config.philipp.yaml


1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34

model: 
  n_cnn_layers: 3
  n_rnn_layers: 5
  rnn_dim: 512
  n_feats: 128 # number of mel features
  stride: 2
  dropout: 0.2 # recommended to be around 0.4-0.6 for smaller datasets, 0.1 for really large datasets

training:
  learning_rate: 0.0005
  batch_size: 32 # recommended to maximum number that fits on the GPU (batch size of 32 fits on a 12GB GPU)
  epochs: 150 
  eval_every_n: 5 # evaluate every n epochs
  num_workers: 4 # number of workers for dataloader
  device: "cuda" # device to run inference on if gpu is available, else "cpu" will be set automatically

dataset:
  download: true
  dataset_root_path: "data" # files will be downloaded into this dir
  language_name: "mls_german_opus"
  limited_supervision: false # set to True if you want to use limited supervision
  dataset_percentage: 1 # percentage of dataset to use (1.0 = 100%)
  shuffle: true

tokenizer:
  tokenizer_path: "data/tokenizers/char_tokenizer_german.json"

checkpoints:
  model_load_path: "data/runs/epoch31" # path to load model from
  model_save_path: "data/runs/epoch" # path to save model to

inference:
  model_load_path: "data/runs/epoch30" # path to load model from
  device: "cuda" # device to run inference on if gpu is available, else "cpu" will be set automatically