onnx2tf 1.29.20__tar.gz → 1.29.21__tar.gz

{onnx2tf-1.29.20 → onnx2tf-1.29.21}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: onnx2tf
-Version: 1.29.20
+Version: 1.29.21
 Summary: Self-Created Tools to convert ONNX files (NCHW) to TensorFlow/TFLite/Keras format (NHWC). The purpose of this tool is to solve the massive Transpose extrapolation problem in onnx-tensorflow (onnx-tf).
 Keywords: onnx,tensorflow,tflite,keras,deep-learning,machine-learning
 Author: Katsuya Hyodo
@@ -18,7 +18,7 @@ Classifier: Programming Language :: Python :: 3.11
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Requires-Dist: requests==2.32.5
 Requires-Dist: numpy==1.26.4
-Requires-Dist: onnx==1.19.0
+Requires-Dist: onnx==1.19.1
 Requires-Dist: onnxruntime==1.23.0
 Requires-Dist: opencv-python==4.11.0.86
 Requires-Dist: onnxsim==0.4.36
@@ -365,7 +365,7 @@ Video speed is adjusted approximately 50 times slower than actual speed.
   docker run --rm -it \
   -v `pwd`:/workdir \
   -w /workdir \
-  ghcr.io/pinto0309/onnx2tf:1.29.20
+  ghcr.io/pinto0309/onnx2tf:1.29.21
 
   or
 
@@ -373,18 +373,18 @@ Video speed is adjusted approximately 50 times slower than actual speed.
   docker run --rm -it \
   -v `pwd`:/workdir \
   -w /workdir \
-  docker.io/pinto0309/onnx2tf:1.29.20
+  docker.io/pinto0309/onnx2tf:1.29.21
 
   or
 
-  pip install -U onnx==1.19.0 \
+  pip install -U onnx==1.19.1 \
   && pip install -U onnx-graphsurgeon==0.5.8 \
   && pip install -U onnxruntime==1.23.0 \
   && pip install -U onnxsim==0.4.36 \
   && pip install -U onnxoptimizer==0.4.2 \
   && pip install -U simple_onnx_processing_tools==1.1.32 \
-  && pip install -U sne4onnx>=1.0.13 \
-  && pip install -U sng4onnx>=1.0.4 \
+  && pip install -U sne4onnx==1.0.15 \
+  && pip install -U sng4onnx==1.0.5 \
   && pip install -U ai_edge_litert==1.2.0 \
   && pip install -U tensorflow==2.19.0 \
   && pip install -U protobuf==3.20.3 \
@@ -630,7 +630,7 @@ After many upgrades, the need for JSON parameter correction has become much less
 
 `-ois` an option to overwrite the input OP to a static size if it has undefined dimensions. `-cotof` option checks the accuracy of all OPs one by one. `-cotoa` is the error value of the threshold for determining an accuracy error. If there are undefined dimensions in the input OP, it is better to fix them to the static geometry to improve the accuracy of the accuracy measurement.
 
-Also, you can use the `-cind` option to specify custom input for `-cotof`, instead of using the default dummy input. Otherwise, all input values will be set to 1. For more information about the `-cind` option, please refer to [here](#cli-parameter).
+Also, you can use the `-cind` option to specify custom input for `-cotof`, instead of using the default dummy input. Otherwise, all input values will be set to 1. You can override the dummy input values with `--value_hints` (scalar only, `*:default` supported). For more information about the `-cind` option, please refer to [here](#cli-parameter).
 
 The `-cotof` option only compares the original ONNX and converted TensorFlow (Keras) models at Float32 precision, not at Float16 or INT8 precision.
 
@@ -644,6 +644,10 @@ onnx2tf -i mobilenetv2-12.onnx -b 1 -cotof -cotoa 1e-1
 or
 
 onnx2tf -i mobilenetv2-12.onnx -cotof -cotoa 1e-1 -cind "input" "/your/path/x.npy"
+
+or
+
+onnx2tf -i mobilenetv2-12.onnx -cotof -cotoa 1e-1 --value_hints "input:0.5" "*:1.0"
 ```
 ![image](https://user-images.githubusercontent.com/33194443/216901668-5fdb1e38-8670-46a4-b4b9-8a774fa7545e.png)
 
@@ -1826,6 +1830,14 @@ optional arguments:
     A value of 1 or more must be specified.
     Numerical values other than dynamic dimensions are ignored.
 
+  -vh VALUE_HINTS [VALUE_HINTS ...], \
+      --value_hints VALUE_HINTS [VALUE_HINTS ...]
+    Value hints for dummy inference input tensors.
+    The format is
+    "input_name_1:value" "input_name_2:value" "*:default_value"
+    "*" applies to all inputs not explicitly specified.
+    Values are scalar only.
+
   -nlt, --no_large_tensor
     Suppresses constant bloat caused by Tile OP when optimizing models in onnxsim.
     See: https://github.com/daquexian/onnx-simplifier/issues/178
@@ -2157,6 +2169,7 @@ convert(
   batch_size: Union[int, NoneType] = None,
   overwrite_input_shape: Union[List[str], NoneType] = None,
   shape_hints: Union[List[str], NoneType] = None,
+  value_hints: Union[List[str], NoneType] = None,
   no_large_tensor: Optional[bool] = False,
   output_nms_with_dynamic_tensor: Optional[bool] = False,
   switch_nms_version: Optional[str] = 'v4',
@@ -2377,6 +2390,13 @@ convert(
       A value of 1 or more must be specified.
       Numerical values other than dynamic dimensions are ignored.
 
+    value_hints: Optional[List[str]]
+      Value hints for dummy inference input tensors.
+      The format is
+      ['input_name_1:value', 'input_name_2:value', '*:default_value']
+      "*" applies to all inputs not explicitly specified.
+      Values are scalar only.
+
     no_large_tensor: Optional[bool]
       Suppresses constant bloat caused by Tile OP when optimizing models in onnxsim.
       See: https://github.com/daquexian/onnx-simplifier/issues/178
@@ -3032,6 +3052,7 @@ The above differences often cannot be dealt with by simply converting the model
 14. [nobuco](https://github.com/AlexanderLutsenko/nobuco)
 15. [onnx2torch](https://github.com/ENOT-AutoDL/onnx2torch)
 16. [ai-edge-torch](https://github.com/google-ai-edge/ai-edge-torch)
+17. [LiteRT.js](https://ai.google.dev/edge/litert/web)
 
 ## Acknowledgement
 1. https://github.com/onnx/models

{onnx2tf-1.29.20 → onnx2tf-1.29.21}/README.md

@@ -323,7 +323,7 @@ Video speed is adjusted approximately 50 times slower than actual speed.
   docker run --rm -it \
   -v `pwd`:/workdir \
   -w /workdir \
-  ghcr.io/pinto0309/onnx2tf:1.29.20
+  ghcr.io/pinto0309/onnx2tf:1.29.21
 
   or
 
@@ -331,18 +331,18 @@ Video speed is adjusted approximately 50 times slower than actual speed.
   docker run --rm -it \
   -v `pwd`:/workdir \
   -w /workdir \
-  docker.io/pinto0309/onnx2tf:1.29.20
+  docker.io/pinto0309/onnx2tf:1.29.21
 
   or
 
-  pip install -U onnx==1.19.0 \
+  pip install -U onnx==1.19.1 \
   && pip install -U onnx-graphsurgeon==0.5.8 \
   && pip install -U onnxruntime==1.23.0 \
   && pip install -U onnxsim==0.4.36 \
   && pip install -U onnxoptimizer==0.4.2 \
   && pip install -U simple_onnx_processing_tools==1.1.32 \
-  && pip install -U sne4onnx>=1.0.13 \
-  && pip install -U sng4onnx>=1.0.4 \
+  && pip install -U sne4onnx==1.0.15 \
+  && pip install -U sng4onnx==1.0.5 \
   && pip install -U ai_edge_litert==1.2.0 \
   && pip install -U tensorflow==2.19.0 \
   && pip install -U protobuf==3.20.3 \
@@ -588,7 +588,7 @@ After many upgrades, the need for JSON parameter correction has become much less
 
 `-ois` an option to overwrite the input OP to a static size if it has undefined dimensions. `-cotof` option checks the accuracy of all OPs one by one. `-cotoa` is the error value of the threshold for determining an accuracy error. If there are undefined dimensions in the input OP, it is better to fix them to the static geometry to improve the accuracy of the accuracy measurement.
 
-Also, you can use the `-cind` option to specify custom input for `-cotof`, instead of using the default dummy input. Otherwise, all input values will be set to 1. For more information about the `-cind` option, please refer to [here](#cli-parameter).
+Also, you can use the `-cind` option to specify custom input for `-cotof`, instead of using the default dummy input. Otherwise, all input values will be set to 1. You can override the dummy input values with `--value_hints` (scalar only, `*:default` supported). For more information about the `-cind` option, please refer to [here](#cli-parameter).
 
 The `-cotof` option only compares the original ONNX and converted TensorFlow (Keras) models at Float32 precision, not at Float16 or INT8 precision.
 
@@ -602,6 +602,10 @@ onnx2tf -i mobilenetv2-12.onnx -b 1 -cotof -cotoa 1e-1
 or
 
 onnx2tf -i mobilenetv2-12.onnx -cotof -cotoa 1e-1 -cind "input" "/your/path/x.npy"
+
+or
+
+onnx2tf -i mobilenetv2-12.onnx -cotof -cotoa 1e-1 --value_hints "input:0.5" "*:1.0"
 ```
 ![image](https://user-images.githubusercontent.com/33194443/216901668-5fdb1e38-8670-46a4-b4b9-8a774fa7545e.png)
 
@@ -1784,6 +1788,14 @@ optional arguments:
     A value of 1 or more must be specified.
     Numerical values other than dynamic dimensions are ignored.
 
+  -vh VALUE_HINTS [VALUE_HINTS ...], \
+      --value_hints VALUE_HINTS [VALUE_HINTS ...]
+    Value hints for dummy inference input tensors.
+    The format is
+    "input_name_1:value" "input_name_2:value" "*:default_value"
+    "*" applies to all inputs not explicitly specified.
+    Values are scalar only.
+
   -nlt, --no_large_tensor
     Suppresses constant bloat caused by Tile OP when optimizing models in onnxsim.
     See: https://github.com/daquexian/onnx-simplifier/issues/178
@@ -2115,6 +2127,7 @@ convert(
   batch_size: Union[int, NoneType] = None,
   overwrite_input_shape: Union[List[str], NoneType] = None,
   shape_hints: Union[List[str], NoneType] = None,
+  value_hints: Union[List[str], NoneType] = None,
   no_large_tensor: Optional[bool] = False,
   output_nms_with_dynamic_tensor: Optional[bool] = False,
   switch_nms_version: Optional[str] = 'v4',
@@ -2335,6 +2348,13 @@ convert(
       A value of 1 or more must be specified.
       Numerical values other than dynamic dimensions are ignored.
 
+    value_hints: Optional[List[str]]
+      Value hints for dummy inference input tensors.
+      The format is
+      ['input_name_1:value', 'input_name_2:value', '*:default_value']
+      "*" applies to all inputs not explicitly specified.
+      Values are scalar only.
+
     no_large_tensor: Optional[bool]
       Suppresses constant bloat caused by Tile OP when optimizing models in onnxsim.
       See: https://github.com/daquexian/onnx-simplifier/issues/178
@@ -2990,6 +3010,7 @@ The above differences often cannot be dealt with by simply converting the model
 14. [nobuco](https://github.com/AlexanderLutsenko/nobuco)
 15. [onnx2torch](https://github.com/ENOT-AutoDL/onnx2torch)
 16. [ai-edge-torch](https://github.com/google-ai-edge/ai-edge-torch)
+17. [LiteRT.js](https://ai.google.dev/edge/litert/web)
 
 ## Acknowledgement
 1. https://github.com/onnx/models

{onnx2tf-1.29.20 → onnx2tf-1.29.21}/onnx2tf/__init__.py

@@ -1,3 +1,3 @@
 from onnx2tf.onnx2tf import convert, main
 
-__version__ = '1.29.20'
+__version__ = '1.29.21'

{onnx2tf-1.29.20 → onnx2tf-1.29.21}/onnx2tf/onnx2tf.py

@@ -43,6 +43,7 @@ from typing import Optional, List, Any, Dict
 from argparse import ArgumentParser
 
 import importlib
+import onnx2tf.utils.common_functions as common_functions
 from onnx2tf.utils.common_functions import (
     dummy_onnx_inference,
     dummy_tf_inference,
@@ -639,6 +640,7 @@ def convert(
     batch_size: Optional[int] = None,
     overwrite_input_shape: Optional[List[str]] = None,
     shape_hints: Optional[List[str]] = None,
+    value_hints: Optional[List[str]] = None,
     no_large_tensor: Optional[bool] = False,
     output_nms_with_dynamic_tensor: Optional[bool] = False,
     switch_nms_version: Optional[str] = 'v4',
@@ -850,6 +852,15 @@ def convert(
         A value of 1 or more must be specified.\n
         Numerical values other than dynamic dimensions are ignored.
 
+    value_hints: Optional[List[str]]
+        Value hints for dummy inference input tensors.\n
+        The format is\n
+        ["input_name_1:value","input_name_2:value","*:default_value"].\n
+        "*" applies to all inputs not explicitly specified.\n
+        Values are scalar only.\n
+        e.g.\n
+        ['input0:0.5','mask:0','*:1.0']\n
+
     no_large_tensor: Optional[bool]
         Suppresses constant bloat caused by Tile OP when optimizing models in onnxsim.\n
         See: https://github.com/daquexian/onnx-simplifier/issues/178
@@ -1110,6 +1121,8 @@ def convert(
     if verbosity is None:
         verbosity = 'debug'
     set_log_level('error' if non_verbose else verbosity)
+    common_functions.set_dummy_shape_hints(shape_hints)
+    common_functions.set_dummy_value_hints(value_hints)
 
     # Either designation required
     if not input_onnx_file_path and not onnx_graph:
@@ -1326,6 +1339,10 @@ def convert(
                 'Failed to optimize the onnx file.'
             )
 
+    has_external_data = False
+    if input_onnx_file_path and os.path.exists(input_onnx_file_path):
+        has_external_data = check_has_external_data(input_onnx_file_path)
+
     # Automatic generation of each OP name - sng4onnx
     if not not_use_opname_auto_generate:
         info('')
@@ -1357,9 +1374,7 @@ def convert(
 
     # Loading Graphs
     # onnx_graph If specified, onnx_graph is processed first
-    has_external_data = False
     if not onnx_graph:
-        has_external_data = check_has_external_data(input_onnx_file_path)
         onnx_graph = onnx.load(input_onnx_file_path)
 
     if not auto_split_model and onnx_graph is not None:
@@ -1686,6 +1701,7 @@ def convert(
                     'batch_size': batch_size,
                     'overwrite_input_shape': overwrite_input_shape,
                     'shape_hints': shape_hints,
+                    'value_hints': value_hints,
                     'no_large_tensor': no_large_tensor,
                     'output_nms_with_dynamic_tensor': output_nms_with_dynamic_tensor,
                     'switch_nms_version': switch_nms_version,
@@ -3874,6 +3890,18 @@ def main():
             'Only applied to dynamic dimensions in inputs. \n' +
             'Only used when -cotof or -coto are specified.'
     )
+    parser.add_argument(
+        '-vh',
+        '--value_hints',
+        type=str,
+        nargs='+',
+        help=\
+            'Value hints for dummy inference input tensors. \n' +
+            'The format is\n' +
+            '"input_name_1:value" "input_name_2:value" "*:default_value". \n' +
+            '"*" applies to all inputs not explicitly specified. \n' +
+            'Values are scalar only.'
+    )
     parser.add_argument(
         '-nlt',
         '--no_large_tensor',
@@ -4359,6 +4387,7 @@ def main():
         batch_size=args.batch_size,
         overwrite_input_shape=args.overwrite_input_shape,
         shape_hints=args.shape_hints,
+        value_hints=args.value_hints,
         no_large_tensor=args.no_large_tensor,
         output_nms_with_dynamic_tensor=args.output_nms_with_dynamic_tensor,
         switch_nms_version=args.switch_nms_version,

{onnx2tf-1.29.20 → onnx2tf-1.29.21}/onnx2tf/utils/common_functions.py

@@ -1,4 +1,5 @@
 import math
+import ast
 import os
 import io
 import re
@@ -45,6 +46,59 @@ from onnx2tf.utils.enums import (
 INF_INDEX_VALUE: int = 4294967296
 ONNX_INF_INDEX_VALUE = sys.maxsize # 9223372036854775807
 
+_DEFAULT_DUMMY_SHAPE_HINTS: Optional[List[str]] = None
+_DEFAULT_DUMMY_VALUE_HINTS: Optional[List[str]] = None
+
+
+def set_dummy_shape_hints(shape_hints: Optional[List[str]]) -> None:
+    global _DEFAULT_DUMMY_SHAPE_HINTS
+    _DEFAULT_DUMMY_SHAPE_HINTS = shape_hints
+
+
+def set_dummy_value_hints(value_hints: Optional[List[str]]) -> None:
+    global _DEFAULT_DUMMY_VALUE_HINTS
+    _DEFAULT_DUMMY_VALUE_HINTS = value_hints
+
+
+def _parse_value_hint_scalar(value: str) -> Optional[Any]:
+    try:
+        parsed = ast.literal_eval(value)
+    except Exception:
+        try:
+            parsed = float(value)
+        except Exception:
+            return None
+    if isinstance(parsed, (list, tuple, dict, set, np.ndarray)):
+        return None
+    if isinstance(parsed, (int, float, bool, np.number)):
+        return parsed
+    return None
+
+
+def _parse_value_hints(
+    value_hints: Optional[List[str]]
+) -> Tuple[Dict[str, Any], Optional[Any], bool]:
+    if not value_hints:
+        return {}, None, False
+    hints: Dict[str, Any] = {}
+    default_value: Optional[Any] = None
+    for hint in value_hints:
+        if not isinstance(hint, str):
+            continue
+        parts = hint.split(':', 1)
+        if len(parts) != 2:
+            continue
+        input_name, value_str = parts[0], parts[1]
+        parsed_value = _parse_value_hint_scalar(value_str)
+        if parsed_value is None:
+            warn(f'Invalid --value_hints entry ignored: {hint}')
+            continue
+        if input_name == '*':
+            default_value = parsed_value
+        else:
+            hints[input_name] = parsed_value
+    return hints, default_value, default_value is not None
+
 
 
 
@@ -3854,6 +3908,7 @@ def dummy_onnx_inference(
     enable_ort_output_memmap: bool = False,
     ort_output_memmap_dir: Optional[str] = None,
     shape_hints: Optional[List[str]] = None,
+    value_hints: Optional[List[str]] = None,
     input_datas_for_validation: Optional[Dict[str, np.ndarray]] = None,
 ) -> List[np.ndarray]:
     """Perform inference on ONNX subgraphs with an all-1 dummy tensor.
@@ -3891,6 +3946,10 @@ def dummy_onnx_inference(
         Directory to store memmap files. If not specified, a temporary
         directory is created and removed on exit.
 
+    value_hints: Optional[List[str]]
+        Value hints for dummy inference input tensors.
+        Format: ["input_name:value", "*:default_value"].
+
     input_datas_for_validation: Optional[Dict[str, np.ndarray]]
         Optional dict to be filled with the input tensors used for inference.
 
@@ -3899,6 +3958,11 @@ def dummy_onnx_inference(
     outputs: List[np.ndarray]
         Results of inference using dummy tensor
     """
+    if shape_hints is None:
+        shape_hints = _DEFAULT_DUMMY_SHAPE_HINTS
+    if value_hints is None:
+        value_hints = _DEFAULT_DUMMY_VALUE_HINTS
+
     # Separate onnx at specified output_names position
     domain: str = onnx_graph.domain
     ir_version: int = onnx_graph.ir_version
@@ -4053,6 +4117,7 @@ def dummy_onnx_inference(
         name: tuple(size) for name, size in zip(input_names, input_sizes)
     }
     input_datas = {}
+    value_hints_dict, default_value, has_default = _parse_value_hints(value_hints)
 
     # -cid
     if custom_input_op_name_np_data_path:
@@ -4086,7 +4151,17 @@ def dummy_onnx_inference(
 
     else:
         for input_name, input_size, input_dtype in zip(input_names, input_sizes, input_dtypes):
-            if test_data_nhwc is None:
+            hint_value = value_hints_dict.get(
+                input_name,
+                default_value if has_default else None,
+            )
+            if hint_value is not None:
+                input_datas[input_name] = np.full(
+                    input_size,
+                    hint_value,
+                    dtype=input_dtype,
+                )
+            elif test_data_nhwc is None:
                 input_datas[input_name] = np.ones(
                     input_size,
                     dtype=input_dtype,
@@ -4245,6 +4320,7 @@ def dummy_tf_inference(
     verification_datas: Optional[List[np.ndarray]] = None,
     custom_input_op_name_np_data_path: Optional[str] = None,
     shape_hints: Optional[List[str]] = None,
+    value_hints: Optional[List[str]] = None,
     input_datas_for_validation: Optional[Dict[str, np.ndarray]] = None,
     prefilled_input_datas: Optional[Dict[str, np.ndarray]] = None,
     keep_shape_absolutely_input_names: Optional[List[str]] = None,
@@ -4269,6 +4345,11 @@ def dummy_tf_inference(
 
     custom_input_op_name_np_data_path
         Path to Numpy file for custom data used for dummy inference
+
+    value_hints: Optional[List[str]]
+        Value hints for dummy inference input tensors.
+        Format: ["input_name:value", "*:default_value"].
+
     input_datas_for_validation: Optional[Dict[str, np.ndarray]]
         Optional dict to be filled with the input tensors used for inference.
 
@@ -4278,6 +4359,11 @@ def dummy_tf_inference(
         Results of inference using dummy tensor.
         Dict of tensorflow node and corresponding ndarray output.
     """
+    if shape_hints is None:
+        shape_hints = _DEFAULT_DUMMY_SHAPE_HINTS
+    if value_hints is None:
+        value_hints = _DEFAULT_DUMMY_VALUE_HINTS
+
     input_names: List[str] = [inp.name for inp in inputs]
     input_sizes: List[int] = [inp.shape for inp in inputs]
     input_size_map = {name: size for name, size in zip(input_names, input_sizes)}
@@ -4353,6 +4439,7 @@ def dummy_tf_inference(
 
     input_dtypes: List[Any] = [inp.dtype for inp in inputs]
     input_datas = {}
+    value_hints_dict, default_value, has_default = _parse_value_hints(value_hints)
 
     # -cid
     if custom_input_op_name_np_data_path:
@@ -4409,7 +4496,17 @@ def dummy_tf_inference(
     else:
         if verification_datas is None:
             for input_name, input_size, input_dtype in zip(input_names, input_sizes, input_dtypes):
-                if test_data_nhwc is None:
+                hint_value = value_hints_dict.get(
+                    input_name,
+                    default_value if has_default else None,
+                )
+                if hint_value is not None:
+                    input_datas[input_name] = np.full(
+                        input_size,
+                        hint_value,
+                        dtype=TF_DTYPES_TO_NUMPY_DTYPES[input_dtype],
+                    )
+                elif test_data_nhwc is None:
                     input_datas[input_name] = np.ones(
                         input_size,
                         dtype=TF_DTYPES_TO_NUMPY_DTYPES[input_dtype],

{onnx2tf-1.29.20 → onnx2tf-1.29.21}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "uv_build"
 
 [project]
 name = "onnx2tf"
-version = "1.29.20"
+version = "1.29.21"
 description = "Self-Created Tools to convert ONNX files (NCHW) to TensorFlow/TFLite/Keras format (NHWC). The purpose of this tool is to solve the massive Transpose extrapolation problem in onnx-tensorflow (onnx-tf)."
 readme = "README.md"
 requires-python = ">=3.10.12"
@@ -28,7 +28,7 @@ classifiers = [
 dependencies = [
     "requests==2.32.5",
     "numpy==1.26.4",
-    "onnx==1.19.0",
+    "onnx==1.19.1",
     "onnxruntime==1.23.0",
     "opencv-python==4.11.0.86",
     "onnxsim==0.4.36",