PyPI - onnxslim - Versions diffs - 0.1.77__tar.gz → 0.1.79__tar.gz - Mend

onnxslim 0.1.77tar.gz → 0.1.79tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (75) hide show

{onnxslim-0.1.77/onnxslim.egg-info → onnxslim-0.1.79}/PKG-INFO RENAMED Viewed

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: onnxslim
-Version: 0.1.77
+Version: 0.1.79
 Summary: OnnxSlim: A Toolkit to Help Optimize Onnx Model
 Home-page: https://github.com/inisis/OnnxSlim
 Author: inisis
@@ -61,7 +61,9 @@ Dynamic: summary
 OnnxSlim can help you slim your onnx model, with less operators, but same accuracy, better inference speed.
-- 🚀 2025/05/17: OnnxSlim is merged into [optimum](https://github.com/huggingface/optimum) 🤗🤗🤗
+- 🚀 2025/11/29: Top 1% on PyPI
+- 🚀 2025/11/27: OnnxSlim is merged into [NVIDIA TensorRT-Model-Optimizer](https://github.com/NVIDIA/TensorRT-Model-Optimizer) 🤗🤗🤗
+- 🚀 2025/05/17: OnnxSlim is merged into [HuggingFace optimum](https://github.com/huggingface/optimum) 🤗🤗🤗
 - 🚀 2025/04/30: Rank 1st in the [AICAS 2025 LLM inference optimization challenge](https://tianchi.aliyun.com/competition/entrance/532289/customize588)
 - 🚀 2025/01/28: Achieved 1M downloads
 - 🚀 2024/06/23: OnnxSlim is merged into [transformers.js](https://github.com/huggingface/transformers.js) 🤗🤗🤗
@@ -119,19 +121,78 @@ For more usage, see onnxslim -h or refer to our [examples](./examples)
 # Projects using OnnxSlim
-- <img src="https://avatars.githubusercontent.com/u/131524?s=48&v=4" width="22" height="22"/>[Mozilla/smart_autofill](https://github.com/mozilla/smart_autofill)
-- <img src="https://avatars.githubusercontent.com/u/1961952?s=48&v=4" width="22" height="22"/>[alibaba/MNN](https://github.com/alibaba/MNN)
-- <img src="https://avatars.githubusercontent.com/u/23534030?s=48&v=4" width="22" height="22"/>[PaddlePaddle/PaddleOCR](https://github.com/PaddlePaddle/PaddleOCR)
-- <img src="https://avatars.githubusercontent.com/u/25720743?s=48&v=4" width="22" height="22"/>[huggingface/transformers.js](https://github.com/huggingface/transformers.js)
-- <img src="https://avatars.githubusercontent.com/u/25720743?s=48&v=4" width="22" height="22"/>[huggingface/optimum](https://github.com/huggingface/optimum)
-- <img src="https://avatars.githubusercontent.com/u/86091366?s=48&v=4" width="22" height="22"/>[THU-MIG/yolov10](https://github.com/THU-MIG/yolov10)
-- <img src="https://avatars.githubusercontent.com/u/26833451?s=48&v=4" width="22" height="22"/>[ultralytics/ultralytics](https://github.com/ultralytics/ultralytics)
-- <img src="https://avatars.githubusercontent.com/u/109945100?s=48&v=4" width="22" height="22"/>[ModelScope/FunASR](https://github.com/modelscope/FunASR)
-- <img src="https://avatars.githubusercontent.com/u/1961952?s=48&v=4" width="22" height="22"/>[alibaba/MNN-LLM](https://github.com/wangzhaode/mnn-llm)
-- <img src="https://avatars.githubusercontent.com/u/126587470?s=48&v=4" width="22" height="22"/>[deepghs/imgutils](https://github.com/deepghs/imgutils)
-- <img src="https://avatars.githubusercontent.com/u/48153283?s=48&v=4" width="22" height="22"/>[sunsmarterjie/yolov12](https://github.com/sunsmarterjie/yolov12)
-- <img src="https://avatars.githubusercontent.com/u/147458884?s=48&v=4" width="22" height="22"/>[nndeploy/nndeploy](https://github.com/nndeploy/nndeploy)
-- <img src="https://avatars.githubusercontent.com/u/111754012?s=48&v=4" width="22" height="22"/>[CVCUDA/CV-CUDA](https://github.com/CVCUDA/CV-CUDA)
+<table style="width:100%; border-collapse:separate; border-spacing:10px;">
+  <tr>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/1728152?s=200&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/NVIDIA/TensorRT-Model-Optimizer" target="_blank">NVIDIA/TensorRT-Model-Optimizer</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/1961952?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/alibaba/MNN" target="_blank">alibaba/MNN</a>
+    </td>
+  </tr>
+  <tr>
+  <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/26833451?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/ultralytics/ultralytics" target="_blank">ultralytics/ultralytics</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/131524?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/mozilla/smart_autofill" target="_blank">Mozilla/smart_autofill</a>
+    </td>
+  </tr>
+  <tr>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/1961952?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/wangzhaode/mnn-llm" target="_blank">alibaba/MNN-LLM</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/25720743?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/huggingface/transformers.js" target="_blank">huggingface/transformers.js</a>
+    </td>
+  </tr>
+  <tr>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/25720743?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/huggingface/optimum" target="_blank">huggingface/optimum</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/23534030?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/PaddlePaddle/PaddleOCR" target="_blank">PaddlePaddle/PaddleOCR</a>
+    </td>
+  </tr>
+  <tr>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/109945100?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/modelscope/FunASR" target="_blank">ModelScope/FunASR</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/111754012?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/CVCUDA/CV-CUDA" target="_blank">CVCUDA/CV-CUDA</a>
+    </td>
+  </tr>
+  <tr>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/86091366?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/THU-MIG/yolov10" target="_blank">THU-MIG/yolov10</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/48153283?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/sunsmarterjie/yolov12" target="_blank">sunsmarterjie/yolov12</a>
+    </td>
+  </tr>
+  <tr>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/147458884?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/nndeploy/nndeploy" target="_blank">nndeploy/nndeploy</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/126587470?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/deepghs/imgutils" target="_blank">deepghs/imgutils</a>
+    </td>
+  </tr>
+</table>
 # References

{onnxslim-0.1.77 → onnxslim-0.1.79}/README.md RENAMED Viewed

@@ -27,7 +27,9 @@
 OnnxSlim can help you slim your onnx model, with less operators, but same accuracy, better inference speed.
-- 🚀 2025/05/17: OnnxSlim is merged into [optimum](https://github.com/huggingface/optimum) 🤗🤗🤗
+- 🚀 2025/11/29: Top 1% on PyPI
+- 🚀 2025/11/27: OnnxSlim is merged into [NVIDIA TensorRT-Model-Optimizer](https://github.com/NVIDIA/TensorRT-Model-Optimizer) 🤗🤗🤗
+- 🚀 2025/05/17: OnnxSlim is merged into [HuggingFace optimum](https://github.com/huggingface/optimum) 🤗🤗🤗
 - 🚀 2025/04/30: Rank 1st in the [AICAS 2025 LLM inference optimization challenge](https://tianchi.aliyun.com/competition/entrance/532289/customize588)
 - 🚀 2025/01/28: Achieved 1M downloads
 - 🚀 2024/06/23: OnnxSlim is merged into [transformers.js](https://github.com/huggingface/transformers.js) 🤗🤗🤗
@@ -85,19 +87,78 @@ For more usage, see onnxslim -h or refer to our [examples](./examples)
 # Projects using OnnxSlim
-- <img src="https://avatars.githubusercontent.com/u/131524?s=48&v=4" width="22" height="22"/>[Mozilla/smart_autofill](https://github.com/mozilla/smart_autofill)
-- <img src="https://avatars.githubusercontent.com/u/1961952?s=48&v=4" width="22" height="22"/>[alibaba/MNN](https://github.com/alibaba/MNN)
-- <img src="https://avatars.githubusercontent.com/u/23534030?s=48&v=4" width="22" height="22"/>[PaddlePaddle/PaddleOCR](https://github.com/PaddlePaddle/PaddleOCR)
-- <img src="https://avatars.githubusercontent.com/u/25720743?s=48&v=4" width="22" height="22"/>[huggingface/transformers.js](https://github.com/huggingface/transformers.js)
-- <img src="https://avatars.githubusercontent.com/u/25720743?s=48&v=4" width="22" height="22"/>[huggingface/optimum](https://github.com/huggingface/optimum)
-- <img src="https://avatars.githubusercontent.com/u/86091366?s=48&v=4" width="22" height="22"/>[THU-MIG/yolov10](https://github.com/THU-MIG/yolov10)
-- <img src="https://avatars.githubusercontent.com/u/26833451?s=48&v=4" width="22" height="22"/>[ultralytics/ultralytics](https://github.com/ultralytics/ultralytics)
-- <img src="https://avatars.githubusercontent.com/u/109945100?s=48&v=4" width="22" height="22"/>[ModelScope/FunASR](https://github.com/modelscope/FunASR)
-- <img src="https://avatars.githubusercontent.com/u/1961952?s=48&v=4" width="22" height="22"/>[alibaba/MNN-LLM](https://github.com/wangzhaode/mnn-llm)
-- <img src="https://avatars.githubusercontent.com/u/126587470?s=48&v=4" width="22" height="22"/>[deepghs/imgutils](https://github.com/deepghs/imgutils)
-- <img src="https://avatars.githubusercontent.com/u/48153283?s=48&v=4" width="22" height="22"/>[sunsmarterjie/yolov12](https://github.com/sunsmarterjie/yolov12)
-- <img src="https://avatars.githubusercontent.com/u/147458884?s=48&v=4" width="22" height="22"/>[nndeploy/nndeploy](https://github.com/nndeploy/nndeploy)
-- <img src="https://avatars.githubusercontent.com/u/111754012?s=48&v=4" width="22" height="22"/>[CVCUDA/CV-CUDA](https://github.com/CVCUDA/CV-CUDA)
+<table style="width:100%; border-collapse:separate; border-spacing:10px;">
+  <tr>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/1728152?s=200&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/NVIDIA/TensorRT-Model-Optimizer" target="_blank">NVIDIA/TensorRT-Model-Optimizer</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/1961952?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/alibaba/MNN" target="_blank">alibaba/MNN</a>
+    </td>
+  </tr>
+  <tr>
+  <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/26833451?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/ultralytics/ultralytics" target="_blank">ultralytics/ultralytics</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/131524?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/mozilla/smart_autofill" target="_blank">Mozilla/smart_autofill</a>
+    </td>
+  </tr>
+  <tr>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/1961952?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/wangzhaode/mnn-llm" target="_blank">alibaba/MNN-LLM</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/25720743?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/huggingface/transformers.js" target="_blank">huggingface/transformers.js</a>
+    </td>
+  </tr>
+  <tr>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/25720743?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/huggingface/optimum" target="_blank">huggingface/optimum</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/23534030?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/PaddlePaddle/PaddleOCR" target="_blank">PaddlePaddle/PaddleOCR</a>
+    </td>
+  </tr>
+  <tr>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/109945100?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/modelscope/FunASR" target="_blank">ModelScope/FunASR</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/111754012?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/CVCUDA/CV-CUDA" target="_blank">CVCUDA/CV-CUDA</a>
+    </td>
+  </tr>
+  <tr>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/86091366?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/THU-MIG/yolov10" target="_blank">THU-MIG/yolov10</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/48153283?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/sunsmarterjie/yolov12" target="_blank">sunsmarterjie/yolov12</a>
+    </td>
+  </tr>
+  <tr>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/147458884?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/nndeploy/nndeploy" target="_blank">nndeploy/nndeploy</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/126587470?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/deepghs/imgutils" target="_blank">deepghs/imgutils</a>
+    </td>
+  </tr>
+</table>
 # References

onnxslim-0.1.79/VERSION ADDED Viewed

	@@ -0,0 +1 @@
1	+ 0.1.79

{onnxslim-0.1.77 → onnxslim-0.1.79}/onnxslim/third_party/onnx_graphsurgeon/exporters/onnx_exporter.py RENAMED Viewed

@@ -21,7 +21,6 @@ from collections.abc import Sequence
 import numpy as np
 import onnx
-import onnx.numpy_helper
 from onnx import IR_VERSION, ModelProto, defs
 from onnxslim.third_party.onnx_graphsurgeon.exporters.base_exporter import BaseExporter
@@ -35,9 +34,11 @@ from onnxslim.third_party.onnx_graphsurgeon.ir.tensor import (
     Tensor,
     Variable,
 )
 from onnxslim.third_party.onnx_graphsurgeon.logger import G_LOGGER
 from onnxslim.third_party.onnx_graphsurgeon.util import misc
+from ml_dtypes import bfloat16, float8_e4m3fn
 def dtype_to_onnx(dtype: np.dtype | onnx.TensorProto.DataType) -> int:
     """Converts a numpy dtype or ONNX data type to its integer representation."""
@@ -86,6 +87,15 @@ def update_import_domains(graph):
     return graph.import_domains
+def float32_to_bfloat16_uint16(x):
+    """Convert a float32 value to bfloat16 represented as uint16."""
+    return bfloat16(x).view(np.uint16)
+def float32_to_float8e4m3(x):
+    """Convert a float32 value to float8e4m3 represented as uint8."""
+    return float8_e4m3fn(x).view(np.uint8)
 class NumpyArrayConverter:
     def __init__(self, container, scalar_converter):
         self.func = np.vectorize(scalar_converter, otypes=[container])
@@ -95,12 +105,10 @@ class NumpyArrayConverter:
 _NUMPY_ARRAY_CONVERTERS = {
-    onnx.TensorProto.BFLOAT16: NumpyArrayConverter(np.uint16, onnx.helper.float32_to_bfloat16),
+    onnx.TensorProto.BFLOAT16: NumpyArrayConverter(np.uint16, float32_to_bfloat16_uint16),
     # FP8 in TensorRT supports negative zeros, no infinities
     # See https://onnx.ai/onnx/technical/float8.html#papers
-    onnx.TensorProto.FLOAT8E4M3FN: NumpyArrayConverter(
-        np.uint8, lambda x: onnx.helper.float32_to_float8e4m3(x, fn=True, uz=False)
-    ),
+    onnx.TensorProto.FLOAT8E4M3FN: NumpyArrayConverter(np.uint8, float32_to_float8e4m3),
 }

{onnxslim-0.1.77 → onnxslim-0.1.79}/onnxslim/third_party/onnx_graphsurgeon/ir/graph.py RENAMED Viewed

@@ -978,17 +978,46 @@ class Graph:
             return next(iter(tensor.values)) if tensor.shape else tensor.values
         def fold_shape(tensor):
-            """Returns the input tensor shape if available, otherwise returns None."""
-            inp = get_input(get_producer(tensor, "Shape"))
+            """Returns the input tensor shape if available, otherwise returns None.
+            Handles Shape node with optional 'start' and 'end' attributes (opset 15+).
+            """
+            shape_node = get_producer(tensor, "Shape")
+            inp = get_input(shape_node)
             if inp is None:
                 return None
             if inp.shape is None or misc.is_dynamic_shape(inp.shape):
                 return None
-            return np.array(inp.shape, dtype=np.int64)
+            full_shape = inp.shape
+            num_dims = len(full_shape)
+            # Get start and end attributes (default: start=0, end=None means full shape)
+            start = shape_node.attrs.get("start", 0)
+            end = shape_node.attrs.get("end", None)
+            # Handle negative indices
+            if start < 0:
+                start = num_dims + start
+            if end is None:
+                end = num_dims
+            elif end < 0:
+                end = num_dims + end
+            # Clamp to valid range
+            start = max(0, min(start, num_dims))
+            end = max(0, min(end, num_dims))
+            if start > end:
+                return None
+            target_shape = full_shape[start:end]
+            return np.array(target_shape, dtype=np.int64)
         def fold_shape_gather(tensor):
-            """Retrieves and returns the shape of the input tensor as a NumPy array, otherwise returns None."""
+            """Retrieves and returns the shape of the input tensor as a NumPy array, otherwise returns None.
+            Handles Shape node with optional 'start' and 'end' attributes (opset 15+).
+            """
             gather = get_producer(tensor, "Gather")
             if gather is None:
                 return None
@@ -996,69 +1025,132 @@ class Graph:
             data = gather.inputs[0]
             indices_tensor = gather.inputs[1]
-            inp = get_input(get_producer(data, "Shape"))
+            shape_node = get_producer(data, "Shape")
+            inp = get_input(shape_node)
             if inp is None or inp.shape is None:
                 return None
             if not isinstance(indices_tensor, Constant):
                 return None
+            # Get the shape slice from Shape node (considering start/end attributes)
+            full_shape = inp.shape
+            num_dims = len(full_shape)
+            start = shape_node.attrs.get("start", 0)
+            end = shape_node.attrs.get("end", None)
+            if start < 0:
+                start = num_dims + start
+            if end is None:
+                end = num_dims
+            elif end < 0:
+                end = num_dims + end
+            start = max(0, min(start, num_dims))
+            end = max(0, min(end, num_dims))
+            if start > end:
+                return None
+            shape_slice = full_shape[start:end]
             indices = indices_tensor.values
             if not indices.shape:  # Scalar-case
-                shape = inp.shape[int(indices)]
+                idx = int(indices)
+                # Handle negative indices relative to shape_slice
+                if idx < 0:
+                    idx = len(shape_slice) + idx
+                if idx < 0 or idx >= len(shape_slice):
+                    return None
+                shape = shape_slice[idx]
                 if misc.is_dynamic_dimension(shape):
                     return None
             else:
-                shape = [inp.shape[index] for index in indices]
+                shape = []
+                for index in indices:
+                    idx = int(index)
+                    # Handle negative indices relative to shape_slice
+                    if idx < 0:
+                        idx = len(shape_slice) + idx
+                    if idx < 0 or idx >= len(shape_slice):
+                        return None
+                    shape.append(shape_slice[idx])
                 if misc.is_dynamic_shape(shape):
                     return None
             return np.array(shape, dtype=np.int64)
         def fold_shape_slice(tensor):
-            """Fold tensor shape slice information into a NumPy array of int64 type."""
-            slice = get_producer(tensor, "Slice")
-            if slice is None:
+            """Fold tensor shape slice information into a NumPy array of int64 type.
+            Handles Shape node with optional 'start' and 'end' attributes (opset 15+).
+            """
+            slice_node = get_producer(tensor, "Slice")
+            if slice_node is None:
                 return None
-            data = slice.inputs[0]
+            data = slice_node.inputs[0]
-            if len(slice.inputs) >= 3:
-                starts, ends = slice.inputs[1:3]
+            if len(slice_node.inputs) >= 3:
+                starts, ends = slice_node.inputs[1:3]
                 if any(not isinstance(t, Constant) for t in [starts, ends]):
                     return None
                 starts, ends = get_scalar_value(starts), get_scalar_value(ends)
-            elif "starts" in slice.attrs and "ends" in slice.attrs:
-                starts, ends = slice.attrs["starts"][0], slice.attrs["ends"][0]
+            elif "starts" in slice_node.attrs and "ends" in slice_node.attrs:
+                starts, ends = slice_node.attrs["starts"][0], slice_node.attrs["ends"][0]
             else:
                 return None
-            inp = get_input(get_producer(data, "Shape"))
+            shape_node = get_producer(data, "Shape")
+            inp = get_input(shape_node)
             if inp is None or inp.shape is None:
                 return None
             # For shape tensors, we can only slice on the 0th dimension.
-            if len(slice.inputs) > 3:
-                axes = slice.inputs[3]
+            if len(slice_node.inputs) > 3:
+                axes = slice_node.inputs[3]
                 if not isinstance(axes, Constant):
                     return None
                 if get_scalar_value(axes) != 0:
                     return None
-            elif "axes" in slice.attrs:
-                if slice.attrs["axes"][0] != 0:
+            elif "axes" in slice_node.attrs:
+                if slice_node.attrs["axes"][0] != 0:
                     return None
             steps = 1
-            if len(slice.inputs) > 4:
-                steps = slice.inputs[4]
+            if len(slice_node.inputs) > 4:
+                steps = slice_node.inputs[4]
                 if not isinstance(steps, Constant):
                     return None
                 steps = get_scalar_value(steps)
-            elif "steps" in slice.attrs:
-                steps = slice.attrs["steps"][0]
+            elif "steps" in slice_node.attrs:
+                steps = slice_node.attrs["steps"][0]
+            # Get the shape slice from Shape node (considering start/end attributes)
+            full_shape = inp.shape
+            num_dims = len(full_shape)
+            shape_start = shape_node.attrs.get("start", 0)
+            shape_end = shape_node.attrs.get("end", None)
+            if shape_start < 0:
+                shape_start = num_dims + shape_start
+            if shape_end is None:
+                shape_end = num_dims
+            elif shape_end < 0:
+                shape_end = num_dims + shape_end
+            shape_start = max(0, min(shape_start, num_dims))
+            shape_end = max(0, min(shape_end, num_dims))
+            if shape_start > shape_end:
+                return None
+            shape_slice = full_shape[shape_start:shape_end]
-            shape = inp.shape[starts:ends:steps]
+            # Apply the Slice operation on the shape_slice
+            shape = shape_slice[starts:ends:steps]
             if misc.is_dynamic_shape(shape):
                 return None

onnxslim-0.1.79/onnxslim/version.py ADDED Viewed

	@@ -0,0 +1 @@
1	+ __version__ = "0.1.79"

{onnxslim-0.1.77 → onnxslim-0.1.79/onnxslim.egg-info}/PKG-INFO RENAMED Viewed

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: onnxslim
-Version: 0.1.77
+Version: 0.1.79
 Summary: OnnxSlim: A Toolkit to Help Optimize Onnx Model
 Home-page: https://github.com/inisis/OnnxSlim
 Author: inisis
@@ -61,7 +61,9 @@ Dynamic: summary
 OnnxSlim can help you slim your onnx model, with less operators, but same accuracy, better inference speed.
-- 🚀 2025/05/17: OnnxSlim is merged into [optimum](https://github.com/huggingface/optimum) 🤗🤗🤗
+- 🚀 2025/11/29: Top 1% on PyPI
+- 🚀 2025/11/27: OnnxSlim is merged into [NVIDIA TensorRT-Model-Optimizer](https://github.com/NVIDIA/TensorRT-Model-Optimizer) 🤗🤗🤗
+- 🚀 2025/05/17: OnnxSlim is merged into [HuggingFace optimum](https://github.com/huggingface/optimum) 🤗🤗🤗
 - 🚀 2025/04/30: Rank 1st in the [AICAS 2025 LLM inference optimization challenge](https://tianchi.aliyun.com/competition/entrance/532289/customize588)
 - 🚀 2025/01/28: Achieved 1M downloads
 - 🚀 2024/06/23: OnnxSlim is merged into [transformers.js](https://github.com/huggingface/transformers.js) 🤗🤗🤗
@@ -119,19 +121,78 @@ For more usage, see onnxslim -h or refer to our [examples](./examples)
 # Projects using OnnxSlim
-- <img src="https://avatars.githubusercontent.com/u/131524?s=48&v=4" width="22" height="22"/>[Mozilla/smart_autofill](https://github.com/mozilla/smart_autofill)
-- <img src="https://avatars.githubusercontent.com/u/1961952?s=48&v=4" width="22" height="22"/>[alibaba/MNN](https://github.com/alibaba/MNN)
-- <img src="https://avatars.githubusercontent.com/u/23534030?s=48&v=4" width="22" height="22"/>[PaddlePaddle/PaddleOCR](https://github.com/PaddlePaddle/PaddleOCR)
-- <img src="https://avatars.githubusercontent.com/u/25720743?s=48&v=4" width="22" height="22"/>[huggingface/transformers.js](https://github.com/huggingface/transformers.js)
-- <img src="https://avatars.githubusercontent.com/u/25720743?s=48&v=4" width="22" height="22"/>[huggingface/optimum](https://github.com/huggingface/optimum)
-- <img src="https://avatars.githubusercontent.com/u/86091366?s=48&v=4" width="22" height="22"/>[THU-MIG/yolov10](https://github.com/THU-MIG/yolov10)
-- <img src="https://avatars.githubusercontent.com/u/26833451?s=48&v=4" width="22" height="22"/>[ultralytics/ultralytics](https://github.com/ultralytics/ultralytics)
-- <img src="https://avatars.githubusercontent.com/u/109945100?s=48&v=4" width="22" height="22"/>[ModelScope/FunASR](https://github.com/modelscope/FunASR)
-- <img src="https://avatars.githubusercontent.com/u/1961952?s=48&v=4" width="22" height="22"/>[alibaba/MNN-LLM](https://github.com/wangzhaode/mnn-llm)
-- <img src="https://avatars.githubusercontent.com/u/126587470?s=48&v=4" width="22" height="22"/>[deepghs/imgutils](https://github.com/deepghs/imgutils)
-- <img src="https://avatars.githubusercontent.com/u/48153283?s=48&v=4" width="22" height="22"/>[sunsmarterjie/yolov12](https://github.com/sunsmarterjie/yolov12)
-- <img src="https://avatars.githubusercontent.com/u/147458884?s=48&v=4" width="22" height="22"/>[nndeploy/nndeploy](https://github.com/nndeploy/nndeploy)
-- <img src="https://avatars.githubusercontent.com/u/111754012?s=48&v=4" width="22" height="22"/>[CVCUDA/CV-CUDA](https://github.com/CVCUDA/CV-CUDA)
+<table style="width:100%; border-collapse:separate; border-spacing:10px;">
+  <tr>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/1728152?s=200&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/NVIDIA/TensorRT-Model-Optimizer" target="_blank">NVIDIA/TensorRT-Model-Optimizer</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/1961952?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/alibaba/MNN" target="_blank">alibaba/MNN</a>
+    </td>
+  </tr>
+  <tr>
+  <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/26833451?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/ultralytics/ultralytics" target="_blank">ultralytics/ultralytics</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/131524?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/mozilla/smart_autofill" target="_blank">Mozilla/smart_autofill</a>
+    </td>
+  </tr>
+  <tr>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/1961952?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/wangzhaode/mnn-llm" target="_blank">alibaba/MNN-LLM</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/25720743?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/huggingface/transformers.js" target="_blank">huggingface/transformers.js</a>
+    </td>
+  </tr>
+  <tr>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/25720743?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/huggingface/optimum" target="_blank">huggingface/optimum</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/23534030?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/PaddlePaddle/PaddleOCR" target="_blank">PaddlePaddle/PaddleOCR</a>
+    </td>
+  </tr>
+  <tr>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/109945100?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/modelscope/FunASR" target="_blank">ModelScope/FunASR</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/111754012?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/CVCUDA/CV-CUDA" target="_blank">CVCUDA/CV-CUDA</a>
+    </td>
+  </tr>
+  <tr>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/86091366?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/THU-MIG/yolov10" target="_blank">THU-MIG/yolov10</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/48153283?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/sunsmarterjie/yolov12" target="_blank">sunsmarterjie/yolov12</a>
+    </td>
+  </tr>
+  <tr>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/147458884?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/nndeploy/nndeploy" target="_blank">nndeploy/nndeploy</a>
+    </td>
+    <td style="vertical-align:middle;">
+      <img src="https://avatars.githubusercontent.com/u/126587470?s=48&v=4" width="22" height="22" style="vertical-align:middle; margin-right:8px;"/>
+      <a href="https://github.com/deepghs/imgutils" target="_blank">deepghs/imgutils</a>
+    </td>
+  </tr>
+</table>
 # References