https://github.com/t04glovern/stylegan-pokemon

Generating Pokemon cards using a mixture of StyleGAN and RNN to create beautiful & vibrant cards ready for battle!
https://github.com/t04glovern/stylegan-pokemon

pokemon rnn rnn-tensorflow stylegan stylegan-model tensorflow

Last synced: about 1 month ago
JSON representation

Generating Pokemon cards using a mixture of StyleGAN and RNN to create beautiful & vibrant cards ready for battle!

• Host: GitHub
• URL: https://github.com/t04glovern/stylegan-pokemon
• Owner: t04glovern
• Created: 2019-05-11T15:24:59.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2020-01-29T14:42:52.000Z (almost 5 years ago)
• Last Synced: 2024-04-21T17:16:34.856Z (9 months ago)
• Topics: pokemon, rnn, rnn-tensorflow, stylegan, stylegan-model, tensorflow
• Language: Python
• Homepage: https://devopstar.com/2019/05/21/stylegan-pokemon-card-generator/
• Size: 6.91 MB
• Stars: 50
• Watchers: 2
• Forks: 10
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Pokemon Card Generator

Generating Pokemon cards using a mixture of StyleGAN and RNN to create beautiful & vibrant cards ready for battle!

![Example](samples/pokemon.png)

Demo: [https://thesepokemondonotexist.com](https://thesepokemondonotexist.com)

## Run StyleGAN

### StyleGAN Environment

```bash

conda env create -f environment.yml

conda activate stylegan-pokemon

cd stylegan

```

#### StyleGAN Dataset

```bash

aws s3 cp s3://devopstar/resources/stylegan-pokemon/network-snapshot-007961.pkl network-snapshot-007961.pkl

aws s3 sync s3://devopstar/resources/stylegan-pokemon/kaggle-one-shot-pokemon kaggle-one-shot-pokemon

mkdir pokemon

```

#### Prepare Images

```bash

python prepare.py

python dataset_tool.py create_from_images datasets/smalls/ ./pokemon/

```

#### StyleGAN Training

```bash

python train.py

```

#### StyleGAN Invoke [Self-trained]

Using the `network-final.pkl` under the most recent entry in the [stylegan/results](stylegan/results) path

```bash

python invoke.py \

    --model_file './results/00000-sgan-custom-1gpu/network-final.pkl' \

    --output_file '../img/pokemon1.png'

```

#### StyleGAN Invoke [Pre-trained]

```bash

python invoke.py \

    --model_file './network-snapshot-007961.pkl' \

    --output_file '../img/pokemon2.png'

```

#### StyleGAN Invoke [MichaelFriese10](https://twitter.com/MichaelFriese10/status/1127614400750346240)

```bash

aws s3 cp s3://devopstar/resources/stylegan-pokemon/MichaelFriese10_pokemon.pkl MichaelFriese10_pokemon.pkl

python invoke.py \

    --model_file './MichaelFriese10_pokemon.pkl' \

    --output_file '../img/pokemon3.png'

```

## Run RNN

### RNN Environment

```bash

conda activate stylegan-pokemon

cd rnn

```

#### RNN Dataset

Dataset is from [armgilles/pokemon.csv](https://gist.github.com/armgilles/194bcff35001e7eb53a2a8b441e8b2c6)

```bash

ls -al data/pokemon/input.txt

```

Moves and Descriptions come from [https://pokemondb.net/move/all](https://pokemondb.net/move/all)

```bash

ls -al data/moves/input.txt

ls -al data/desc/input.txt

```

#### RNN Training

```bash

# Pokemon

python train.py \

    --data_dir=./data/pokemon \

    --save_dir=./save/pokemon \

    --seq_length=12

# Moves

python train.py \

    --data_dir=./data/moves \

    --save_dir=./save/moves \

    --rnn_size=300 \

    --seq_length=12

# Description

python train.py \

    --data_dir=./data/desc \

    --save_dir=./save/desc \

    --seq_length=25

```

#### RNN Invoke

```bash

# Pokemon

python sample.py \

    --save_dir=./save/pokemon \

    --output_dir=../samples/pokemon.txt

# Moves

python sample.py \

    --save_dir=./save/moves \

    --output_dir=../samples/moves.txt

# Description

python sample.py \

    --save_dir=./save/desc \

    --output_dir=../samples/desc.txt \

    -n=1500

```

### Cropping

```bash

# Define all fields adhoc

./cropper.sh \

    "Pokemon Name" \

    "Pokemon Attack 1" \

    "Pokemon Attack Description 1" \

    "Pokemon Attack 2" \

    "Pokemon Attack Description 2" \

    "img/pokemon1.png"

# Random

./cropper-shuffle.sh \

    "samples/pokemon.txt" \

    "samples/moves.txt" \

    "samples/desc.txt" \

    "img/pokemon3.png"

```

### Web Hosting

```bash

aws s3 mb s3://thesepokemondonotexist.com

aws s3 cp index.html s3://thesepokemondonotexist.com/index.html

```

## Attribution

* [One-Shot-Pokemon Images](https://www.kaggle.com/aaronyin/oneshotpokemon)

* [Inspired from Andrej Karpathy's char-rnn](https://github.com/karpathy/char-rnn)

* [Pokemon generated by neural network](https://aiweirdness.com/post/147834883707/pokemon-generated-by-neural-network)

* [MichaelFriese10 Pixel art model](https://twitter.com/MichaelFriese10/status/1127614400750346240)

* [Pokemon Blank Card Templates](https://www.deviantart.com/levelinfinitum/art/Pokemon-Blank-Card-Templates-Basic-474601445)