Comments (13)
from convolutional-pose-machine-tf.
我看这个可能是你输入通道的问题。请你检查你的输入图像。 Regards
…
在 2021年5月14日,19:01,FlyFish0101 @.***> 写道: [CAUTION: Non-UBC Email] Closed #23. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe.
万分感谢,把puttext那一段话注释掉就运行成功了
from convolutional-pose-machine-tf.
from convolutional-pose-machine-tf.
应该是你的图像带 alpha 通道,你检查下
…
在 2021年5月14日,20:10,FlyFish0101 @.> 写道: [CAUTION: Non-UBC Email] 我看这个可能是你输入通道的问题。请你检查你的输入图像。 Regards … 在 2021年5月14日,19:01,FlyFish0101 @.> 写道: [CAUTION: Non-UBC Email] Closed #23. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe. 万分感谢,把puttext那一段话注释掉就运行成功了 — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.
请问将npy转换成ckpt的操作大致是什么呢?我搜到一些代码但是感觉不完全适用,谢谢
from convolutional-pose-machine-tf.
from convolutional-pose-machine-tf.
使用 npy load 模型,再存 ckpt
…
在 2021年5月14日,20:18,FlyFish0101 @.***> 写道: [CAUTION: Non-UBC Email] 应该是你的图像带 alpha 通道,你检查下 … 在 2021年5月14日,20:10,FlyFish0101 @.> 写道: [CAUTION: Non-UBC Email] 我看这个可能是你输入通道的问题。请你检查你的输入图像。 Regards … 在 2021年5月14日,19:01,FlyFish0101 @.> 写道: [CAUTION: Non-UBC Email] Closed #23. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe. 万分感谢,把puttext那一段话注释掉就运行成功了 — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. 请问将npy转换成ckpt的操作大致是什么呢?我搜到一些代码但是感觉不完全适用,谢谢 — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.
请问我能否在CPM.py里修改build_model函数,在self.sess.run(tf.global_variables_initializer())这句话后面添加
self.saver = tf.train.Saver()
self.saver.save(self.sess, os.path.join(self.log_dir, "model.ckpt"))
来实现加载并赋值后模型的保存呢
因为npy转ckpt好像要先引入结构再load数据,我是新手感觉您的代码的结构很难自己搭出来,想借用您现成有的模板(CPM.py),不知道这样可行否
from convolutional-pose-machine-tf.
from convolutional-pose-machine-tf.
不可以,你在构建模型以后,要确保使用了npy的参数才可以save。可以参考 demo 中的代码,npy 读取以后 save session
…
在 2021年5月15日,11:02,FlyFish0101 @.> 写道: [CAUTION: Non-UBC Email] 使用 npy load 模型,再存 ckpt … 在 2021年5月14日,20:18,FlyFish0101 @.> 写道: [CAUTION: Non-UBC Email] 应该是你的图像带 alpha 通道,你检查下 … 在 2021年5月14日,20:10,FlyFish0101 @.> 写道: [CAUTION: Non-UBC Email] 我看这个可能是你输入通道的问题。请你检查你的输入图像。 Regards … 在 2021年5月14日,19:01,FlyFish0101 @.> 写道: [CAUTION: Non-UBC Email] Closed #23. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe. 万分感谢,把puttext那一段话注释掉就运行成功了 — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. 请问将npy转换成ckpt的操作大致是什么呢?我搜到一些代码但是感觉不完全适用,谢谢 — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. 请问我能否在CPM.py里修改build_model函数,在self.sess.run(tf.global_variables_initializer())这句话后面添加 self.saver = tf.train.Saver() self.saver.save(self.sess, os.path.join(self.log_dir, "model.ckpt")) 来实现加载并赋值后模型的保存呢 — You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub, or unsubscribe.
np.load(xxx.npy)是在init部分,buildmodel代码运行时间线迟于init的吧
from convolutional-pose-machine-tf.
from convolutional-pose-machine-tf.
我也有几年没有做 tensorflow v1 了。如果能保证 init variable 不会初始化 weight 就可以了。之后就导出 session 就行了
…
在 2021年5月15日,11:13,FlyFish0101 @.***> 写道: [CAUTION: Non-UBC Email] 不可以,你在构建模型以后,要确保使用了npy的参数才可以save。可以参考 demo 中的代码,npy 读取以后 save session … 在 2021年5月15日,11:02,FlyFish0101 @.> 写道: [CAUTION: Non-UBC Email] 使用 npy load 模型,再存 ckpt … 在 2021年5月14日,20:18,FlyFish0101 @.> 写道: [CAUTION: Non-UBC Email] 应该是你的图像带 alpha 通道,你检查下 … 在 2021年5月14日,20:10,FlyFish0101 @.> 写道: [CAUTION: Non-UBC Email] 我看这个可能是你输入通道的问题。请你检查你的输入图像。 Regards … 在 2021年5月14日,19:01,FlyFish0101 @.> 写道: [CAUTION: Non-UBC Email] Closed #23. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe. 万分感谢,把puttext那一段话注释掉就运行成功了 — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. 请问将npy转换成ckpt的操作大致是什么呢?我搜到一些代码但是感觉不完全适用,谢谢 — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. 请问我能否在CPM.py里修改build_model函数,在self.sess.run(tf.global_variables_initializer())这句话后面添加 self.saver = tf.train.Saver() self.saver.save(self.sess, os.path.join(self.log_dir, "model.ckpt")) 来实现加载并赋值后模型的保存呢 — You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub, or unsubscribe. np.load(xxx.npy)是在init部分,buildmodel代码运行时间线迟于init的吧 — You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub, or unsubscribe.
谢谢大佬!!意思是self.sess.run(tf.global_variables_initializer())可能会不用npy里的参数初始化数据嘛,所以要确保这点嘛?
另附一点疑惑:依据大佬做骨骼点检测的经验,有没有办法只关注某三个点(比如左侧的手腕,胳膊肘和肩)的训练方法呢(因为我们网络的使用场景不用做其他部分检测),或许可以把joints_list的内容改成3个?(大佬见笑了)
from convolutional-pose-machine-tf.
from convolutional-pose-machine-tf.
第一点,是 第二点,可以,只需要配合你的标注更改一下 pred channel 就可以。 你客气了,回答你的问题是我的义务。
…
您好,您程序里提供的只关注某些interst joints相关的api非常实用,已经在跑训练了,还想问下,这个mobileCPM网络还有没有进一步裁剪结构的可能,比如把stage改为2可以吗,或者把前面的mobilenet的backbone裁剪一点
from convolutional-pose-machine-tf.
from convolutional-pose-machine-tf.
Related Issues (20)
- 关于net的疑惑 HOT 1
- Unable to access the model from offered site HOT 2
- Error when trainning. HOT 2
- about trained model HOT 2
- train自己的数据报错 HOT 15
- 能得到关节点的具体坐标信息吗 HOT 3
- demo下面没有demo,model下面也不是model HOT 1
- 训练MPII数据集 HOT 3
- 关于自己测试图片的问题
- MPII数据结果和论文差别大吗? HOT 2
- 关于预训练模型vgg.和mobile v1的问题 HOT 1
- 关于TensorFlow的ckpt模型问题 HOT 1
- 您好,model下载网址好像失效了 HOT 2
- 报错 求指导 HOT 8
- 关于训练中的train中build_accuracy的问题 HOT 1
- 关于valPCK 问题 HOT 6
- 关于数据 HOT 1
- bug:在datagen_mpii.py padding时,高度的padding好像有错误 HOT 1
- A bug in net/CPM.py HOT 1
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 convolutional-pose-machine-tf.