Scrape quotes from Wisdom Quotes, specifying various options like string length:
$ python scrape.py --help
usage: scrape.py [-h] [-i INPUTFILE] [-o OUTPUTFILE] [--maxlen MAXLEN]
[--minlen MINLEN]
optional arguments:
-h, --help show this help message and exit
-i INPUTFILE, --inputfile INPUTFILE
Name of the input file that stores links, one link per
line
-o OUTPUTFILE, --outputfile OUTPUTFILE
Name of the file to store quotes
--maxlen MAXLEN Maximum length of the quotes stored
--minlen MINLEN Minimum length of the quotes stored
Train with scraped text:
$ python train.py
In train.py
, config entries are hardcoded. You will need to change them manually.
model_cfg = {
'word_level': True,
'rnn_size': 128,
'rnn_layers': 6,
'rnn_bidirectional': False,
'max_length': 3,
'max_words': 10000
}
train_cfg = {
'line_delimited': True,
'num_epochs': 10,
'gen_epochs': 10,
'train_size': 1,
'dropout': 0.3,
'validation': False,
'is_csv': False
}
Generate random quotes (make sure the configs are the same as those in train.py
:
$ python generatetext.py
Overlay "exactly" 3 lines of text over a generated image (use generated text or supply your own text file):
$ python imagegen.py
usage: imagegen.py [-h] [-x X] [-y Y] [-n NUM_NEURONS] [-l NUM_LAYERS]
[-c COUNT] [-f FILE]
optional arguments:
-h, --help show this help message and exit
-x X Width
-y Y Height
-n NUM_NEURONS, --num_neurons NUM_NEURONS
Number of neurons
-l NUM_LAYERS, --num_layers NUM_LAYERS
Number of layers
-c COUNT, --count COUNT
Number of images to be generated
-f FILE, --file FILE Name of the quotes file
- Implement with Instagram Graph API (Facebook needs to open the API to everyone)
- JSON config file
- Maybe a GUI