbUwdq90I3bGFLtOghoZ1WVFGcs,841
 datachain/__main__.py,sha256=hG3Y4ARGEqe1AWwNMd259rBlqtphx1Wk39YbueQ0yV8,91
 datachain/asyn.py,sha256=CKCFQJ0CbB3r04S7mUTXxriKzPnOvdUaVPXjM8vCtJw,7644
 datachain/cache.py,sha256=FaPWrqWznPffmskTb1pdPkt2jAMMf__9FC2zEnP0vDU,4022
-datachain/cli.py,sha256=lInqYMhk8YuPY-ZWkfWZmE-ZmdIChJgbs305-a_MWpo,32457
+datachain/cli.py,sha256=gikzwEXTDKyzY1xOAUziXN2-OVqnOhDMJTd7SHq0Jxc,32406
 datachain/cli_utils.py,sha256=jrn9ejGXjybeO1ur3fjdSiAyCHZrX0qsLLbJzN9ErPM,2418
 datachain/config.py,sha256=PfC7W5yO6HFO6-iMB4YB-0RR88LPiGmD6sS_SfVbGso,1979
 datachain/dataset.py,sha256=MZezyuJWNj_3PEtzr0epPMNyWAOTrhTSPI5FmemV6L4,14470
 datachain/error.py,sha256=GY9KYTmb7GHXn2gGHV9X-PBhgwLj3i7VpK7tGHtAoGM,1279
 datachain/listing.py,sha256=sX8vZNzAzoTel1li6VJiYeHUJwseUERVEoW9D5P7tII,8192
-datachain/node.py,sha256=fHe7k5ajI2g2qnzsG-_NQR_T-QdBYctVeEa8c8dsu_Y,5703
+datachain/node.py,sha256=fsQDJUmRMSRHhL1u6qQlWgreHbH760Ls-yDzFLhbW-U,5724
 datachain/nodes_fetcher.py,sha256=kca19yvu11JxoVY1t4_ydp1FmchiV88GnNicNBQ9NIA,831
 datachain/nodes_thread_pool.py,sha256=ZyzBvUImIPmi4WlKC2SW2msA0UhtembbTdcs2nx29A0,3191
 datachain/progress.py,sha256=7_8FtJs770ITK9sMq-Lt4k4k18QmYl4yIG_kCoWID3o,4559
@@ -16,7 +16,7 @@ datachain/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 datachain/storage.py,sha256=RiSJLYdHUjnrEWkLBKPcETHpAxld_B2WxLg711t0aZI,3733
 datachain/utils.py,sha256=12yQAV8tfyCHqp_xJcJBeNnr1L_BO8e2bOPyXdM68gs,10759
 datachain/catalog/__init__.py,sha256=g2iAAFx_gEIrqshXlhSEbrc8qDaEH11cjU40n3CHDz4,409
-datachain/catalog/catalog.py,sha256=pulKGJgAmxqSmFqBhA-J0wCKdBqGX4vqpV0cAvV6vUw,79578
+datachain/catalog/catalog.py,sha256=A5W9Ffoz1lZkzl6A3igaMC5jrus8VIYVLJLX8JTVKrk,79603
 datachain/catalog/datasource.py,sha256=D-VWIVDCM10A8sQavLhRXdYSCG7F4o4ifswEF80_NAQ,1412
 datachain/catalog/loader.py,sha256=GJ8zhEYkC7TuaPzCsjJQ4LtTdECu-wwYzC12MikPOMQ,7307
 datachain/catalog/subclass.py,sha256=B5R0qxeTYEyVAAPM1RutBPSoXZc8L5mVVZeSGXki9Sw,2096
@@ -32,51 +32,53 @@ datachain/data_storage/db_engine.py,sha256=rgBuqJ-M1j5QyqiUQuJRewctuvRRj8LBDL54-
 datachain/data_storage/id_generator.py,sha256=VlDALKijggegAnNMJwuMETJgnLoPYxpkrkld5DNTPQw,3839
 datachain/data_storage/job.py,sha256=w-7spowjkOa1P5fUVtJou3OltT0L48P0RYWZ9rSJ9-s,383
 datachain/data_storage/metastore.py,sha256=y-4fYvuOPnWeYxAvqhDnw6CdlTvQiurg0Gg4TaG9LR0,54074
-datachain/data_storage/schema.py,sha256=bY3q2OUaUraos0s5BnwWkhgce8YpeNmIl7M1ifshoes,8074
+datachain/data_storage/schema.py,sha256=hUykqT-As-__WffMdWTrSZwv9k5EYYowRke3OENQ3aY,8102
 datachain/data_storage/serializer.py,sha256=6G2YtOFqqDzJf1KbvZraKGXl2XHZyVml2krunWUum5o,927
-datachain/data_storage/sqlite.py,sha256=F68Q_AIqNAObZ5kJ0GnBqRC6e2D2sRehkQo8UzrHgtI,25079
-datachain/data_storage/warehouse.py,sha256=h35JiJoCGtwkMctis_x3NHxkwEejX5sIWvJOluZxrOI,33132
-datachain/image/__init__.py,sha256=g3l7vJFzg0-s5OAmBtGargsxt12TuKU4Ex6S0fOmEeY,101
+datachain/data_storage/sqlite.py,sha256=cIYobczfH72c4l-iMkxpkgcTuuvvT8Xi64iP7Zr3Skw,25084
+datachain/data_storage/warehouse.py,sha256=UbD37_jqaM4BY2SsQaTiJre-eSa7HcPejrTp936L080,33170
 datachain/lib/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-datachain/lib/arrow.py,sha256=FF3WWUOjB6Prw8ygfiLsrVfrdob0S01lPzEazuGqoO8,2556
-datachain/lib/cached_stream.py,sha256=t2ifK0hZVZiVn0MQ8D3FaFK1-qK84TwJW2Dw1SRsw9g,1066
-datachain/lib/claude.py,sha256=iAauA1zNVNONpLzUo1t0QN5PZ5Ot6cZkfib7Ka_c638,1969
-datachain/lib/clip.py,sha256=rDeZlFGs0DXBlpmh5ZQJhR9Sz13bWAZGQjfYm1hsUI4,5388
-datachain/lib/dc.py,sha256=D3cgib-U0Mo0x5wEK1_NfgymAldHqCvooZwtyohi53Q,34426
-datachain/lib/feature.py,sha256=iMwbMyQUyjRUeB-vhAucnx59kNSVvX_xEChTW5B9klY,12244
-datachain/lib/feature_registry.py,sha256=K3jGQzBp2HZDjR9hdGe1BZaXOAne8RpkCRRQdTVjkTs,1622
-datachain/lib/feature_utils.py,sha256=2yLdZd9o4AJ5QQX7kqgbCxCT78aT7HE12CLxQ6QRpbc,4982
-datachain/lib/file.py,sha256=LGBwC7tFU7VcSWk5kjPpEWPBQas5me69L2uTDNvYXGM,8326
-datachain/lib/gpt4_vision.py,sha256=idyXVZVWzltstGaVIu5RYE5UNbdqcPEjIWy81O1MwkM,2922
-datachain/lib/hf_image_to_text.py,sha256=HiPSWzJRDT-vnz9DXJbJBNCMNl9wmpxiSS3PbbVz8SE,3310
-datachain/lib/hf_pipeline.py,sha256=f0AH_XCziOF1OKN3d1w1swTBLaeajMJ8xgdsX37i5-o,2287
-datachain/lib/image.py,sha256=ZYfDqr9p-RRmWBeWFQwXLS1J3vQS616ykfMUvQVpqBY,2717
-datachain/lib/image_transform.py,sha256=NXWtnVOcofWBgl_YMxb4ABpaT7JTBMx7tLKvErH1IC4,3024
-datachain/lib/iptc_exif_xmp.py,sha256=xrbxFeY-wRP6T5JsUgE3EXfTxKvZVymRaRD_VIfxD0A,2236
-datachain/lib/meta_formats.py,sha256=wIVVLRLp45Zk4vjZRd_P1UtD24vpDCb-vILWtcsACwk,6630
-datachain/lib/pytorch.py,sha256=Ea1sXhborF6zcywQjLpXgKnbr1lTez4Bfu3m0Gr78FI,5843
+datachain/lib/arrow.py,sha256=ttSiH8Xr08zxypAa3-BNTxMO2NBuZfYICwmG1qQwvWU,3268
+datachain/lib/clip.py,sha256=YRa15Whnn6C8BMA-OAu0mYjc4h9i_n7pffRGdtfrTBA,5222
+datachain/lib/data_model.py,sha256=DpV_-1JqJptCf0w4cnzPlHm5Yl4FQaveRgVCDZFaHXs,2012
+datachain/lib/dc.py,sha256=Px7zj1mrAsL3sBLu1pezssBQkvY0YAoGJm4VbT2yRwc,34699
+datachain/lib/feature_registry.py,sha256=LUrBvDom-k1shFuCv46-OdgntbIUQ5008oyIS0iPM6Q,2298
+datachain/lib/file.py,sha256=Uik1sq2l-uknpikH4Gdm7ZR0EcQYP2TrNg-urECjbW4,8304
+datachain/lib/gpt4_vision.py,sha256=CZ-a64olZNp9TNmLGngmbN6b02UYImzwK3dPClnjxTI,2716
+datachain/lib/hf_image_to_text.py,sha256=uVl4mnUl8gnHrJ3wfSZlxBevH-cxqOswxLArLAHxRrE,3077
+datachain/lib/hf_pipeline.py,sha256=MBFzixVa25_6QVR9RyOq8Rr9UIQ-sFVcBHducx_sZcY,2069
+datachain/lib/image.py,sha256=K0n_P7kmobWTgxe-rDbr5yY3vBrOPnseziE3DXwFFVo,2325
+datachain/lib/image_transform.py,sha256=hfgvIrSMGBx_MEXECyvrFoO1NyPBHoDb28j2lT2dsf8,2953
+datachain/lib/iptc_exif_xmp.py,sha256=rmlxjOmAP31OCgbGBAwIgd1F_6QVBoSWsOPG6UsBg_w,2007
+datachain/lib/meta_formats.py,sha256=SF7UPPe-U-1HL6DBO1NfwZLIChjkHrHasgHf5ztCUoU,6436
+datachain/lib/model_store.py,sha256=JFpI1P0WFpsO6eAU49AdWmff5T8azqLrqOMB08pYJjg,2331
+datachain/lib/pytorch.py,sha256=7fd2g0dI9zrMfRl3IVwIvXRH0v6TwSAyZGAbqKdEjcI,5505
 datachain/lib/settings.py,sha256=6Nkoh8riETrftYwDp3aniK53Dsjc07MdztL8N0cW1D8,2849
-datachain/lib/signal_schema.py,sha256=hD56hyO1H3A5H2oyTUwPcNu6UOQ_XY0DeA0nrXBqFaU,11492
-datachain/lib/text.py,sha256=PUT1O0jNJoQGsuhff2LgDpzTWk2eMdwIKqEDBrE448M,1307
-datachain/lib/udf.py,sha256=axMvqYz4tdyg_C3nyuOcDsu3Aqr19jWv2vl54U_8LQM,6595
-datachain/lib/udf_signature.py,sha256=CUKgoVpM_N8CgvMncpAw2RYchoiJdAGdDSdluoP0hIk,7161
+datachain/lib/signal_schema.py,sha256=xzVHauGrhGcS5aOE1UMqT5YjJeZIMAZYQq76tZShhnY,13550
+datachain/lib/text.py,sha256=d2V-52cqzVm5OT68BcLYyHrglvFMVR5DPzsbtRRv3D0,1063
+datachain/lib/udf.py,sha256=RqCiGuNKL5P8eS84s_mmVYjK1gvkuRYdnIKm9qe-i2U,9698
+datachain/lib/udf_signature.py,sha256=R81QqZseG_xeBFzJSgt-wrTQeUU-1RrWkHckLm_HEUU,7135
 datachain/lib/unstructured.py,sha256=9Y6rAelXdYqkNbPaqz6DhXjhS8d6qXcP0ieIsWkzvkk,1143
 datachain/lib/utils.py,sha256=5-kJlAZE0D9nXXweAjo7-SP_AWGo28feaDByONYaooQ,463
 datachain/lib/vfile.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-datachain/lib/webdataset.py,sha256=GWB_pocfRZGoU4Lhd7Wh3hx2Rnm_fJWXX4S_zXJIEmk,8286
-datachain/lib/webdataset_laion.py,sha256=HAtSCbVvEQqzKkoRamRxDKaQALSB3QmJRU2yWRFNxwY,2147
+datachain/lib/webdataset.py,sha256=eqIDSqfBOhEK43JMp-6lYdYy2x3Ge5lwqR-hKGV8aG0,8259
+datachain/lib/webdataset_laion.py,sha256=PQP6tQmUP7Xu9fPuAGK1JDBYA6T5UufYMUTGaxgspJA,2118
+datachain/lib/convert/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+datachain/lib/convert/flatten.py,sha256=XdAj0f9W32ABjOo8UyYm0y0H_yHDn3qEHERTyXuhJxk,1592
+datachain/lib/convert/type_converter.py,sha256=W-wvCIcb6OwWjRJ3EWJE4-LbpoqxsRBd6gYNpFlm8qo,2643
+datachain/lib/convert/unflatten.py,sha256=Ogvh_5wg2f38_At_1lN0D_e2uZOOpYEvwvB2xdq56Tw,2012
+datachain/lib/convert/values_to_tuples.py,sha256=MWz9pHT-AaPQN8hNMUYfuOHstyuNv0QEckwXlKgFbLA,3088
 datachain/query/__init__.py,sha256=tv-spkjUCYamMN9ys_90scYrZ8kJ7C7d1MTYVmxGtk4,325
-datachain/query/batch.py,sha256=sOMxXbaNii7lVyFIEZ2noqbhy_S8qtZ-WWxrka72shc,3474
+datachain/query/batch.py,sha256=j-_ZcuQra2Ro3Wj4crtqQCg-7xuv-p84hr4QHdvT7as,3479
 datachain/query/builtins.py,sha256=ZKNs49t8Oa_OaboCBIEqtXZt7c1Qe9OR_C_HpoDriIU,2781
-datachain/query/dataset.py,sha256=vpu2wQYC5uWc-LdZrNV-PV7xQapbYCtqyrXiiIa77DI,64982
-datachain/query/dispatch.py,sha256=ZeL5dga5d4cJDBftK7gAQ_mx4C7zq6t3z0Hdt7mcZYY,13094
+datachain/query/dataset.py,sha256=Pmaz16phEummJpWJD3x-8SMMbCb6xcOtWTyMdsFOdOE,64414
+datachain/query/dispatch.py,sha256=Qv5QpP5-K9JAmZLntifRzS5_AYHbK82Ahreo302Ntq8,13218
 datachain/query/metrics.py,sha256=vsECqbZfoSDBnvC3GQlziKXmISVYDLgHP1fMPEOtKyo,640
 datachain/query/params.py,sha256=O_j89mjYRLOwWNhYZl-z7mi-rkdP7WyFmaDufsdTryE,863
 datachain/query/schema.py,sha256=n1NBOj6JO2I26mZD4vSURmVC2rk3mjIkJQheeLogoy4,7748
 datachain/query/session.py,sha256=e4_vv4RqAjU-g3KK0avgLd9MEsmJBzRVEj1w8v7fP1k,3663
-datachain/query/udf.py,sha256=gnLDM7LKH8_bbdDeVHnlDKaBdbWc_NAbwvYCc4i-OlU,7101
+datachain/query/udf.py,sha256=c0IOTkcedpOQEmX-Idlrrl1__1IecNXL0N9oUO9Dtkg,7755
 datachain/remote/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-datachain/remote/studio.py,sha256=bZb85WjtqMNFBoRuPbH-TEGpAyz0afROR7E9UgIef_Y,7438
+datachain/remote/studio.py,sha256=f5s6qSZ9uB4URGUoU_8_W1KZRRQQVSm6cgEBkBUEfuE,7226
 datachain/sql/__init__.py,sha256=A2djrbQwSMUZZEIKGnm-mnRA-NDSbiDJNpAmmwGNyIo,303
 datachain/sql/selectable.py,sha256=fBM-wS1TUA42kVEAAiwqGtibIevyZAEritwt8PZGyLQ,1589
 datachain/sql/types.py,sha256=BzUm0nCcMPASvdqpQouX5bdVcK3G3DBfeeNhau7X_hA,10234
@@ -94,9 +96,10 @@ datachain/sql/sqlite/base.py,sha256=nPMF6_FF04hclDNZev_YfxMgbJAsWEdF-rU2pUhqBtc,
 datachain/sql/sqlite/types.py,sha256=oP93nLfTBaYnN0z_4Dsv-HZm8j9rrUf1esMM-z3JLbg,1754
 datachain/sql/sqlite/vector.py,sha256=ncW4eu2FlJhrP_CIpsvtkUabZlQdl2D5Lgwy_cbfqR0,469
 datachain/text/__init__.py,sha256=-yxHL2gVl3H0Zxam6iWUO6F1Mc4QAFHX6z-5fjHND74,72
-datachain-0.2.9.dist-info/LICENSE,sha256=8DnqK5yoPI_E50bEg_zsHKZHY2HqPy4rYN338BHQaRA,11344
-datachain-0.2.9.dist-info/METADATA,sha256=k1Q5NcgOeFu71IXx7Fygiuh8gt7GIU6jzN6X1-oHcPM,16475
-datachain-0.2.9.dist-info/WHEEL,sha256=Z4pYXqR_rTB7OWNDYFOm1qRk0RX6GFP2o8LgvP453Hk,91
-datachain-0.2.9.dist-info/entry_points.txt,sha256=0GMJS6B_KWq0m3VT98vQI2YZodAMkn4uReZ_okga9R4,49
-datachain-0.2.9.dist-info/top_level.txt,sha256=lZPpdU_2jJABLNIg2kvEOBi8PtsYikbN1OdMLHk8bTg,10
-datachain-0.2.9.dist-info/RECORD,,
+datachain/torch/__init__.py,sha256=9QJW8h0FevIXEykRsxQ7XzMDXvdIkv3kVf_UY95CTyg,600
+datachain-0.2.10.dist-info/LICENSE,sha256=8DnqK5yoPI_E50bEg_zsHKZHY2HqPy4rYN338BHQaRA,11344
+datachain-0.2.10.dist-info/METADATA,sha256=bWvqTD9c2joLmkDGpdcutjjF_s1LpccbSCLbkIaKQYQ,16732
+datachain-0.2.10.dist-info/WHEEL,sha256=Z4pYXqR_rTB7OWNDYFOm1qRk0RX6GFP2o8LgvP453Hk,91
+datachain-0.2.10.dist-info/entry_points.txt,sha256=0GMJS6B_KWq0m3VT98vQI2YZodAMkn4uReZ_okga9R4,49
+datachain-0.2.10.dist-info/top_level.txt,sha256=lZPpdU_2jJABLNIg2kvEOBi8PtsYikbN1OdMLHk8bTg,10
+datachain-0.2.10.dist-info/RECORD,,

datachain/image/__init__.py

@@ -1,3 +0,0 @@
-from datachain.lib.image import ImageFile, convert_images
-
-__all__ = ["ImageFile", "convert_images"]

datachain/lib/cached_stream.py

@@ -1,38 +0,0 @@
-from abc import ABC
-from contextlib import AbstractContextManager
-
-from datachain.cache import UniqueId
-
-
-class AbstractCachedStream(AbstractContextManager, ABC):
-    def __init__(self, catalog, uid: UniqueId):
-        self.catalog = catalog
-        self.uid = uid
-        self.mode = "rb"
-
-    def set_mode(self, mode):
-        self.mode = mode
-
-
-class PreCachedStream(AbstractCachedStream):
-    def __init__(self, catalog, uid: UniqueId):
-        super().__init__(catalog, uid)
-        self.client = self.catalog.get_client(self.uid.storage)
-        self.cached_file = None
-
-    def get_path_in_cache(self):
-        return self.catalog.cache.path_from_checksum(self.uid.get_hash())
-
-    def __enter__(self):
-        self.client.download(self.uid)
-        self.cached_file = open(self.get_path_in_cache(), self.mode)
-        return self.cached_file
-
-    def __exit__(self, *args):
-        self.cached_file.close()
-
-
-class PreDownloadStream(PreCachedStream):
-    def __exit__(self, *args):
-        super().__exit__(*args)
-        self.catalog.cache.remove(self.uid)

datachain/lib/claude.py

@@ -1,69 +0,0 @@
-import os
-from typing import Callable, Literal, Optional
-
-import anthropic
-
-from datachain.lib.feature import Feature
-from datachain.lib.file import File
-
-default_model_name = "claude-3-haiku-20240307"
-DEFAULT_OUTPUT_TOKENS = 1024
-
-# This classes can be auto-generated:
-# >> from anthropic.types.message import Message
-# >> ClaudeMessage = pydantic_to_feature(Message)
-# However, auto-generated pydentic classes do not work in multithreading mode.
-
-
-class UsageFr(Feature):
-    input_tokens: int = 0
-    output_tokens: int = 0
-
-
-class TextBlockFr(Feature):
-    text: str = ""
-    type: str = "text"
-
-
-class ClaudeMessage(Feature):
-    id: str = ""
-    content: list[TextBlockFr]
-    model: str = ""
-    role: str = ""
-    stop_reason: Optional[Literal["end_turn", "max_tokens", "stop_sequence"]] = None
-    stop_sequence: Optional[str] = None
-    type: Literal["message"] = "message"
-    usage: UsageFr = UsageFr()
-
-
-def claude_processor(
-    prompt: str,
-    messages: Optional[list] = None,
-    model: str = "claude-3-haiku-20240307",
-    api_key: Optional[str] = "",
-    max_retries: int = 5,
-    temperature: float = 0.9,
-    max_tokens: int = 1024,
-    **kwargs,
-) -> Callable:
-    if not messages:
-        messages = []
-    api_key = api_key or os.environ.get("ANTHROPIC_API_KEY")
-
-    def claude_func(file) -> ClaudeMessage:
-        try:
-            data = file.get_value() if isinstance(file, File) else file
-            client = anthropic.Anthropic(api_key=api_key, max_retries=max_retries)
-            response = client.messages.create(
-                model=model,
-                system=prompt,
-                messages=[{"role": "user", "content": data}, *messages],
-                temperature=temperature,
-                max_tokens=max_tokens,
-                **kwargs,
-            )
-            return ClaudeMessage(**response.model_dump())
-        except Exception:  # noqa: BLE001
-            return ClaudeMessage(content=[])
-
-    return claude_func

datachain/lib/feature.py

@@ -1,412 +0,0 @@
-import copy
-import inspect
-import re
-import warnings
-from collections.abc import Iterable, Sequence
-from datetime import datetime
-from enum import Enum
-from functools import lru_cache
-from types import GenericAlias
-from typing import (
-    TYPE_CHECKING,
-    Any,
-    ClassVar,
-    Literal,
-    Union,
-    get_args,
-    get_origin,
-)
-
-import attrs
-import numpy as np
-import pandas as pd
-from pydantic import BaseModel
-from typing_extensions import Literal as LiteralEx
-
-from datachain.lib.feature_registry import Registry
-from datachain.query import C
-from datachain.query.schema import DEFAULT_DELIMITER
-from datachain.sql.types import (
-    JSON,
-    Array,
-    Binary,
-    Boolean,
-    DateTime,
-    Float,
-    Int,
-    Int32,
-    Int64,
-    NullType,
-    SQLType,
-    String,
-)
-
-if TYPE_CHECKING:
-    from datachain.catalog import Catalog
-
-FeatureStandardType = Union[
-    type[int],
-    type[str],
-    type[float],
-    type[bool],
-    type[list],
-    type[dict],
-    type[bytes],
-    type[datetime],
-]
-
-FeatureType = Union[type["Feature"], FeatureStandardType]
-FeatureTypeNames = "Feature, int, str, float, bool, list, dict, bytes, datetime"
-
-
-TYPE_TO_DATACHAIN = {
-    int: Int64,
-    str: String,
-    Literal: String,
-    LiteralEx: String,
-    Enum: String,
-    float: Float,
-    bool: Boolean,
-    datetime: DateTime,  # Note, list of datetime is not supported yet
-    bytes: Binary,  # Note, list of bytes is not supported yet
-    list: Array,
-    dict: JSON,
-}
-
-DATACHAIN_TO_TYPE = {
-    Int: int,
-    Int32: int,
-    Int64: int,
-    String: str,
-    Float: float,
-    Boolean: bool,
-    DateTime: datetime,
-    Binary: bytes,
-    Array(NullType): list,
-    JSON: dict,
-}
-
-
-NUMPY_TO_DATACHAIN = {
-    np.dtype("int8"): Int,
-    np.dtype("int16"): Int,
-    np.dtype("int32"): Int,
-    np.dtype("int64"): Int,
-    np.dtype("uint8"): Int,
-    np.dtype("uint16"): Int,
-    np.dtype("uint32"): Int,
-    np.dtype("uint64"): Int,
-    np.dtype("float16"): Float,
-    np.dtype("float32"): Float,
-    np.dtype("float64"): Float,
-    np.dtype("object"): String,
-    pd.CategoricalDtype(): String,
-}
-
-
-# Disable Pydantic warning, see https://github.com/iterative/dvcx/issues/1285
-warnings.filterwarnings(
-    "ignore",
-    message="Field name .* shadows an attribute in parent",
-    category=UserWarning,
-)
-
-
-# Optimization: Store feature classes in this lookup variable so extra checks can be
-# skipped within loops.
-feature_classes_lookup: dict[type, bool] = {}
-
-
-class Feature(BaseModel):
-    """A base class for defining data classes that serve as inputs and outputs for
-    DataChain processing functions like `map()`, `gen()`, etc. Inherits from
-    `pydantic`'s BaseModel.
-    """
-
-    _is_file: ClassVar[bool] = False
-    _version: ClassVar[int] = 1
-
-    @classmethod
-    def _is_hidden(cls):
-        return cls.__name__.startswith("_")
-
-    def get_value(self, *args: Any, **kwargs: Any) -> Any:
-        name = self.__class__.__name__
-        raise NotImplementedError(f"get_value() is not defined for feature '{name}'")
-
-    @classmethod
-    def _name(cls) -> str:
-        return f"{cls.__name__}@{cls._version}"
-
-    @classmethod
-    def __pydantic_init_subclass__(cls):
-        Registry.add(cls)
-        for name, field_info in cls.model_fields.items():
-            attr_value = _resolve(cls, name, field_info, prefix=[])
-            setattr(cls, name, RestrictedAttribute(attr_value, cls, name))
-
-    @classmethod
-    def _prefix(cls) -> str:
-        return cls._normalize(cls.__name__)
-
-    @classmethod
-    def _normalize(cls, name: str) -> str:
-        if DEFAULT_DELIMITER in name:
-            raise RuntimeError(
-                f"variable '{name}' cannot be used "
-                f"because it contains {DEFAULT_DELIMITER}"
-            )
-        return Feature._to_snake_case(name)
-
-    @staticmethod
-    def _to_snake_case(name: str) -> str:
-        """Convert a CamelCase name to snake_case."""
-        s1 = re.sub("(.)([A-Z][a-z]+)", r"\1_\2", name)
-        return re.sub("([a-z0-9])([A-Z])", r"\1_\2", s1).lower()
-
-    def _set_stream(self, catalog: "Catalog", caching_enabled: bool = False) -> None:
-        pass
-
-    @classmethod
-    def get_file_signals(cls, path: list[str]) -> Iterable[list[str]]:
-        if cls._is_file:
-            yield path
-
-        for name, f_info in cls.model_fields.items():
-            anno = f_info.annotation
-            if Feature.is_feature(anno):
-                yield from anno.get_file_signals([*path, name])  # type: ignore[union-attr]
-
-    @classmethod
-    def is_feature(cls, anno) -> bool:
-        if anno in feature_classes_lookup:
-            # Optimization: Skip expensive subclass checks if already checked.
-            return feature_classes_lookup[anno]
-        is_class = inspect.isclass(anno)
-        result = (
-            is_class
-            and not isinstance(anno, GenericAlias)
-            and issubclass(anno, Feature)
-        )
-        if is_class:
-            # Only cache types in the feature classes lookup dict (not instances).
-            feature_classes_lookup[anno] = result
-        return result
-
-    @classmethod
-    def is_standard_type(cls, t: type) -> bool:
-        return any(
-            t is ft or t is get_args(ft)[0] for ft in get_args(FeatureStandardType)
-        )
-
-    @classmethod
-    def is_feature_type(cls, t: type) -> bool:
-        if cls.is_standard_type(t):
-            return True
-        if get_origin(t) is list and len(get_args(t)) == 1:
-            return cls.is_feature_type(get_args(t)[0])
-        return cls.is_feature(t)
-
-    def _flatten_fields_values(self, fields, model):
-        for name, f_info in fields.items():
-            anno = f_info.annotation
-            # Optimization: Access attributes directly to skip the model_dump() call.
-            value = getattr(model, name)
-
-            if isinstance(value, list):
-                yield [
-                    val.model_dump() if Feature.is_feature(type(val)) else val
-                    for val in value
-                ]
-            elif isinstance(value, dict):
-                yield {
-                    key: val.model_dump() if Feature.is_feature(type(val)) else val
-                    for key, val in value.items()
-                }
-            elif Feature.is_feature(anno):
-                yield from self._flatten_fields_values(anno.model_fields, value)
-            else:
-                yield value
-
-    def _flatten(self):
-        return tuple(self._flatten_fields_values(self.model_fields, self))
-
-    @staticmethod
-    def _flatten_list(objs):
-        return tuple(
-            val
-            for obj in objs
-            for val in obj._flatten_fields_values(obj.model_fields, obj)
-        )
-
-    @classmethod
-    def _unflatten_with_path(cls, dump, name_path: list[str]):
-        res = {}
-        for name, f_info in cls.model_fields.items():
-            anno = f_info.annotation
-            name_norm = cls._normalize(name)
-            lst = copy.copy(name_path)
-
-            if inspect.isclass(anno) and issubclass(anno, Feature):
-                lst.append(name_norm)
-                val = anno._unflatten_with_path(dump, lst)
-                res[name] = val
-            else:
-                lst.append(name_norm)
-                curr_path = DEFAULT_DELIMITER.join(lst)
-                res[name] = dump[curr_path]
-        return cls(**res)
-
-    @classmethod
-    def _unflatten(cls, dump):
-        return cls._unflatten_with_path(dump, [])
-
-    @classmethod
-    def _unflatten_to_json(cls, row: Sequence[Any], pos=0) -> dict:
-        return cls._unflatten_to_json_pos(row, pos)[0]
-
-    @classmethod
-    def _unflatten_to_json_pos(cls, row: Sequence[Any], pos=0) -> tuple[dict, int]:
-        res = {}
-        for name, f_info in cls.model_fields.items():
-            anno = f_info.annotation
-            origin = get_origin(anno)
-            if (
-                origin not in (list, dict)
-                and inspect.isclass(anno)
-                and issubclass(anno, Feature)
-            ):
-                res[name], pos = anno._unflatten_to_json_pos(row, pos)
-            else:
-                res[name] = row[pos]
-                pos += 1
-        return res, pos
-
-    @classmethod
-    @lru_cache(maxsize=1000)
-    def build_tree(cls):
-        res = {}
-
-        for name, f_info in cls.model_fields.items():
-            anno = f_info.annotation
-            subtree = anno.build_tree() if Feature.is_feature(anno) else None
-            res[name] = (anno, subtree)
-
-        return res
-
-
-class RestrictedAttribute:
-    """Descriptor implementing an attribute that can only be accessed through
-    the defining class and not from subclasses or instances.
-
-    Since it is a non-data descriptor, instance dicts have precedence over it.
-    Cannot be used with slotted classes.
-    """
-
-    def __init__(self, value, cls=None, name=None):
-        self.cls = cls
-        self.value = value
-        self.name = name
-
-    def __get__(self, instance, owner):
-        if owner is not self.cls:
-            raise AttributeError(
-                f"'{type(owner).__name__}' object has no attribute '{self.name}'"
-            )
-        if instance is not None:
-            raise RuntimeError(
-                f"Invalid attempt to access class attribute '{self.name}' through "
-                f"'{type(owner).__name__}' instance"
-            )
-        return self.value
-
-    def __set_name__(self, cls, name):
-        self.cls = cls
-        self.name = name
-
-
-@attrs.define
-class FeatureAttributeWrapper:
-    cls: type[Feature]
-    prefix: list[str]
-
-    @property
-    def name(self) -> str:
-        return DEFAULT_DELIMITER.join(self.prefix)
-
-    def __getattr__(self, name):
-        field_info = self.cls.model_fields.get(name)
-        if field_info:
-            return _resolve(self.cls, name, field_info, prefix=self.prefix)
-        raise AttributeError(
-            f"'{type(self).__name__}' object has no attribute '{name}'"
-        )
-
-
-def _resolve(cls, name, field_info, prefix: list[str]):
-    """Resolve feature attributes so they can be used in select(), join()
-    and similar functions.
-
-    Users just use `MyClass.sub_attr1.sub_attr2.field` and it will return a DB column
-    with a proper name (with default naming - `my_class__sub_attr1__sub_attr2__field`).
-    """
-    anno = field_info.annotation
-    norm_name = cls._normalize(name)
-
-    if not cls.is_feature(anno):
-        try:
-            anno_sql_class = convert_type_to_datachain(anno)
-        except TypeError:
-            anno_sql_class = NullType
-        new_prefix = copy.copy(prefix)
-        new_prefix.append(norm_name)
-        return C(DEFAULT_DELIMITER.join(new_prefix), anno_sql_class)
-
-    return FeatureAttributeWrapper(anno, [*prefix, norm_name])
-
-
-def convert_type_to_datachain(typ):  # noqa: PLR0911
-    if inspect.isclass(typ):
-        if issubclass(typ, SQLType):
-            return typ
-        if issubclass(typ, Enum):
-            return str
-
-    res = TYPE_TO_DATACHAIN.get(typ)
-    if res:
-        return res
-
-    orig = get_origin(typ)
-
-    if orig in (Literal, LiteralEx):
-        return String
-
-    args = get_args(typ)
-    if inspect.isclass(orig) and (issubclass(list, orig) or issubclass(tuple, orig)):
-        if args is None or len(args) != 1:
-            raise TypeError(f"Cannot resolve type '{typ}' for flattening features")
-
-        args0 = args[0]
-        if Feature.is_feature(args0):
-            return Array(JSON())
-
-        next_type = convert_type_to_datachain(args0)
-        return Array(next_type)
-
-    if inspect.isclass(orig) and issubclass(dict, orig):
-        return JSON
-
-    if orig == Union and len(args) == 2 and (type(None) in args):
-        return convert_type_to_datachain(args[0])
-
-    # Special case for list in JSON: Union[dict, list[dict]]
-    if orig == Union and len(args) >= 2:
-        args_no_nones = [arg for arg in args if arg != type(None)]
-        if len(args_no_nones) == 2:
-            args_no_dicts = [arg for arg in args_no_nones if arg is not dict]
-            if len(args_no_dicts) == 1 and get_origin(args_no_dicts[0]) is list:
-                arg = get_args(args_no_dicts[0])
-                if len(arg) == 1 and arg[0] is dict:
-                    return JSON
-
-    raise TypeError(f"Cannot recognize type {typ}")