Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/dark-art108/artistic-style-transfer-cnn
https://github.com/dark-art108/artistic-style-transfer-cnn
cnn-architecture colab-notebooks cuda pil vgg19
Last synced: 28 days ago
JSON representation
- Host: GitHub
- URL: https://github.com/dark-art108/artistic-style-transfer-cnn
- Owner: DARK-art108
- License: apache-2.0
- Created: 2020-01-24T17:13:19.000Z (about 5 years ago)
- Default Branch: master
- Last Pushed: 2020-01-24T17:47:04.000Z (about 5 years ago)
- Last Synced: 2024-11-12T20:18:35.315Z (3 months ago)
- Topics: cnn-architecture, colab-notebooks, cuda, pil, vgg19
- Language: Jupyter Notebook
- Size: 1.06 MB
- Stars: 2
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
#######REQUIREMENTS ########
RESEARCH PAPER:GATYS :A NEURAL ALGORITHM OF ARTISTIC STYLE
1.RECOMMENDED TO USE VGG-19 MODEL TO TRAIN THE CONVOLUTIONAL NEURAL NETWORK AS THERE ARE MULTIPLE IMAGE FILTERS AND MULTIPLE IMAGE MAX POOLING OVER THE CONTENTT IMAGE TO IMPOSE STYLE OVER THE CONTENT IMAGE BUT NOT DISTURBING THE CONTOURS OF THE IMAGES .
------------------------------------------------------------------------------------------------------------------------------------------------------
2. USE THE PIL PAKAGE (PYTHON IMAGE LIBRARY) INFACT IF YOU ARE NOT USING GOOGLE COLAB NOTEBOOK WITH CUDA GPU ENABLED YOU ALSO USE [PYCHARM] BY IMPORTING PILLOW PAKAGE IN PLACE OF PIL.
------------------------------------------------------------------------------------------------------------------------------------------------------------
3. USE COLAB NOTEBOOKS TO WITH GPU ENABLE AS TO RECONTRUCT STYLE OF THE IMAGE YOU SHOULD HAVE TO COMPUTE THE RELATION BETWEEN NUERONS USING GRAM MATRIX,SO TO DO STYLE IMPOSITION YOU SHOULD HAVE TO USE CUDA IN PLACE OF YOUR OWN CPU.
-----------------------------------------------------------------------------------------------------------------------------------------------
4. ANY VERSION OF TENSORFLOW OR PYTORCH (TORCH) CAN BE USED.
--------------------------------------------------------------------------------------------------------------------------------------
5. IMPORT MATPLOTLIB TO GET THE PROPER SCALING OF THE IMAGE.
------------------------------------------------------------------------------------------------------------------------------------------------
---------------------------------HOW TO USE SAMPLE CODE?------------------------------------------------------------------>>>
1...OPEN THE GOOGLE COLAB AND GO TO THE EDIT TAB SELECT NEW PYTHON 3 NOTEBOOK AND THE GO TO EDIT AND ENABLE THE GPU FOR YOUR PYTHON 3 COLAB NOTEBOOK ,]
2...RUN THE .IPYNB FILE.