DeepCreamPy/docs/INSTALLATION.md
2022-09-22 00:29:44 +03:00

81 lines
3.1 KiB
Markdown

# Installation
## Download Prebuilt Binaries
You can download the latest release [here](https://github.com/deeppomf/DeepCreamPy/releases/latest) or find all previous releases [here](https://github.com/deeppomf/DeepCreamPy/releases).
Binary only available for Windows 64-bit.
## Run Code Yourself
If you want to run the code yourself, you can clone this repo and download the model from https://drive.google.com/open?id=1IMwzqZUuRnTv5jcuKdvZx-RZweknww5x. Unzip the file into the /models/ folder.
If you want access to older models, see https://drive.google.com/open?id=1_A0xFeJhrqpmulA6cC-a7RxJoQOD2RKm.
## running the code using Docker
Once the input images and model have been placed in `decensor_input` and `models` respectively,
the code can be run in the command line using docker (or podman), to avoid managing dependencies manually.
to build the container image use the command:
```
docker build -t deepcreampy .
```
then to desensor bar censors run the following command:
```
docker run --rm -v $(pwd)/models:/opt/DeepCreamPy/models -v $(pwd)/decensor_input:/opt/DeepCreamPy/decensor_input -v $(pwd)/decensor_output:/opt/DeepCreamPy/decensor_output deepcreampy
```
to desensor mosaics run the following command:
```
docker run --rm -v $(pwd)/models:/opt/DeepCreamPy/models -v $(pwd)/decensor_input:/opt/DeepCreamPy/decensor_input -v $(pwd)/decensor_input_original:/opt/DeepCreamPy/decensor_input_original -v $(pwd)/decensor_output:/opt/DeepCreamPy/decensor_output deepcreampy --is_mosaic=true
```
the contents of `decensor_input` and `decensor_input_original` are explained in the [decensoring tutorial](USAGE.md).
### Dependencies (for running the code yourself)
- Python 3.6.7
- TensorFlow 1.14
- Keras 2.2.4
- Pillow
- Scipy
- OpenCV
No GPU required! Tested on Ubuntu 16.04 and Windows. Tensorflow on Windows is compatible with Python 3 and not Python 2. Tensorflow is not compatible with Python 3.7.
Tensorflow, Keras, Pillow, and h5py can all be installed by running in the command line
```
$ pip install -r requirements.txt
```
## Run Code Yourself on CPUs that don't support AVX instructions
CPUs that don't support AVX instructions may experience this error when using the above install instructions:
```
ModuleNotFoundError: No module named '_pywrap_tensorflow_internal'
```
Follow these alternate install instructions if that happens:
1. Start from a clean Python 3.6.7 install.
2. Download a version of tensorflow that does not support AVX instructions from (https://github.com/fo40225/tensorflow-windows-wheel/tree/master/1.10.0/py36/CPU/sse2). I assume you picked tensorflow-1.10.0-cp36-cp36m-win_amd64.whl for 64-bit and the other for 32-bit computers.
3. Open the command line in the same directory as the file downloaded in step 2. Run
```
pip install tensorflow-1.10.0-cp36-cp36m-win_amd64.whl
```
or
```
pip install tensorflow-1.10.0-cp36-cp36m-win32.whl
```
depending on what you installed in step 2.
4. Open the command line in the directory of "DeepCreamPy-master" and run
```
pip install -r requirements.txt
```
Instructions are from https://github.com/deeppomf/DeepCreamPy/issues/26#issuecomment-434043166.