Differences

Writing /home/fac/arijit/public_html/dokuwiki/data/cache/0/0f3dae15bd579c4b78137f3d0ad7d105.xhtml failed

This shows you the differences between two versions of the page.

--- courses:2017:installation_guide [2017/02/11 15:04]
arijit created
+++ courses:2017:installation_guide [2017/02/12 11:56] (current)
arijit
@@ Line 1: / Line 1: @@
 ======Installation guide for Keras======
+<file>
+Step 1. An easy way to install most of the dependencies is to install Anaconda.
+		- Download page https://www.continuum.io/downloads
+			--install
+					bash Anaconda2-4.2.0-Linux-x86_64.sh
+			--allow set up path at the end of installation
+Step 2. Check PATH variable echo $PATH.
+		- If not set, then set the path in .bashrc file.
+				--export PATH="/home/niraj/anaconda2/bin:$PATH"
+		- Restart terminal and check again.
-  - Installing anaconda from link http://deeplearning.net/software/theano/install.html#compiler-osx which points to anaconda download page https://www.continuum.io/downloads
-    - download done
+Step 3. Installing theano using pip command
-    - installing using bash command
+		- If pip not installed, install it.
-      - allow set up path
+			-- https://pip.pypa.io/en/stable/installing/
-  - Check PATH variable echo $PATH. If not set, then set the path in .bashrc file. Restart terminal and check again.
+		- pip install Theano
-  - Installing theano using pip command [ref to page http://deeplearning.net/software/theano/install.html#compiler-osx]
-  - Install same page next dependency
+Step 4. install additional dependency
-  - Install nose-parameterized ''pip install --user nose-parameterized''
+		-conda install pydot-ng
-  - Running test ''python -c "import theano; theano.test()"''
-    - Ok (about 30min)
+Step 5. install nose-parameterized
-  - Testing installation [ref to page https://elitedatascience.com/keras-tutorial-deep-learning-in-python]
+		-pip install --user nose-parameterized
-  - Installing keras using pip [ref to page https://elitedatascience.com/keras-tutorial-deep-learning-in-python]
-  - Very immportant editing .json file   <code>
+Step 6. running test
-  ~/.keras/keras.json.
+		-python -c "import theano; theano.test()"
-  {
+			--ok(about 30min)
-    "image_dim_ordering": "th",
-    "epsilon": 1e-07,
+Step 7. installing keras using pip
-    "floatx": "float32",
+		-pip install keras
-    "backend": "theano"
-  }
+Step 8. edit .json file
-  </code>
-  - Check keras installation
+		~/.keras/keras.json.
-    - Update if required
+		{
+		"image_dim_ordering": "th",
+		"epsilon": 1e-07,
+		"floatx": "float32",
+		"backend": "theano"
+		}
+Step 9. check keras installation
+		-python -c "import keras; print keras.__version__"
+</file>