corvic-engine 0.3.0rc61__tar.gz → 0.3.0rc63__tar.gz

{corvic_engine-0.3.0rc61 → corvic_engine-0.3.0rc63}/Cargo.lock

@@ -8,12 +8,6 @@ version = "1.4.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "ace50bade8e6234aa140d9a2f552bbee1db4d353f69b8217bc503490fc1a9f26"
 
-[[package]]
-name = "bitflags"
-version = "2.9.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5c8214115b7bf84099f1309324e63141d4c5d7cc26862f97a0a857dbefe165bd"
-
 [[package]]
 name = "cfg-if"
 version = "1.0.0"
@@ -22,7 +16,7 @@ checksum = "baf1de4339761588bc0619e3cbc0120ee582ebb74b53b4efbf79117bd2da40fd"
 
 [[package]]
 name = "corvic-engine"
-version = "0.3.0-rc.61"
+version = "0.3.0-rc.63"
 dependencies = [
  "numpy",
  "pyo3",
@@ -63,16 +57,16 @@ checksum = "48c757948c5ede0e46177b7add2e67155f70e33c07fea8284df6576da70b3719"
 
 [[package]]
 name = "engine-walk"
-version = "0.3.0-rc.61"
+version = "0.3.0-rc.63"
 dependencies = [
  "version_check",
 ]
 
 [[package]]
 name = "heck"
-version = "0.4.1"
+version = "0.5.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "95505c38b4572b2d910cecb0281560f54b440a19336cbbcb27bf6ce6adc6f5a8"
+checksum = "2304e00983f87ffb38b55b444b5e3b60a884b5d30c0fca7d82fe33449bbe55ea"
 
 [[package]]
 name = "indoc"
@@ -86,16 +80,6 @@ version = "0.2.171"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "c19937216e9d3aa9956d9bb8dfc0b0c8beb6058fc4f7a4dc4d850edf86a237d6"
 
-[[package]]
-name = "lock_api"
-version = "0.4.12"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "07af8b9cdd281b7915f413fa73f29ebd5d55d0d3f0155584dade1ff18cea1b17"
-dependencies = [
- "autocfg",
- "scopeguard",
-]
-
 [[package]]
 name = "matrixmultiply"
 version = "0.3.9"
@@ -117,14 +101,16 @@ dependencies = [
 
 [[package]]
 name = "ndarray"
-version = "0.15.6"
+version = "0.16.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "adb12d4e967ec485a5f71c6311fe28158e9d6f4bc4a447b474184d0f91a8fa32"
+checksum = "882ed72dce9365842bf196bdeedf5055305f11fc8c03dee7bb0194a6cad34841"
 dependencies = [
  "matrixmultiply",
  "num-complex",
  "num-integer",
  "num-traits",
+ "portable-atomic",
+ "portable-atomic-util",
  "rawpointer",
 ]
 
@@ -157,9 +143,9 @@ dependencies = [
 
 [[package]]
 name = "numpy"
-version = "0.21.0"
+version = "0.24.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ec170733ca37175f5d75a5bea5911d6ff45d2cd52849ce98b685394e4f2f37f4"
+checksum = "a7cfbf3f0feededcaa4d289fe3079b03659e85c5b5a177f4ba6fb01ab4fb3e39"
 dependencies = [
  "libc",
  "ndarray",
@@ -167,6 +153,7 @@ dependencies = [
  "num-integer",
  "num-traits",
  "pyo3",
+ "pyo3-build-config",
  "rustc-hash",
 ]
 
@@ -177,34 +164,20 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "42f5e15c9953c5e4ccceeb2e7382a716482c34515315f7b03532b8b4e8393d2d"
 
 [[package]]
-name = "parking_lot"
-version = "0.12.3"
+name = "portable-atomic"
+version = "1.11.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "f1bf18183cf54e8d6059647fc3063646a1801cf30896933ec2311622cc4b9a27"
-dependencies = [
- "lock_api",
- "parking_lot_core",
-]
+checksum = "350e9b48cbc6b0e028b0473b114454c6316e57336ee184ceab6e53f72c178b3e"
 
 [[package]]
-name = "parking_lot_core"
-version = "0.9.10"
+name = "portable-atomic-util"
+version = "0.2.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "1e401f977ab385c9e4e3ab30627d6f26d00e2c73eef317493c4ec6d468726cf8"
+checksum = "d8a2f0d8d040d7848a709caf78912debcc3f33ee4b3cac47d73d1e1069e83507"
 dependencies = [
- "cfg-if",
- "libc",
- "redox_syscall",
- "smallvec",
- "windows-targets",
+ "portable-atomic",
 ]
 
-[[package]]
-name = "portable-atomic"
-version = "1.11.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "350e9b48cbc6b0e028b0473b114454c6316e57336ee184ceab6e53f72c178b3e"
-
 [[package]]
 name = "proc-macro2"
 version = "1.0.94"
@@ -216,15 +189,15 @@ dependencies = [
 
 [[package]]
 name = "pyo3"
-version = "0.21.2"
+version = "0.24.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a5e00b96a521718e08e03b1a622f01c8a8deb50719335de3f60b3b3950f069d8"
+checksum = "17da310086b068fbdcefbba30aeb3721d5bb9af8db4987d6735b2183ca567229"
 dependencies = [
  "cfg-if",
  "indoc",
  "libc",
  "memoffset",
- "parking_lot",
+ "once_cell",
  "portable-atomic",
  "pyo3-build-config",
  "pyo3-ffi",
@@ -234,9 +207,9 @@ dependencies = [
 
 [[package]]
 name = "pyo3-build-config"
-version = "0.21.2"
+version = "0.24.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7883df5835fafdad87c0d888b266c8ec0f4c9ca48a5bed6bbb592e8dedee1b50"
+checksum = "e27165889bd793000a098bb966adc4300c312497ea25cf7a690a9f0ac5aa5fc1"
 dependencies = [
  "once_cell",
  "target-lexicon",
@@ -244,9 +217,9 @@ dependencies = [
 
 [[package]]
 name = "pyo3-ffi"
-version = "0.21.2"
+version = "0.24.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "01be5843dc60b916ab4dad1dca6d20b9b4e6ddc8e15f50c47fe6d85f1fb97403"
+checksum = "05280526e1dbf6b420062f3ef228b78c0c54ba94e157f5cb724a609d0f2faabc"
 dependencies = [
  "libc",
  "pyo3-build-config",
@@ -254,9 +227,9 @@ dependencies = [
 
 [[package]]
 name = "pyo3-macros"
-version = "0.21.2"
+version = "0.24.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "77b34069fc0682e11b31dbd10321cbf94808394c56fd996796ce45217dfac53c"
+checksum = "5c3ce5686aa4d3f63359a5100c62a127c9f15e8398e5fdeb5deef1fed5cd5f44"
 dependencies = [
  "proc-macro2",
  "pyo3-macros-backend",
@@ -266,9 +239,9 @@ dependencies = [
 
 [[package]]
 name = "pyo3-macros-backend"
-version = "0.21.2"
+version = "0.24.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "08260721f32db5e1a5beae69a55553f56b99bd0e1c3e6e0a5e8851a9d0f5a85c"
+checksum = "f4cf6faa0cbfb0ed08e89beb8103ae9724eb4750e3a78084ba4017cbe94f3855"
 dependencies = [
  "heck",
  "proc-macro2",
@@ -312,32 +285,11 @@ dependencies = [
  "crossbeam-utils",
 ]
 
-[[package]]
-name = "redox_syscall"
-version = "0.5.10"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0b8c0c260b63a8219631167be35e6a988e9554dbd323f8bd08439c8ed1302bd1"
-dependencies = [
- "bitflags",
-]
-
 [[package]]
 name = "rustc-hash"
-version = "1.1.0"
+version = "2.1.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "08d43f7aa6b08d49f382cde6a7982047c3426db949b1424bc4b7ec9ae12c6ce2"
-
-[[package]]
-name = "scopeguard"
-version = "1.2.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "94143f37725109f92c262ed2cf5e59bce7498c01bcc1502d7b9afe439a4e9f49"
-
-[[package]]
-name = "smallvec"
-version = "1.14.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7fcf8323ef1faaee30a44a340193b1ac6814fd9b7b4e88e9d4519a3e4abe1cfd"
+checksum = "357703d41365b4b27c590e3ed91eabb1b663f07c4c084095e60cbed4362dff0d"
 
 [[package]]
 name = "syn"
@@ -352,9 +304,9 @@ dependencies = [
 
 [[package]]
 name = "target-lexicon"
-version = "0.12.16"
+version = "0.13.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "61c41af27dd6d1e27b1b16b489db798443478cef1f06a660c96db617ba5de3b1"
+checksum = "e502f78cdbb8ba4718f566c418c52bc729126ffd16baee5baa718cf25dd5a69a"
 
 [[package]]
 name = "thread_local"
@@ -383,67 +335,3 @@ name = "version_check"
 version = "0.9.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "0b928f33d975fc6ad9f86c8f283853ad26bdd5b10b7f1542aa2fa15e2289105a"
-
-[[package]]
-name = "windows-targets"
-version = "0.52.6"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9b724f72796e036ab90c1021d4780d4d3d648aca59e491e6b98e725b84e99973"
-dependencies = [
- "windows_aarch64_gnullvm",
- "windows_aarch64_msvc",
- "windows_i686_gnu",
- "windows_i686_gnullvm",
- "windows_i686_msvc",
- "windows_x86_64_gnu",
- "windows_x86_64_gnullvm",
- "windows_x86_64_msvc",
-]
-
-[[package]]
-name = "windows_aarch64_gnullvm"
-version = "0.52.6"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "32a4622180e7a0ec044bb555404c800bc9fd9ec262ec147edd5989ccd0c02cd3"
-
-[[package]]
-name = "windows_aarch64_msvc"
-version = "0.52.6"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "09ec2a7bb152e2252b53fa7803150007879548bc709c039df7627cabbd05d469"
-
-[[package]]
-name = "windows_i686_gnu"
-version = "0.52.6"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8e9b5ad5ab802e97eb8e295ac6720e509ee4c243f69d781394014ebfe8bbfa0b"
-
-[[package]]
-name = "windows_i686_gnullvm"
-version = "0.52.6"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0eee52d38c090b3caa76c563b86c3a4bd71ef1a819287c19d586d7334ae8ed66"
-
-[[package]]
-name = "windows_i686_msvc"
-version = "0.52.6"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "240948bc05c5e7c6dabba28bf89d89ffce3e303022809e73deaefe4f6ec56c66"
-
-[[package]]
-name = "windows_x86_64_gnu"
-version = "0.52.6"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "147a5c80aabfbf0c7d901cb5895d1de30ef2907eb21fbbab29ca94c5b08b1a78"
-
-[[package]]
-name = "windows_x86_64_gnullvm"
-version = "0.52.6"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "24d5b23dc417412679681396f2b49f3de8c1473deb516bd34410872eff51ed0d"
-
-[[package]]
-name = "windows_x86_64_msvc"
-version = "0.52.6"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "589f6da84c646204747d1270a2a5661ea66ed1cced2631d546fdfb155959f9ec"

{corvic_engine-0.3.0rc61 → corvic_engine-0.3.0rc63}/Cargo.toml

@@ -3,7 +3,7 @@ resolver = "2"
 members = ["python"]
 
 [workspace.package]
-version = "0.3.0-rc.61"
+version = "0.3.0-rc.63"
 edition = "2021"
 publish = false
 
@@ -12,9 +12,9 @@ arrow-array = { version = ">=41", default-features = false }
 arrow-buffer = { version = ">=41", default-features = false }
 arrow-data = { version = ">=41", default-features = false }
 arrow-schema = { version = ">=41", default-features = false }
-pyo3 = "0.21"
+pyo3 = "0.24.1"
 tikv-jemallocator = { version = "0.5.0" }
-numpy = "0.21"
+numpy = "0.24"
 rayon = "1.8"
 version_check = "0.9.4"
 thread_local = "1.1.7"

{corvic_engine-0.3.0rc61 → corvic_engine-0.3.0rc63}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: corvic-engine
-Version: 0.3.0rc61
+Version: 0.3.0rc63
 Classifier: Environment :: Console
 Classifier: License :: Other/Proprietary License
 Classifier: Programming Language :: Python :: Implementation :: CPython

{corvic_engine-0.3.0rc61 → corvic_engine-0.3.0rc63}/pyproject.toml

@@ -20,7 +20,7 @@ authors = [
 ]
 readme = "README.md"
 description = "Seamless embedding generation and retrieval."
-version = "0.3.0-rc.61" # sync this with tool.poetry.version below
+version = "0.3.0-rc.63" # sync this with tool.poetry.version below
 
 # TODO(ddn): Pull dependencies from tool.poetry.dependencies. We use maturin as
 # the build system but poetry for the development environment.
@@ -74,7 +74,7 @@ packages = [
 ]
 description = ""
 authors = []
-version = "0.3.0-rc.61"
+version = "0.3.0-rc.63"
 
 [tool.poetry.scripts]
 check = "corvic_check.cli:main"

{corvic_engine-0.3.0rc61 → corvic_engine-0.3.0rc63}/python/corvic/embed/node2vec.py

@@ -39,6 +39,8 @@ class KeyedVectors:
           index_to_key: mapping of index to key struct
           key_field_order: order of key struct fields used for index operations
         """
+        if dim <= 0:
+            raise InvalidArgumentError("number of dimensions must be positive")
         self.dim = dim
         self._index_to_key = index_to_key
         self._key_field_order = key_field_order

{corvic_engine-0.3.0rc61 → corvic_engine-0.3.0rc63}/python/corvic/op_graph/_transformations.py

@@ -73,7 +73,7 @@ def _replace_concat_op_source(
     return op.concat(new_tables, concat_op.how)
 
 
-def replace_op_source(  # noqa: C901
+def replace_op_source(  # noqa: C901, PLR0915
     root_op: op.Op, source_to_replace: op.Op, new_source: op.Op
 ) -> Ok[op.Op] | InvalidArgumentError:
     for source in root_op.sources():
@@ -98,6 +98,8 @@ def replace_op_source(  # noqa: C901
             return new_source.select_columns(root_op.columns)
         case op.LimitRows():
             return new_source.limit_rows(root_op.num_rows)
+        case op.OffsetRows():
+            return new_source.offset_rows(root_op.num_rows)
         case op.OrderBy():
             return new_source.order_by(root_op.columns, desc=root_op.desc)
         case op.FilterRows():

{corvic_engine-0.3.0rc61 → corvic_engine-0.3.0rc63}/python/corvic/op_graph/ops.py

@@ -71,6 +71,7 @@ ProtoOp = (
     | table_pb2.JoinOp
     | table_pb2.SelectColumnsOp
     | table_pb2.LimitRowsOp
+    | table_pb2.OffsetRowsOp
     | table_pb2.OrderByOp
     | table_pb2.FilterRowsOp
     | table_pb2.DistinctRowsOp
@@ -162,6 +163,15 @@ def from_proto(
 ) -> LimitRows: ...
 
 
+@overload
+def from_proto(
+    proto: table_pb2.OffsetRowsOp,
+    parent_ops: list[Op] | None = None,
+    *,
+    skip_validate: bool = False,
+) -> OffsetRows: ...
+
+
 @overload
 def from_proto(
     proto: table_pb2.OrderByOp,
@@ -490,6 +500,8 @@ def from_proto(  # noqa: C901, PLR0915
             proto = table_pb2.TableComputeOp(select_columns=proto)
         case table_pb2.LimitRowsOp():
             proto = table_pb2.TableComputeOp(limit_rows=proto)
+        case table_pb2.OffsetRowsOp():
+            proto = table_pb2.TableComputeOp(offset_rows=proto)
         case table_pb2.OrderByOp():
             proto = table_pb2.TableComputeOp(order_by=proto)
         case table_pb2.FilterRowsOp():
@@ -992,6 +1004,12 @@ class _Base(OneofProtoWrapper[table_pb2.TableComputeOp], ABC):
             proto = table_pb2.LimitRowsOp(source=self._proto, num_rows=num_rows)
         return Ok(from_proto(proto, skip_validate=True))
 
+    def offset_rows(self, num_rows: int) -> InvalidArgumentError | Ok[OffsetRows]:
+        if num_rows <= 0:
+            return InvalidArgumentError("num_rows must be positive")
+        proto = table_pb2.OffsetRowsOp(source=self._proto, num_rows=num_rows)
+        return Ok(from_proto(proto, skip_validate=True))
+
     def order_by(
         self, columns: Sequence[str], *, desc: bool
     ) -> InvalidArgumentError | Ok[OrderBy]:
@@ -1903,6 +1921,23 @@ class LimitRows(_Base):
         return [self.source]
 
 
+class OffsetRows(_Base):
+    """Limit the number of rows in a table."""
+
+    @property
+    def num_rows(self) -> int:
+        return self._proto.offset_rows.num_rows
+
+    @property
+    def source(self) -> Op:
+        if self._parents:
+            return self._parents[0]
+        return from_proto(self._proto.offset_rows.source, skip_validate=True)
+
+    def sources(self):
+        return [self.source]
+
+
 class OrderBy(_Base):
     """Order the rows in a table."""
 
@@ -2756,6 +2791,7 @@ Op = (
     | Join
     | SelectColumns
     | LimitRows
+    | OffsetRows
     | OrderBy
     | FilterRows
     | DistinctRows
@@ -2798,6 +2834,7 @@ _COMPUTE_OP_FIELD_NAME_TO_OP: Final = {
     "join": Join,
     "select_columns": SelectColumns,
     "limit_rows": LimitRows,
+    "offset_rows": OffsetRows,
     "order_by": OrderBy,
     "filter_rows": FilterRows,
     "distinct_rows": DistinctRows,
@@ -3323,6 +3360,7 @@ class Schema(Sequence[Field]):
 
             case (
                 LimitRows()
+                | OffsetRows()
                 | OrderBy()
                 | FilterRows()
                 | DistinctRows()

{corvic_engine-0.3.0rc61 → corvic_engine-0.3.0rc63}/python/corvic/sql/parse_ops.py

@@ -24,6 +24,7 @@ _SqlComputableOp = (
     | op_graph.op.Join
     | op_graph.op.SelectColumns
     | op_graph.op.LimitRows
+    | op_graph.op.OffsetRows
     | op_graph.op.OrderBy
     | op_graph.op.FilterRows
     | op_graph.op.DistinctRows
@@ -74,6 +75,7 @@ def can_be_sql_computed(
             | op_graph.op.Join()
             | op_graph.op.SelectColumns()
             | op_graph.op.LimitRows()
+            | op_graph.op.OffsetRows()
             | op_graph.op.OrderBy()
             | op_graph.op.FilterRows()
             | op_graph.op.DistinctRows()
@@ -362,6 +364,12 @@ class _OpLogParser:
     ) -> Ok[sqlglot.exp.Query] | InvalidArgumentError | NoRowsError:
         return self.parse(op.source).map(lambda query: query.limit(op.num_rows))
 
+    def _offset_rows_to_sql(
+        self,
+        op: op_graph.op.OffsetRows,
+    ) -> Ok[sqlglot.exp.Query] | InvalidArgumentError | NoRowsError:
+        return self.parse(op.source).map(lambda query: query.offset(op.num_rows))
+
     def _order_by_to_sql(
         self,
         op: op_graph.op.OrderBy,
@@ -715,6 +723,8 @@ class _OpLogParser:
                 return self._select_columns_to_sql(op)
             case op_graph.op.LimitRows():
                 return self._limit_rows_to_sql(op)
+            case op_graph.op.OffsetRows():
+                return self._offset_rows_to_sql(op)
             case op_graph.op.OrderBy():
                 return self._order_by_to_sql(op)
             case op_graph.op.FilterRows():

{corvic_engine-0.3.0rc61 → corvic_engine-0.3.0rc63}/python/corvic/system/_embedder.py

@@ -1,6 +1,6 @@
 import dataclasses
 from collections.abc import Sequence
-from typing import Any, Literal
+from typing import TYPE_CHECKING, Any, Literal
 
 import numpy as np
 import polars as pl
@@ -9,6 +9,12 @@ from typing_extensions import Protocol
 from corvic import orm
 from corvic.result import InternalError, InvalidArgumentError, Ok
 
+if TYPE_CHECKING:
+    from transformers import (
+        CLIPModel,
+        CLIPProcessor,
+    )
+
 
 @dataclasses.dataclass
 class EmbedTextContext:
@@ -64,6 +70,12 @@ class ImageEmbedder(Protocol):
     ) -> Ok[EmbedImageResult] | InvalidArgumentError | InternalError: ...
 
 
+@dataclasses.dataclass
+class ClipModels:
+    model: "CLIPModel"
+    processor: "CLIPProcessor"
+
+
 class ClipText(TextEmbedder):
     """Clip Text embedder.
 
@@ -76,28 +88,39 @@ class ClipText(TextEmbedder):
     overcoming several major challenges in computer vision.
     """
 
-    def embed(
-        self, context: EmbedTextContext
-    ) -> Ok[EmbedTextResult] | InvalidArgumentError | InternalError:
-        import torch
+    def _load_models(self):
         from transformers import (
             CLIPModel,
             CLIPProcessor,
         )
 
         model: CLIPModel = CLIPModel.from_pretrained(  # pyright: ignore[reportUnknownMemberType]
-            "openai/clip-vit-base-patch32"
+            pretrained_model_name_or_path="openai/clip-vit-base-patch32",
+            revision="5812e510083bb2d23fa43778a39ac065d205ed4d",
         )
         processor: CLIPProcessor = CLIPProcessor.from_pretrained(  # pyright: ignore[reportUnknownMemberType, reportAssignmentType]
-            "openai/clip-vit-base-patch32"
+            pretrained_model_name_or_path="openai/clip-vit-base-patch32",
+            revision="5812e510083bb2d23fa43778a39ac065d205ed4d",
+            use_fast=False,
         )
-        model.eval()
+        return ClipModels(model=model, processor=processor)
+
+    def embed(
+        self, context: EmbedTextContext
+    ) -> Ok[EmbedTextResult] | InvalidArgumentError | InternalError:
         match context.expected_coordinate_bitwidth:
             case 64:
                 coord_dtype = pl.Float64()
             case 32:
                 coord_dtype = pl.Float32()
 
+        models = self._load_models()
+        model = models.model
+        processor = models.processor
+        model.eval()
+
+        import torch
+
         with torch.no_grad():
             inputs: dict[str, torch.Tensor] = processor(  # pyright: ignore[reportAssignmentType]
                 text=context.inputs,

{corvic_engine-0.3.0rc61 → corvic_engine-0.3.0rc63}/python/corvic/system/_image_embedder.py

@@ -1,3 +1,4 @@
+import dataclasses
 from io import BytesIO
 from typing import TYPE_CHECKING, Any
 
@@ -13,6 +14,10 @@ from corvic.system._embedder import (
 
 if TYPE_CHECKING:
     from PIL import Image
+    from transformers import (
+        CLIPModel,
+        CLIPProcessor,
+    )
 
 
 class RandomImageEmbedder(ImageEmbedder):
@@ -58,6 +63,12 @@ def image_from_bytes(
         return InvalidArgumentError("invalid image format")
 
 
+@dataclasses.dataclass
+class ClipModels:
+    model: "CLIPModel"
+    processor: "CLIPProcessor"
+
+
 class Clip(ImageEmbedder):
     """Clip image embedder.
 
@@ -70,6 +81,23 @@ class Clip(ImageEmbedder):
     overcoming several major challenges in computer vision.
     """
 
+    def _load_models(self):
+        from transformers import (
+            CLIPModel,
+            CLIPProcessor,
+        )
+
+        model: CLIPModel = CLIPModel.from_pretrained(  # pyright: ignore[reportUnknownMemberType]
+            pretrained_model_name_or_path="openai/clip-vit-base-patch32",
+            revision="5812e510083bb2d23fa43778a39ac065d205ed4d",
+        )
+        processor: CLIPProcessor = CLIPProcessor.from_pretrained(  # pyright: ignore[reportUnknownMemberType, reportAssignmentType]
+            pretrained_model_name_or_path="openai/clip-vit-base-patch32",
+            revision="5812e510083bb2d23fa43778a39ac065d205ed4d",
+            use_fast=False,
+        )
+        return ClipModels(model=model, processor=processor)
+
     def embed(
         self, context: EmbedImageContext
     ) -> Ok[EmbedImageResult] | InvalidArgumentError | InternalError:
@@ -99,20 +127,13 @@ class Clip(ImageEmbedder):
                 )
             )
 
-        import torch
-        from transformers import (
-            CLIPModel,
-            CLIPProcessor,
-        )
-
-        model: CLIPModel = CLIPModel.from_pretrained(  # pyright: ignore[reportUnknownMemberType]
-            "openai/clip-vit-base-patch32"
-        )
-        processor: CLIPProcessor = CLIPProcessor.from_pretrained(  # pyright: ignore[reportUnknownMemberType, reportAssignmentType]
-            "openai/clip-vit-base-patch32"
-        )
+        models = self._load_models()
+        model = models.model
+        processor = models.processor
         model.eval()
 
+        import torch
+
         with torch.no_grad():
             inputs: dict[str, torch.FloatTensor] = processor(  # pyright: ignore[reportAssignmentType]
                 images=images, return_tensors="pt"