Comments (6)
Ijustakid
commented on May 20, 2024
mobilenet-v1.yml
from mace-models.
lee-bin
commented on May 20, 2024
@Ijustakid 在导出TensorFlow模型的时候需要设置is_training为false
from mace-models.
Ijustakid
commented on May 20, 2024
@Ijustakid 在导出TensorFlow模型的时候需要设置is_training为false
convert_pb.py中：
out=sess.run(output_tensor_name, feed_dict={input_image_tensor: im,
input_keep_prob_tensor:1.0,
input_is_training_tensor:False})
已把is_training_tensor设为False
另：keep_prob的placeholder如何处理？
十分感谢！
br
from mace-models.
Ijustakid
commented on May 20, 2024
@Ijustakid 在导出TensorFlow模型的时候需要设置is_training为false
convert_pb.py
`input_image_tensor = sess.graph.get_tensor_by_name("input:0")
input_keep_prob_tensor = sess.graph.get_tensor_by_name("keep_prob:0")
input_is_training_tensor = sess.graph.get_tensor_by_name("is_training:0")
# 定义输出的张量名称
output_tensor_name = sess.graph.get_tensor_by_name("MobilenetV1/Logits/SpatialSqueeze:0")
# 读取测试图片
im=read_image(image_path,resize_height,resize_width,normalization=True)
im=im[np.newaxis,:]
# 测试读出来的模型是否正确，注意这里传入的是输出和输入节点的tensor的名字，不是操作节点的名字
# out=sess.run("InceptionV3/Logits/SpatialSqueeze:0", feed_dict={'input:0': im,'keep_prob:0':1.0,'is_training:0':False})
out=sess.run(output_tensor_name, feed_dict={input_image_tensor: im,
input_keep_prob_tensor:1.0,
input_is_training_tensor:False})`
from mace-models.
lee-bin
commented on May 20, 2024
没看到模型构建和导出的代码，只看到了session.run运行的代码，需要在模型构建时设置is_training和keep_prob
from mace-models.
Ijustakid
commented on May 20, 2024
没看到模型构建和导出的代码，只看到了session.run运行的代码，需要在模型构建时设置is_training和keep_prob
i will try it, thanks!
from mace-models.
Related Issues (20)
- detection_output.cc is missing HOT 2
- sre_16, Exception: Unexpected fc input ndim. HOT 1
- deploy Yolov3 with mace HOT 4
- configure output for yolov3 model HOT 1
- Support object detection models HOT 2
- how to set input data, after kaldi convert to onnx model, HOT 1
- YOLOV3模型裁剪三个输出 HOT 1
- yolo v3的fps HOT 1
- How to make model: caffe ssd-mobilenet-v1 support hexagon dsp . HOT 1
- Is there any description on how the input for each model should be preprocessed? HOT 1
- Please provide training scripts (or at least more details) for realtime-style-transfer model. HOT 1
- run on PC HOT 5
- fast-style-transfer模型的计算用时 HOT 5
- Nnet3 with input & i-vector
- Is there any plan to add EfficientDet into this repo? HOT 1
- how to add multi-inputs? or how to test the benchmark on multi inputs? HOT 4
- changing the input size of yolov3 HOT 3
- Support for tf.keras models(models from micro-models) HOT 1
- MobileAI Compute
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from mace-models.