- 전체
- Sample DB
- database modeling
- [표준 SQL] Standard SQL
- G-SQL
- 10-Min
- ORACLE
- MS SQLserver
- MySQL
- SQLite
- postgreSQL
- 데이터아키텍처전문가 - 국가공인자격
- 데이터 분석 전문가 [ADP]
- [국가공인] SQL 개발자/전문가
- NoSQL
- hadoop
- hadoop eco system
- big data (빅데이터)
- stat(통계) R 언어
- XML DB & XQuery
- spark
- DataBase Tool
- 데이터분석 & 데이터사이언스
- Engineer Quality Management
- [기계학습] machine learning
- 데이터 수집 및 전처리
- 국가기술자격 빅데이터분석기사
- 암호화폐 (비트코인, cryptocurrency, bitcoin)
[기계학습] machine learning Get Started With Keras For Beginners
2018.07.11 18:28
Get Started With Keras For Beginners
Sep 8, 2015
I'm trying researching on deep learning, and I decided to use Keras, which runs with very simple code. I installed Keras, its requirements with miniconda on python 2.x. miniconda is for very simple dependency management, and python 2.x is for compatibility for some library (not happened, but just in case).
Keras is a minimalist, highly modular neural network library ..., that uses Theano ...
at http://keras.io/.
Update
- 2016-06-22: I wrote new post for Keras with TensorFlow. Click here.
Environment
$ uname -a
Linux 2a49ce30b6a8 3.16.0-30-generic #40~14.04.1-Ubuntu SMP Thu Jan 15 17:43:14 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux
Install miniconda and python 2.x
See also here.
$ wget https://repo.continuum.io/miniconda/Miniconda-latest-Linux-`uname -p`.sh
$ bash Miniconda-latest-Linux-`uname -p`.sh -b
$ rm Miniconda-latest-Linux-`uname -p`.sh
-b option makes script run with default setting.
Reopen shell or run $ . .bashrc for setting up the environment $PATH.
Install Theano
$ conda install -y theano
Conda does not ask for confirmation with -y option.
Install Keras
Conda cannot install Keras, so install requirements of Keras first, and install Keras with pip.
$ conda install -y h5py
$ pip install keras
Install build essential (optional)
$ sudo apt-get install -y build-essential
Without g++, you will get this message when run Keras and Keras will be very slow.
WARNING (theano.configdefaults): g++ not detected ! Theano will be unable to exe
cute optimized C-implementations (for both CPU and GPU) and will default to Pyth
on implementations. Performance will be severely degraded. To remove this warnin
g, set Theano flags cxx to an empty string.
Test Keras
$ curl -sSL https://github.com/fchollet/keras/raw/master/examples/mnist_mlp.py | python
And check the output.
60000 train samples
10000 test samples
Train on 60000 samples, validate on 10000 samples
Epoch 0
5s - loss: 0.4356 - acc: 0.8716 - val_loss: 0.1863 - val_acc: 0.9421
Epoch 1
5s - loss: 0.1961 - acc: 0.9414 - val_loss: 0.1274 - val_acc: 0.9601
Epoch 2
5s - loss: 0.1451 - acc: 0.9564 - val_loss: 0.1010 - val_acc: 0.9691
Epoch 3
5s - loss: 0.1189 - acc: 0.9642 - val_loss: 0.0847 - val_acc: 0.9752
Epoch 4
5s - loss: 0.1019 - acc: 0.9691 - val_loss: 0.0850 - val_acc: 0.9735
Epoch 5
5s - loss: 0.0903 - acc: 0.9721 - val_loss: 0.0749 - val_acc: 0.9777
Epoch 6
5s - loss: 0.0822 - acc: 0.9745 - val_loss: 0.0753 - val_acc: 0.9762
Epoch 7
5s - loss: 0.0758 - acc: 0.9762 - val_loss: 0.0743 - val_acc: 0.9796
Epoch 8
5s - loss: 0.0705 - acc: 0.9780 - val_loss: 0.0720 - val_acc: 0.9784
Epoch 9
5s - loss: 0.0648 - acc: 0.9790 - val_loss: 0.0688 - val_acc: 0.9793
Epoch 10
5s - loss: 0.0592 - acc: 0.9819 - val_loss: 0.0663 - val_acc: 0.9797
Epoch 11
5s - loss: 0.0567 - acc: 0.9824 - val_loss: 0.0677 - val_acc: 0.9815
Epoch 12
5s - loss: 0.0536 - acc: 0.9833 - val_loss: 0.0711 - val_acc: 0.9796
Epoch 13
5s - loss: 0.0520 - acc: 0.9834 - val_loss: 0.0684 - val_acc: 0.9806
Epoch 14
5s - loss: 0.0500 - acc: 0.9837 - val_loss: 0.0664 - val_acc: 0.9807
Epoch 15
5s - loss: 0.0471 - acc: 0.9850 - val_loss: 0.0683 - val_acc: 0.9809
Epoch 16
5s - loss: 0.0449 - acc: 0.9856 - val_loss: 0.0682 - val_acc: 0.9812
Epoch 17
5s - loss: 0.0433 - acc: 0.9860 - val_loss: 0.0675 - val_acc: 0.9813
Epoch 18
5s - loss: 0.0401 - acc: 0.9869 - val_loss: 0.0683 - val_acc: 0.9819
Epoch 19
5s - loss: 0.0383 - acc: 0.9874 - val_loss: 0.0705 - val_acc: 0.9820
Test score: 0.0704572771238
Test accuracy: 0.982
[출처] http://ermaker.github.io/blog/2015/09/08/get-started-with-keras-for-beginners.html
광고 클릭에서 발생하는 수익금은 모두 웹사이트 서버의 유지 및 관리, 그리고 기술 콘텐츠 향상을 위해 쓰여집니다.
