aidiary / keras-examples Goto Github PK
View Code? Open in Web Editor NEWKerasのサンプルプログラム
License: MIT License
Kerasのサンプルプログラム
License: MIT License
sudo apt-get update
sudo apt-get install emacs gcc
sudo apt-get install -y python-qt4
wget https://repo.continuum.io/archive/Anaconda3-4.2.0-Linux-x86_64.sh
pip install pydot-ng
conda install graphviz
conda install tensorflow
conda install keras
http://machinelearningmastery.com/check-point-deep-learning-models-keras/
http://www.world-of-lucid-dreaming.com/inside-robot-dreams-what-the-google-ai-bots-think-about.html
https://github.com/danielvarga/keras-deep-dream/blob/master/deep_dream.py
https://github.com/google/deepdream/blob/master/dream.ipynb
https://github.com/danielvarga/keras-deep-dream/blob/master/deep_dream.py
Region-CNN
物体検出
http://cs231n.stanford.edu/slides/winter1516_lecture8.pdf
https://www.researchgate.net/post/In_Keras_How_can_I_extract_the_exact_location_of_the_detected_object_or_objects_within_image_that_includes_a_background
r-cnn
fast r-cnn
faster r-cnn
In finetuning.py of 17Flowers how you are performing fine tuning without loading a pretrained weights?
As I can see you commented out the following line (Line no. 51):
# top_model.load_weights(os.path.join(result_dir, 'bottleneck_fc_model.h5'))
Whereas in finetuning.py of dogs_vs_cats you are executing that line. Is it possible to fine tune without loading a pretrained model?
ねこと画像処理 part 3 – Deep Learningで猫の品種識別
http://www.robots.ox.ac.uk/~vgg/data/pets/
https://github.com/fchollet/keras/blob/master/examples/neural_style_transfer.py
https://elix-tech.github.io/ja/2016/08/22/art.html
各手法について変換前・変換後を見比べる
データ拡張を使うと精度が向上するか確認する
フィルタの可視化
https://github.com/fchollet/keras/blob/master/examples/conv_filter_visualization.py
https://blog.keras.io/how-convolutional-neural-networks-see-the-world.html
https://github.com/raghakot/keras-vis
https://github.com/jacobgil/keras-filter-visualization
https://jacobgil.github.io/deeplearning/filter-visualizations
https://blog.keras.io/how-convolutional-neural-networks-see-the-world.html
http://ankivil.com/visualizing-deep-neural-networks-classes-and-features/
http://yosinski.com/deepvis
https://blog.keras.io/how-convolutional-neural-networks-see-the-world.html
https://jacobgil.github.io/deeplearning/filter-visualizations
http://yosinski.com/deepvis
指定したフィルタの活性化(出力?)を最大化するような入力画像を生成する
繰り返しgradを足し込む?
data gradient
http://stackoverflow.com/questions/38135950/meaning-of-weight-gradient-in-cnn
http://yosinski.com/deepvis
http://qiita.com/knao124/items/fdb47674ada389e70c6e
http://ankivil.com/visualizing-deep-neural-networks-classes-and-features/
CNNで解いてみる
17 Category Flower Dataset
http://www.robots.ox.ac.uk/~vgg/data/flowers/17/
102 Category Flower Dataset
http://www.robots.ox.ac.uk/~vgg/data/flowers/102/
スクラッチから学習したモデル
VGG16から転移学習させたモデル
を比較したい
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.