Skip to content

Commit

Permalink
update auto compress classication readme (PaddlePaddle#1191)
Browse files Browse the repository at this point in the history
  • Loading branch information
ceci3 authored Jun 24, 2022
1 parent 52fba48 commit 9a07386
Showing 1 changed file with 6 additions and 6 deletions.
12 changes: 6 additions & 6 deletions demo/auto_compression/image_classification/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -16,7 +16,7 @@
本示例将以图像分类模型MobileNetV1为例,介绍如何使用PaddleClas中Inference部署模型进行自动压缩。本示例使用的自动压缩策略为量化训练和蒸馏。

## 2. Benchmark
- MobileNetV1模型
- PaddlePaddle MobileNetV1模型

| 模型 | 策略 | Top-1 Acc | 耗时(ms) threads=4 |
|:------:|:------:|:------:|:------:|
Expand All @@ -25,14 +25,14 @@

- 测试环境:`SDM710 2*A75(2.2GHz) 6*A55(1.7GHz)`

- MobileNetV1模型
- TensorFlow MobileNetV1模型

| 模型 | 策略 | Top-1 Acc | 耗时(ms) threads=4 | Inference模型 |
| 模型 | 策略 | Top-1 Acc | 耗时(ms) threads=1 | Inference模型 |
|:------:|:------:|:------:|:------:|:------:|
| MobileNetV1 | Base模型 | 71.0 | - | [Model]() |
| MobileNetV1 | 量化+蒸馏 | 70.22 | -| [Model]() |
| MobileNetV1 | Base模型 | 71.0 | 30.45 | [Model](https://paddle-slim-models.bj.bcebos.com/act/mobilenetv1_inference_model_tf2paddle.tar) |
| MobileNetV1 | 量化+蒸馏 | 70.22 | 15.86 | [Model](https://paddle-slim-models.bj.bcebos.com/act/mobilenetv1_quant.tar) |

- 测试环境:
- 测试环境:`骁龙865 4*A77 4*A55`

说明:
- MobileNetV1模型源自[tensorflow/models](http://download.tensorflow.org/models/mobilenet_v1_2018_02_22/mobilenet_v1_1.0_224.tgz),通过[X2Paddle](https://github.com/PaddlePaddle/X2Paddle)工具转换MobileNetV1预测模型步骤:
Expand Down

0 comments on commit 9a07386

Please sign in to comment.