EfficientNet Models
Overview
EfficientNets employ a streamlined approach to scaling by utilizing a simple yet effective compound coefficient to uniformly adjust the depth, width, and resolution of the network. Building upon this, a novel baseline network was designed using neural architecture search, which was then scaled to produce a series of models known as EfficientNets. These models offer superior accuracy and efficiency compared to traditional ConvNets. Additionally, EfficientNets demonstrate strong performance in transfer learning tasks.
Getting Started
Follow these steps to use and convert EfficientNet models using PyTorch and TorchVision.
Install Required Libraries:
Ensure you have the necessary libraries installed:
pip install torch torchvision
Load and Convert EfficientNet Model:
Load a pretrained EfficientNet model using PyTorch and TorchVision, create a dummy input tensor for tracing, trace the model to convert it to TorchScript, and finally save the traced model.
import torch import torchvision model = torchvision.models.efficientnet_b0(pretrained=True) trace_data = torch.randn(1, 3, 224, 224) trace_model = torch.jit.trace(model.cpu().eval(), trace_data) torch.jit.save(trace_model, 'efficientnet_b0.pt')
How It Works ?
Before you begin, ensure that the NeuroPilot Converter Tool is installed.
Quant8 Conversion Process
Generate Calibration Data:
The following script creates a directory named data and generates 100 batches of random input data, each saved as a .npy file. This data is used for calibration during the quantization process.
import os import numpy as np os.mkdir('data') for i in range(100): data = np.random.randn(1, 3, 224, 224).astype(np.float32) np.save('data/batch_{}.npy'.format(i), data)
Convert to Quantized TFLite Format:
Use the following command to convert the model to a quantized TFLite format using the generated calibration data:
mtk_pytorch_converter \ --input_script_module_file=efficientnet_b0.pt \ --output_file=efficientnet_b0_ptq_quant.tflite \ --input_shapes=1,3,224,224 \ --quantize=True \ --input_value_ranges=-1,1 \ --calibration_data_dir=data/ \ --calibration_data_regexp=batch_.*\.npy
FP32 Conversion Process
To convert the model to a non-quantized (FP32) TFLite format, use the following command:
mtk_pytorch_converter \
--input_script_module_file=efficientnet_b0.pt \
--output_file=efficientnet_b0.tflite \
--input_shapes=1,3,224,224
Model Details
General Information
Property |
Value |
---|---|
Category |
Classification |
Input Size |
224x224 |
GFLOPS |
0.39 |
#Params (M) |
5.28 |
Training Framework |
PyTorch |
Inference Framework |
TFLite |
Quant8 Model package |
|
Float32 Model package |
Model Properties
Quant8
Format: TensorFlow Lite v3
Description: Exported by NeuroPilot converter v7.14.1+release
Inputs
Property |
Value |
---|---|
Name |
x.2 |
Tensor |
int8[1,3,224,224] |
Identifier |
155 |
Quantization |
Linear |
Quantization Range |
-1.0039 ≤ 0.0078 * q ≤ 0.9961 |
Outputs
Property |
Value |
---|---|
Name |
1245 |
Tensor |
int8[1,1000] |
Identifier |
115 |
Quantization |
Linear |
Quantization Range |
-1.0636 ≤ 0.0102 * (q + 24) ≤ 1.5443 |
Fp32
Format: TensorFlow Lite v3
Description: Exported by NeuroPilot converter v7.14.1+release
Inputs
Property |
Value |
---|---|
Name |
x.2 |
Tensor |
float32[1,3,224,224] |
Identifier |
344 |
Outputs
Property |
Value |
---|---|
Name |
1245 |
Tensor |
float32[1,1000] |
Identifier |
139 |
Performance Benchmarks
EfficientNet-quant8
Run model (.tflite) 10 times |
CPU (Thread:8) |
GPU |
ARMNN(GpuAcc) |
ARMNN(CpuAcc) |
Neuron Stable Delegate(APU) |
APU(MDLA) |
APU(VPU) |
G350 |
N/A |
N/A |
N/A |
N/A |
N/A |
N/A |
N/A |
G510 |
N/A |
N/A |
N/A |
N/A |
N/A |
4.05 ms |
N/A |
G700 |
N/A |
N/A |
N/A |
N/A |
N/A |
3.00 ms |
N/A |
G1200 |
N/A |
N/A |
N/A |
N/A |
N/A |
3.05 ms |
N/A |
EfficientNet-fp32
Run model (.tflite) 10 times |
CPU (Thread:8) |
GPU |
ARMNN(GpuAcc) |
ARMNN(CpuAcc) |
Neuron Stable Delegate(APU) |
APU(MDLA) |
APU(VPU) |
G350 |
198.458 ms (Thread:4) |
73.741 ms |
102.228 ms |
123.873 ms |
N/A |
N/A |
1517.21 ms |
G510 |
174.425 ms |
27.616 ms |
40.920 ms |
37.749 ms |
8.272 ms |
9.03 ms |
N/A |
G700 |
49.287 ms |
19.659 ms |
30.424 ms |
31.363 ms |
5.749 ms |
6.04 ms |
N/A |
G1200 |
44.765 ms |
14.796 ms |
22.689 ms |
20.943 ms |
6.636 ms |
6.05 ms |
N/A |
Widespread: CPU only, light workload.
Performance: CPU and GPU, medium workload.
Ultimate: CPU, GPU, and APUs, heavy workload.