forked from mrq/DL-Art-School
42 lines
523 B
Plaintext
42 lines
523 B
Plaintext
# Fundamentals
|
|
numpy
|
|
pyyaml
|
|
tb-nightly
|
|
future
|
|
scp
|
|
tqdm
|
|
matplotlib
|
|
scipy
|
|
munch
|
|
tqdm
|
|
scp
|
|
tensorboard
|
|
orjson
|
|
einops
|
|
lambda-networks
|
|
|
|
# For image generation stuff
|
|
opencv-python
|
|
kornia
|
|
pytorch_ssim
|
|
gsa-pytorch
|
|
pytorch_fid==0.1.1
|
|
|
|
# For audio generation stuff
|
|
inflect==0.2.5
|
|
librosa==0.6.0
|
|
Unidecode==1.0.22
|
|
tgt == 1.4.4
|
|
pyworld == 0.2.10
|
|
audio2numpy
|
|
|
|
# For text stuff
|
|
transformers
|
|
tokenizers
|
|
|
|
# lucidrains stuff
|
|
vector_quantize_pytorch
|
|
linear_attention_transformer
|
|
rotary-embedding-torch
|
|
axial_positional_embedding
|
|
g-mlp-pytorch |