keyird / deeplearning-tensorflow2 Goto Github PK
View Code? Open in Web Editor NEW本仓库我将使用谷歌TensorFlow2框架逐一复现经典的卷积神经网络:LeNet、AlexNet、VGG系列、GooLeNet、ResNet 系列、DenseNet 系列,以及现在比较经典的目标检测网络、语义分割网络等。
本仓库我将使用谷歌TensorFlow2框架逐一复现经典的卷积神经网络:LeNet、AlexNet、VGG系列、GooLeNet、ResNet 系列、DenseNet 系列,以及现在比较经典的目标检测网络、语义分割网络等。
Thanks for your share, i want to ask you that these code can direct run in gpu mode.
SSD,Yolo目标检测继续更新吗?
Hello,I found a performance issue in 7.%20ResNet/train.py
,
train_db = train_db.shuffle(50000).map was called without num_parallel_calls.
I think it will increase the efficiency of your program if you add this.
The same issues also exist in test_db = test_db.map ,
db = db.shuffle(1000).map,
db = db.shuffle(1000).map,
Here is the documemtation of tensorflow to support this thing.
Looking forward to your reply. Btw, I am very glad to create a PR to fix it if you are too busy.
hi, 我使用你的前两个案例进行训练,发现模型输出的损失和准确率都不变
一下输出是第2 个案例,2.快速搭建MNIST分类器/FastNet.py
Epoch 1/20
469/469 [==============================] - 2s 4ms/step - loss: 2.2167 - accuracy: 0.2436
Epoch 2/20
469/469 [==============================] - 2s 4ms/step - loss: 2.3567 - accuracy: 0.1044
Epoch 3/20
469/469 [==============================] - 2s 4ms/step - loss: 2.3567 - accuracy: 0.1044
Epoch 4/20
469/469 [==============================] - 2s 4ms/step - loss: 2.3567 - accuracy: 0.1044
Epoch 5/20
469/469 [==============================] - 2s 4ms/step - loss: 2.3567 - accuracy: 0.1044 - val_loss: 2.3584 - val_accuracy: 0.1028
Epoch 6/20
469/469 [==============================] - 2s 4ms/step - loss: 2.3567 - accuracy: 0.1044
Epoch 7/20
469/469 [==============================] - 2s 4ms/step - loss: 2.3567 - accuracy: 0.1044
Epoch 8/20
469/469 [==============================] - 2s 4ms/step - loss: 2.3567 - accuracy: 0.1044
Epoch 9/20
469/469 [==============================] - 2s 4ms/step - loss: 2.3567 - accuracy: 0.1044
Epoch 10/20
469/469 [==============================] - 2s 4ms/step - loss: 2.3567 - accuracy: 0.1044 - val_loss: 2.3584 - val_accuracy: 0.1028
Epoch 11/20
469/469 [==============================] - 2s 4ms/step - loss: 2.3567 - accuracy: 0.1044
Epoch 12/20
469/469 [==============================] - 2s 4ms/step - loss: 2.3567 - accuracy: 0.1044
Epoch 13/20
469/469 [==============================] - 2s 4ms/step - loss: 2.3567 - accuracy: 0.1044
Epoch 14/20
469/469 [==============================] - 2s 4ms/step - loss: 2.3567 - accuracy: 0.1044
Epoch 15/20
469/469 [==============================] - 2s 4ms/step - loss: 2.3567 - accuracy: 0.1044 - val_loss: 2.3584 - val_accuracy: 0.1028
Epoch 16/20
469/469 [==============================] - 2s 4ms/step - loss: 2.3567 - accuracy: 0.1044
Epoch 17/20
469/469 [==============================] - 2s 4ms/step - loss: 2.3567 - accuracy: 0.1044
Epoch 18/20
469/469 [==============================] - 2s 4ms/step - loss: 2.3567 - accuracy: 0.1044
Epoch 19/20
469/469 [==============================] - 2s 4ms/step - loss: 2.3567 - accuracy: 0.1044
Epoch 20/20
469/469 [==============================] - 2s 4ms/step - loss: 2.3567 - accuracy: 0.1044 - val_loss: 2.3584 - val_accuracy: 0.1028
Hello! I've found a performance issue in your project: batch()
should be called before map()
, which could make your program more efficient. Here is the tensorflow document to support it.
Detailed description is listed below:
.batch(128)
(here) should be called before .map(preprocess)
(here)..batch(128)
(here) should be called before .map(preprocess)
(here)..batch(32)
(here) should be called before .map(preprocess)
(here)..batch(16)
(here) should be called before .map(preprocess)
(here).Besides, you need to check the function called in map()
(e.g., preprocess
called in .map(preprocess)
) whether to be affected or not to make the changed code work properly. For example, if preprocess
needs data with shape (x, y, z) as its input before fix, it would require data with shape (batch_size, x, y, z).
Looking forward to your reply. Btw, I am very glad to create a PR to fix it if you are too busy.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.