ketiswp/google-coral-DeepLabV3-MobileNetV2-0.5-PascalVOC-fp32-onnx

Google Coral DeepLabV3 MobileNetV2 0.5 Pascal VOC FP32 ONNX

FP32 ONNX version of Google Coral DeepLabV3 MobileNetV2 0.5 Pascal VOC for semantic segmentation.

Model Files

File	Purpose	Format
model.onnx	Downloadable converted model	ONNX
source/model.pb	Original model	TensorFlow GraphDef
graphs/netron.png	ONNX graph visualization	PNG
mlir/onnx.mlir	ONNX-MLIR import result	MLIR text

Parameter Summary

Item	Value
Precision	FP32
ONNX file size	2.67 MiB
Initializer tensors	134
Stored initializer elements	689,172
External weight files	None
Original format	TensorFlow GraphDef

Stored initializer elements includes weights, biases, quantization scales, zero-points, and other constant tensors. It is not a trainable-parameter count.

Original Model Inference

pip install huggingface_hub numpy tensorflow

import numpy as np
import tensorflow as tf
from huggingface_hub import hf_hub_download

repo_id = "ketiswp/google-coral-DeepLabV3-MobileNetV2-0.5-PascalVOC-fp32-onnx"
model_path = hf_hub_download(repo_id=repo_id, filename="source/model.pb")
graph_def = tf.compat.v1.GraphDef()
graph_def.ParseFromString(open(model_path, "rb").read())
graph = tf.Graph()
with graph.as_default():
    tf.import_graph_def(graph_def, name="")

input_tensor = graph.get_tensor_by_name("ImageTensor:0")
output_tensor = graph.get_tensor_by_name("SemanticPredictions:0")
input_value = np.zeros((1, 513, 513, 3), dtype=input_tensor.dtype.as_numpy_dtype)

config = tf.compat.v1.ConfigProto(
    intra_op_parallelism_threads=1,
    inter_op_parallelism_threads=1,
    device_count={"GPU": 0},
)
with tf.compat.v1.Session(graph=graph, config=config) as session:
    output = session.run(output_tensor, feed_dict={input_tensor: input_value})
print(output.shape, output.dtype)

Converted ONNX Inference

pip install huggingface_hub numpy onnxruntime

import numpy as np
import onnxruntime as ort
from huggingface_hub import hf_hub_download

repo_id = "ketiswp/google-coral-DeepLabV3-MobileNetV2-0.5-PascalVOC-fp32-onnx"
model_path = hf_hub_download(repo_id=repo_id, filename="model.onnx")

options = ort.SessionOptions()
options.intra_op_num_threads = 1
options.inter_op_num_threads = 1
options.execution_mode = ort.ExecutionMode.ORT_SEQUENTIAL
options.graph_optimization_level = ort.GraphOptimizationLevel.ORT_ENABLE_ALL
session = ort.InferenceSession(
    model_path,
    sess_options=options,
    providers=["CPUExecutionProvider"],
)

dtype_by_ort_type = {
    "tensor(float)": np.float32,
    "tensor(double)": np.float64,
    "tensor(float16)": np.float16,
    "tensor(int64)": np.int64,
    "tensor(int32)": np.int32,
    "tensor(int16)": np.int16,
    "tensor(int8)": np.int8,
    "tensor(uint8)": np.uint8,
    "tensor(bool)": np.bool_,
}
feeds = {}
for item in session.get_inputs():
    shape = [dim if isinstance(dim, int) and dim > 0 else 1 for dim in item.shape]
    feeds[item.name] = np.zeros(shape, dtype=dtype_by_ort_type[item.type])

outputs = session.run(None, feeds)
print([(item.name, value.shape, str(value.dtype))
       for item, value in zip(session.get_outputs(), outputs)])

Paired Model

UINT8 version

Source

• Original artifact
• Source project

Project Validation

FP32/quantized comparison, conversion results, and reproduction code

This repository also includes the Netron graph, ONNX Dialect MLIR, and static MLIR dependency graph for this model variant.