wellton/Emosic

Fork 0

T

Charlie Hewitt 579f192ad4 new trained + updated net

2017-12-16 09:47:22 +00:00

app/aff-song

remove write up etc.

2017-12-12 23:44:14 +00:00

data

update net

2017-12-14 13:58:35 +00:00

data_p

update net

2017-12-14 13:58:35 +00:00

icon.png

Add app

2017-12-10 16:34:39 +00:00

icon.psd

Add app

2017-12-10 16:34:39 +00:00

net.py

new trained + updated net

2017-12-16 09:47:22 +00:00

OUT_128D.h5

new trained + updated net

2017-12-16 09:47:22 +00:00

OUT_512D.h5

new trained + updated net

2017-12-16 09:47:22 +00:00

parse.py

first trained

2017-12-15 22:43:14 +00:00

README.md

Create README.md

2017-12-15 09:56:38 +00:00

TRAIN_128D.txt

new trained + updated net

2017-12-16 09:47:22 +00:00

training_labels.npy

update net + add dataset files

2017-12-14 10:12:18 +00:00

training_paths.npy

update net + add dataset files

2017-12-14 10:12:18 +00:00

validation_labels.npy

update net + add dataset files

2017-12-14 10:12:18 +00:00

validation_paths.npy

update net + add dataset files

2017-12-14 10:12:18 +00:00

README.md

Affectone

Song recommendations based on user affect using CNNs for emotion classification and valence/arousal regression. Predictions made using the Spotify Web API.

CNNs trained on AffectNet using Keras and deployed to iOS by conversion to coreML using coremltools.