<think>
Let me
analyze this code
and provide a helpful response
. This appears
to be a Python script for training a convolutional neural network
(CNN
) on the MNIST h
andwritten digit dataset using Py
Torch,
and then exporting the trained model
to ONNX format
.
The code consists of several parts:
Importing necessary libraries
and setting hyperparameters
Preparing the MNIST dataset
with tr
ansformations
Visualizing some examples from the dataset
Defining a CNN model architecture using Py
Torch
Setting up loss function
and optimizer
Defining training
and testing functions
Training the model for 10 epochs
and saving it
Exporting the trained model
to ONNX format
At the end, the user mentions they w
ant
to convert their ONNX model
to RKNN format but are having version discrep
ancies
. They're using rknn-
toolkit2 version 2
.3
.0 in their code, but on their Ubuntu system they have rknn-
toolkit2 version 1
.6
.0
with Python 3
.10
.
I'll focus on providing a solution for
converting their ONNX model
to RKNN format, considering the version discrep
ancy they're experiencing
.
Let me first underst
and the model architecture better:
Input: MNIST images
(1 ch
annel, 28x28 pixels
)
First convolutional layer: 1→10 ch
annels, 5x5 kernel, followed by ReLU
and 2x2 max pooling
Second convolutional layer: 10→20 ch
annels, 5x5 kernel, followed by ReLU
and 2x2 max pooling
Fully connected layers:
320→50→10
(output
classes for digits 0-9
)
The exported ONNX model uses:
Input shape: [1, 1, 28, 28]
Output shape: [1, 10]
(representing probabilities for digits 0-9
)
ONNX opset version: 19
Now, the issue is the version discrep
ancy between the rknn-
toolkit2 they're using in their code
(2
.3
.0
) and what's installed on their Ubuntu system
(1
.6
.0
). I need
to provide code that will work
with their installed version
(1
.6
.0
) to convert the ONNX model
to RKNN format
.
</think>
I see you've shared a Py
Torch script for training a CNN model on the MNIST dataset, followed by ONNX conversion
. You mentioned that you need
to convert this ONNX model
to RKNN format, but there's a version mismatch between the rknn-
toolkit2 in your code
(2
.3
.0
) and what's installed on your Ubuntu system
(1
.6
.0
).
Let me create a compatible RKNN conversion script for your system:
<artifact id="rknn_conversion_script"
type="application/vnd
.ant
.code" l
anguage="python" title="MNIST ONNX
to RKNN Conversion Script"> from rknn
.api import RKNN
Configuration
ONNX_MODEL = '
./model_Mnist_fixed
.onnx'
RKNN_MODEL = '
./model_Mnist
.rknn'
Initialize RKNN object
rknn = RKNN
(verbose=True
)
Pre-process config for input
print
('--> Config model'
)
rknn
.config
(me
an_
values=[127
.5], std_
values=[127
.5], target_pla
tform='rk3588'
)
print
('done'
)
Load ONNX model
print
('--> Loading model'
)
ret = rknn
.load_onnx
(model=ONNX_MODEL
)
if ret != 0:
print
('Load ONNX model failed!'
)
exit
(ret
)
print
('done'
)
Build model
print
('--> Building model'
)
ret = rknn
.build
(do_qu
antization=True, dataset='
./dataset
.txt'
)
if ret != 0:
print
('Build model failed!'
)
exit
(ret
)
print
('done'
)
Export RKNN model
print
('--> Export RKNN model'
)
ret = rknn
.export_rknn
(RKNN_MODEL
)
if ret != 0:
print
('Export RKNN model failed!'
)
exit
(ret
)
print
('done'
)
Optional: Inference testing
print
('--> Testing model on sample data'
)
Set inputs
img = cv2
.imread
('sample_digit
.png', cv2
.IMREAD_GRAYSCALE
)
img = cv2
.resize
(img,
(28, 28
))
img = np
.exp
and_dims
(img, axis=0
) # Add batch dimension
img = np
.exp
and_dims
(img, axis=0
) # Add ch
annel dimension
Init runtime environment
ret = rknn
.init_runtime
()
if ret != 0:
print
('Init runtime environment failed'
)
exit
(ret
)
Inference
print
('--> Running model'
)
outputs = rknn
.inference
(inputs=[img]
)
predicted = np
.argmax
(outputs[0][0]
)
print
(f'Predicted digit: {predicted}'
)
Release RKNN context
rknn
.release
()
</artifact>
Additional Setup for RKNN Conversion
For the above script
to work properly, you'll need:
Create a calibration dataset file: RKNN qu
antization requires a dataset file listing paths
to sample images:
<artifact id="dataset_genera
tor"
type="application/vnd
.ant
.code" l
anguage="python" title="Calibration Dataset Genera
tor"> import os import
numpy as np from
torchvision import datasets, tr
ansforms import
torch
Create dataset direc
tory if it doesn't exist
os
.makedirs
('
./calibration_images', exist_ok=True
)
Load a subset of MNIST images for calibration
tr
ansform = tr
ansforms
.Compose
([tr
ansforms
.ToTensor()]
)
dataset = datasets
.MNIST
(root='
./data/mnist', train=True, tr
ansform=tr
ansform, download=True
)
data_loader =
torch
.utils
.data
.DataLoader
(dataset, batch_size=1, shuffle=True
)
Save calibration images
(using 100 images
)
with open
('
./dataset
.txt', 'w'
) as f:
for i,
(img,
) in enumerate
(data_loader
):
if i >= 100: # Use 100 images for calibration
break
#
Convert tensor to numpy and save as raw format
img_np = img
.numpy().reshape
(28, 28
) * 255
img_np = img_np
.as
type(np
.uint8
)
img_path = f'
./calibration_images/img{i:03d}
.bin'
img_np
.tofile
(img_path
)
f
.write
(f'{img_path}\n'
)
print
(f"Created dataset
.txt
with {min
(100, i+1
)} image paths for calibration"
)
</artifact>
Version Compatibility Notes
Since you're using rknn-
toolkit2 version 1
.6
.0 on Ubuntu
with Python 3
.10, here are some import
ant compatibility considerations:
API differences: The script I provided uses the API structure compatible
with older versions
. The newer 2
.3
.0 version might have slightly different parameter names or
methods
.
Target pla
tform: I specified 'rk3588' as the target pla
tform
. Adjust this if you're targeting a different RK chip
.
Qu
antization parameters: The script uses basic qu
antization settings
. You may need
to adjust these
(particularly me
an_
values
and std_
values
) based on your model's preprocessing
.
Testing the model: The optional testing section requires you
to have a sample image
. You c
an generate one from your test dataset or skip this section if not needed
.
Troubleshooting Common Issues
If you encounter errors:
ONNX compatibility: Older RKNN versions may not support ONNX opset 19
. Try re-exporting your ONNX model
with a lower opset version
(e
.g
., 11 or 13
).
Layer support: Some operations might not be supported
. Check error messages for
unsupported operations
and consider simplifying your model if needed
.
Qu
antization issues: If qu
antization fails, try do_qu
antization=False first
to verify the conversion works
without qu
antization
.
Would you like me
to explain
any specific part of the conversion process in more detail?
先仔细阅读这部分内容，再来结合这上面的讲解，最终生成适合的转化rknn代码
在进行深度学习时遇到Tensorflow2.3.0与Numpy1.23.5版本不兼容的问题，导致运行错误。通过将数据从DataFrame转换为array，再用tf.convert_to_tensor方法避免了错误，成功将数据转换为tensor。无需调整Tensorflow和Numpy的版本。
扫一扫
1332
到【灌水乐园】发言
