Skip to content

Latest commit

 

History

History

compressed_sensing

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
datasets/sources
datasets/sources
 
 
README.md
README.md
 
 
TODOs.md
TODOs.md
 
 
data_exploration.py
data_exploration.py
 
 
main.py
main.py
 
 
requirements.txt
requirements.txt
 
 
tf_network.py
tf_network.py
 
 
tf_py_func_test
tf_py_func_test
 
 
transform_exploration.py
transform_exploration.py
 
 
utils.py
utils.py
 
 

README.md

learn_cs_param

Installation with docker

TODO

Manual installation

Warning: TensorFlow does not support Windows yet.

  1. Install Python 3.x * Linux: Use your package manager to install Python 3.5 (or newer) * Mac: Either via the Homebrew package manager using brew install python3 or via Anaconda * Windows: Use Anaconda

  2. Clone the repository * ssh:

```bash
git clone [email protected]:dperdios/learn_cs_param.git
cd learn_cs_param
```

* https:

```bash
git clone https://github.com/dperdios/learn_cs_param.git
cd learn_cs_param
```
  1. (optional) Create a dedicated environment. The code have been developed on Python 3.5 * Using Anaconda:
```bash
conda create -n learn_cs_param python=3.5
```

* Using `pyenv`:

```bash
pyvenv /path/to/new/virtual/env . /path/to/new/virtual/env/bin/activate
```
  1. Install the dependencies from requirements.txt. Please edit the TensorFlow version (CPU/GPU, Linux/OSX) you want to install. The code has been developed with TensorFlow 0.11. * Depending on your installation, pip may refer to Python 2 (you can verify with pip -V). In that case, use pip3 instead of pip.
pip install --upgrapde pip
pip install -r requirements.txt

  1. TODO Check you have a working installation

  2. Launch Jupyter notebook.

jupyter notebook

License

TODO