# wmt14_en_de.yaml
save_data: data/wmt/run/example

# Corpus opts:
data:
    commoncrawl:
        path_src: data/wmt/commoncrawl.de-en.en
        path_tgt: data/wmt/commoncrawl.de-en.de
        transforms: [sentencepiece, filtertoolong]
        weight: 23
    europarl:
        path_src: data/wmt/europarl-v7.de-en.en
        path_tgt: data/wmt/europarl-v7.de-en.de
        transforms: [sentencepiece, filtertoolong]
        weight: 19
    news_commentary:
        path_src: data/wmt/news-commentary-v11.de-en.en
        path_tgt: data/wmt/news-commentary-v11.de-en.de
        transforms: [sentencepiece, filtertoolong]
        weight: 3
    valid:
        path_src: data/wmt/valid.en
        path_tgt: data/wmt/valid.de
        transforms: [sentencepiece]

### Transform related opts:
#### Subword
src_subword_model: data/wmt/wmtende.model
tgt_subword_model: data/wmt/wmtende.model
# src_subword_type: sentencepiece
# tgt_subword_type: sentencepiece
# onmttok_kwargs: "{'mode': 'none', 'spacer_annotate': True}"

subword_nbest: 1
subword_alpha: 0.0
#### Filter
src_seq_length: 150
tgt_seq_length: 150

# silently ignore empty lines in the data
skip_empty_level: silent


# # Vocab opts
# ### vocab:
src_vocab: data/wmt/run/example.vocab.src
tgt_vocab: data/wmt/run/example.vocab.tgt
src_vocab_size: 32000
tgt_vocab_size: 32000
vocab_size_multiple: 8
src_words_min_frequency: 1
tgt_words_min_frequency: 1
share_vocab: True

# # Model training parameters

# General opts
save_model: data/wmt/run/model
keep_checkpoint: 50
save_checkpoint_steps: 5000
average_decay: 0.0005
seed: 1234
report_every: 100
train_steps: 100000
valid_steps: 5000

# Batching
bucket_size: 32768
world_size: 2
gpu_ranks: [0, 1]
num_workers: 4
batch_type: "tokens"
batch_size: 4096
valid_batch_size: 2048
batch_size_multiple: 8
max_generator_batches: 0
accum_count: [3]
accum_steps: [0]

# Optimization
model_dtype: "fp32"
optim: "adam"
learning_rate: 2
warmup_steps: 6000
decay_method: "noam"
adam_beta2: 0.998
max_grad_norm: 0
label_smoothing: 0.1
param_init: 0
param_init_glorot: true
normalization: "tokens"

# Model
encoder_type: transformer
decoder_type: transformer
enc_layers: 6
dec_layers: 6
heads: 8
hidden_size: 512
word_vec_size: 512
transformer_ff: 2048
dropout_steps: [0]
dropout: [0.1]
attention_dropout: [0.1]
share_decoder_embeddings: true
share_embeddings: true
position_encoding: true