onnx2tf 1.26.9__py3-none-any.whl → 1.27.0__py3-none-any.whl

onnx2tf/__init__.py

@@ -1,3 +1,3 @@
 from onnx2tf.onnx2tf import convert, main
 
-__version__ = '1.26.9'
+__version__ = '1.27.0'

onnx2tf/utils/common_functions.py

@@ -16,6 +16,7 @@ import subprocess
 import numpy as np
 np.random.seed(0)
 import tensorflow as tf
+from ai_edge_litert.interpreter import Interpreter
 import tf_keras
 from tensorflow.python.keras.layers import Lambda
 from tensorflow.python.keras.utils import conv_utils
@@ -4143,8 +4144,7 @@ def weights_export(
         Path to file in hdf5 format to save the extracted weights
     """
     import h5py
-    from tensorflow.lite.python import interpreter as interpreter_wrapper
-    interpreter = interpreter_wrapper.Interpreter(
+    interpreter = Interpreter(
         model_path=extract_target_tflite_file_path,
     )
     interpreter.allocate_tensors()

{onnx2tf-1.26.9.dist-info → onnx2tf-1.27.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: onnx2tf
-Version: 1.26.9
+Version: 1.27.0
 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).
 Home-page: https://github.com/PINTO0309/onnx2tf
 Author: Katsuya Hyodo
@@ -285,9 +285,11 @@ Video speed is adjusted approximately 50 times slower than actual speed.
 - onnx-simplifier==0.4.33 or 0.4.30 `(onnx.onnx_cpp2py_export.shape_inference.InferenceError: [ShapeInferenceError] (op_type:Slice, node name: /xxxx/Slice): [ShapeInferenceError] Inferred shape and existing shape differ in rank: (x) vs (y))`
 - onnx_graphsurgeon
 - simple_onnx_processing_tools
-- tensorflow==2.17.0, Special bugs: [#436](https://github.com/PINTO0309/onnx2tf/issues/436)
+- tensorflow==2.19.0, Special bugs: [#436](https://github.com/PINTO0309/onnx2tf/issues/436)
+- tf-keras==2.19.0
+- ai-edge-litert==1.2.0
 - psutil==5.9.5
-- ml_dtypes==0.3.2
+- ml_dtypes==0.5.1
 - flatbuffers-compiler (Optional, Only when using the `-coion` option. Executable file named `flatc`.)
 - flatbuffers>=23.1.21
   ```bash
@@ -331,7 +333,7 @@ Video speed is adjusted approximately 50 times slower than actual speed.
   docker run --rm -it \
   -v `pwd`:/workdir \
   -w /workdir \
-  ghcr.io/pinto0309/onnx2tf:1.26.9
+  ghcr.io/pinto0309/onnx2tf:1.27.0
 
   or
 
@@ -339,11 +341,11 @@ Video speed is adjusted approximately 50 times slower than actual speed.
   docker run --rm -it \
   -v `pwd`:/workdir \
   -w /workdir \
-  docker.io/pinto0309/onnx2tf:1.26.9
+  docker.io/pinto0309/onnx2tf:1.27.0
 
   or
 
-  pip install -U onnx==1.16.1 \
+  pip install -U onnx==1.17.0 \
   && pip install -U nvidia-pyindex \
   && pip install -U onnx-graphsurgeon \
   && pip install -U onnxruntime==1.18.1 \
@@ -351,13 +353,14 @@ Video speed is adjusted approximately 50 times slower than actual speed.
   && pip install -U simple_onnx_processing_tools \
   && pip install -U sne4onnx>=1.0.13 \
   && pip install -U sng4onnx>=1.0.4 \
-  && pip install -U tensorflow==2.17.0 \
+  && pip install -U ai_edge_litert==1.2.0 \
+  && pip install -U tensorflow==2.19.0 \
   && pip install -U protobuf==3.20.3 \
   && pip install -U onnx2tf \
   && pip install -U h5py==3.11.0 \
   && pip install -U psutil==5.9.5 \
-  && pip install -U ml_dtypes==0.3.2 \
-  && pip install -U tf-keras~=2.16 \
+  && pip install -U ml_dtypes==0.5.1 \
+  && pip install -U tf-keras==2.19.0 \
   && pip install flatbuffers>=23.5.26
 
   or
@@ -381,8 +384,9 @@ or
     && sudo chmod +x flatc \
     && sudo mv flatc /usr/bin/
   !pip install -U pip \
-    && pip install tensorflow==2.17.0 \
-    && pip install -U onnx==1.16.1 \
+    && pip install tensorflow==2.19.0 \
+    && pip install ai_edge_litert==1.2.0 \
+    && pip install -U onnx==1.17.0 \
     && python -m pip install onnx_graphsurgeon \
           --index-url https://pypi.ngc.nvidia.com \
     && pip install -U onnxruntime==1.18.1 \
@@ -392,8 +396,8 @@ or
     && pip install -U protobuf==3.20.3 \
     && pip install -U h5py==3.11.0 \
     && pip install -U psutil==5.9.5 \
-    && pip install -U ml_dtypes==0.3.2 \
-    && pip install -U tf-keras~=2.16 \
+    && pip install -U ml_dtypes==0.5.1 \
+    && pip install -U tf-keras==2.19.0 \
     && pip install flatbuffers>=23.5.26
   ```
 
@@ -608,7 +612,7 @@ import onnxruntime
 import numpy as np
 import onnx2tf
 import tensorflow as tf
-from tensorflow.lite.python import interpreter as tflite_interpreter
+from ai_edge_litert.interpreter import Interpreter
 
 class Model(torch.nn.Module):
     def forward(self, x, y):
@@ -647,7 +651,7 @@ onnx2tf.convert(
 )
 
 # Now, test the newer TFLite model
-interpreter = tf.lite.Interpreter(model_path="model.tf/model_float32.tflite")
+interpreter = Interpreter(model_path="model.tf/model_float32.tflite")
 tf_lite_model = interpreter.get_signature_runner()
 inputs = {
   'x': np.asarray([10], dtype=np.int64),
@@ -1061,10 +1065,10 @@ Now, let's try inference with the TFLite runtime instead of the TensorFlow runti
   import time
   import numpy as np
   np.random.seed(0)
-  import tensorflow as tf
+  from ai_edge_litert.interpreter import Interpreter
 
   # Load TFLite model
-  interpreter = tf.lite.Interpreter(model_path="./saved_model/prelu_check_float32.tflite")
+  interpreter = Interpreter(model_path="./saved_model/prelu_check_float32.tflite")
   interpreter.allocate_tensors()
   tensor_shape = (256, 20)
   input_data = {'waveform': np.random.randn(*tensor_shape).astype(np.float32)}
@@ -1232,10 +1236,10 @@ You can use `signature_runner` to handle dynamic input tensors by performing inf
 - `test.py` - Batch size: `5`
   ```python
   import numpy as np
-  import tensorflow as tf
+  from ai_edge_litert.interpreter import Interpreter
   from pprint import pprint
 
-  interpreter = tf.lite.Interpreter(model_path="saved_model/osnet_x0_25_msmt17_float32.tflite")
+  interpreter = Interpreter(model_path="saved_model/osnet_x0_25_msmt17_float32.tflite")
   tf_lite_model = interpreter.get_signature_runner()
   inputs = {
       'images': np.ones([5,256,128,3], dtype=np.float32),
@@ -1263,10 +1267,10 @@ You can use `signature_runner` to handle dynamic input tensors by performing inf
 - `test.py` - Batch size: `3`
   ```python
   import numpy as np
-  import tensorflow as tf
+  from ai_edge_litert.interpreter import Interpreter
   from pprint import pprint
 
-  interpreter = tf.lite.Interpreter(model_path="saved_model/osnet_x0_25_msmt17_float32.tflite")
+  interpreter = Interpreter(model_path="saved_model/osnet_x0_25_msmt17_float32.tflite")
   tf_lite_model = interpreter.get_signature_runner()
   inputs = {
       'images': np.ones([3,256,128,3], dtype=np.float32),
@@ -1350,11 +1354,11 @@ The relationship between the ONNX before conversion and the TFLite file after co
 Use the generated TFLite file to inference and ensure that it always contains fixed value output.
 
 ```python
-import tensorflow as tf
+from ai_edge_litert.interpreter import Interpreter
 import numpy as np
 from pprint import pprint
 
-interpreter = tf.lite.Interpreter(model_path="saved_model/toy_with_constant_float32.tflite")
+interpreter = Interpreter(model_path="saved_model/toy_with_constant_float32.tflite")
 interpreter.allocate_tensors()
 
 input_details = interpreter.get_input_details()

{onnx2tf-1.26.9.dist-info → onnx2tf-1.27.0.dist-info}/RECORD

@@ -1,4 +1,4 @@
-onnx2tf/__init__.py,sha256=14sed6LJ9qrrV90jfdiUVlID3DheS0YBDgOQlRIbt_U,66
+onnx2tf/__init__.py,sha256=lHhAWe8iqi6znmPVuWY05CVVp6042_wHUvkziLiOvA8,66
 onnx2tf/__main__.py,sha256=2RSCQ7d4lc6CwD-rlGn9UicPFg-P5du7ZD_yh-kuBEU,57
 onnx2tf/onnx2tf.py,sha256=IEnfIs3Dy8Y5F3iJ4HY7bWkn3QuB6lq_gHa1q5E3tMI,124745
 onnx2tf/ops/Abs.py,sha256=V7btmCG_ZvK_qJovUsguq0ZMJ349mhNQ4FHSgzP_Yuo,4029
@@ -185,13 +185,13 @@ onnx2tf/ops/_Loop.py,sha256=eo5sNfrfOnKV6_I737AWsM5LJTY9DVOxQEvhanxtP4g,11322
 onnx2tf/ops/__Loop.py,sha256=ClwMcbNS4hqUtW_pzwjMa9Cqg7ONvz9aplke55A0uJ0,19704
 onnx2tf/ops/__init__.py,sha256=jnmUWWa-3dHzBZV9bmPzXu6eoz2dumJTzO7i8JdcgSM,25
 onnx2tf/utils/__init__.py,sha256=E9FM9He68VIASDnYp-OrxvHFVn55GzWqw2OEkCqn1zg,27
-onnx2tf/utils/common_functions.py,sha256=lhhWuNVDwM_mVFc9DD2kYcYEyyT46CDlJMvZzi9KWD4,241473
+onnx2tf/utils/common_functions.py,sha256=HTDca3DGXB3xvc1S50RNscgB57TCiq4yC5Nrafs6ka4,241430
 onnx2tf/utils/enums.py,sha256=7c5TqetqB07VjyHoxJHfLgtqBqk9ZRyUF33fPOJR1IM,1649
 onnx2tf/utils/logging.py,sha256=yUCmPuJ_XiUItM3sZMcaMO24JErkQy7zZwVTYWAuiKg,1982
-onnx2tf-1.26.9.dist-info/LICENSE,sha256=5v_Kxihy8i6mzHVl349ikSREaIdsl9YeUnX1KBDLD2w,1070
-onnx2tf-1.26.9.dist-info/LICENSE_onnx-tensorflow,sha256=gK4GtS9S5YcyINu6uuNNWdo-kBClyEM4MFLFGiNTeRM,11231
-onnx2tf-1.26.9.dist-info/METADATA,sha256=YlHKAl6m5rKRbF5pN8-jvIFUDvVTX9Mg-eWdkJvUkKM,147499
-onnx2tf-1.26.9.dist-info/WHEEL,sha256=52BFRY2Up02UkjOa29eZOS2VxUrpPORXg1pkohGGUS8,91
-onnx2tf-1.26.9.dist-info/entry_points.txt,sha256=gDPK8ToCFPKMvm8jr9xrGOkXtORJJVh4736fBEKO5k0,41
-onnx2tf-1.26.9.dist-info/top_level.txt,sha256=WgfPiEy3f6vZ_FOpAIEA2CF3TCx1eYrhGw93Ih6b9Fw,8
-onnx2tf-1.26.9.dist-info/RECORD,,
+onnx2tf-1.27.0.dist-info/LICENSE,sha256=5v_Kxihy8i6mzHVl349ikSREaIdsl9YeUnX1KBDLD2w,1070
+onnx2tf-1.27.0.dist-info/LICENSE_onnx-tensorflow,sha256=gK4GtS9S5YcyINu6uuNNWdo-kBClyEM4MFLFGiNTeRM,11231
+onnx2tf-1.27.0.dist-info/METADATA,sha256=Qjvfs5wxntQGejt30wlRjfCY1PA2gtg8d2UOLJ_8GEk,147683
+onnx2tf-1.27.0.dist-info/WHEEL,sha256=52BFRY2Up02UkjOa29eZOS2VxUrpPORXg1pkohGGUS8,91
+onnx2tf-1.27.0.dist-info/entry_points.txt,sha256=gDPK8ToCFPKMvm8jr9xrGOkXtORJJVh4736fBEKO5k0,41
+onnx2tf-1.27.0.dist-info/top_level.txt,sha256=WgfPiEy3f6vZ_FOpAIEA2CF3TCx1eYrhGw93Ih6b9Fw,8
+onnx2tf-1.27.0.dist-info/RECORD,,