File size: 961 Bytes
bcdb559
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
log_dir: "Checkpoint_40b_Cotlet_Lite"
mixed_precision: no
data_folder: "/home/ubuntu/001_PLBERT_JA/40B_dataset_W_fixed"
batch_size: 26
save_interval: 10000
log_interval: 100
num_process: 1 # number of GPUs
num_steps: 1000000

dataset_params:
    tokenizer: ""
    token_separator: " " # token used for phoneme separator (space)
    token_mask: "M" # token used for phoneme mask (M)
    word_separator: 2 # token used for word separator (<formula>)
    token_maps: "/home/ubuntu/001_PLBERT_JA/token_maps_jp_200k_0.pkl" # token map path
    
    max_mel_length: 512 # max phoneme length
    
    word_mask_prob: 0.15 # probability to mask the entire word
    phoneme_mask_prob: 0.1 # probability to mask each phoneme
    replace_prob: 0.2 # probablity to replace phonemes
    
model_params:
    vocab_size: 178
    hidden_size: 768
    num_attention_heads: 12
    intermediate_size: 2048
    max_position_embeddings: 512
    num_hidden_layers: 12
    dropout: 0.1