[paths]
train = "assets/clausecat/train.spacy"
dev = "assets/clausecat/dev.spacy"
vectors = ${paths.ner_model}
init_tok2vec = "assets/pretrained_weights_clausecat.bin"
ner_model = "training/ner/config_tok2vec/model-best"

[system]
gpu_allocator = "pytorch"
seed = 0

[nlp]
lang = "en"
pipeline = ["sentencizer","tok2vec","ner","benepar","segmentation","clausecat","aggregation"]
batch_size = 128
disabled = []
before_creation = null
after_creation = null
after_pipeline_creation = null
tokenizer = {"@tokenizers":"spacy.Tokenizer.v1"}

[components]

[components.aggregation]
factory = "healthsea.aggregation.v1"

[components.benepar]
factory = "benepar"
disable_tagger = false
model = "benepar_en3"
subbatch_max_tokens = 500

[components.clausecat]
factory = "healthsea.clausecat.v1"
threshold = 0.5

[components.clausecat.model]
@architectures = "healthsea.clausecat_model.v1"
blinder = {"@layers":"healthsea.blinder.v1"}

[components.clausecat.model.textcat]
@architectures = "spacy.TextCatEnsemble.v2"
nO = null

[components.clausecat.model.textcat.linear_model]
@architectures = "spacy.TextCatBOW.v2"
exclusive_classes = true
ngram_size = 1
no_output_layer = false
nO = null

[components.clausecat.model.textcat.tok2vec]
@architectures = "spacy.Tok2Vec.v2"

[components.clausecat.model.textcat.tok2vec.embed]
@architectures = "spacy.MultiHashEmbed.v2"
width = 64
rows = [2000,2000,1000,1000,1000,1000]
attrs = ["ORTH","LOWER","PREFIX","SUFFIX","SHAPE","ID"]
include_static_vectors = false

[components.clausecat.model.textcat.tok2vec.encode]
@architectures = "spacy.MaxoutWindowEncoder.v2"
width = ${components.clausecat.model.textcat.tok2vec.embed.width}
window_size = 1
maxout_pieces = 3
depth = 2

[components.ner]
factory = "ner"
incorrect_spans_key = null
moves = null
scorer = {"@scorers":"spacy.ner_scorer.v1"}
update_with_oracle_cut_size = 100

[components.ner.model]
@architectures = "spacy.TransitionBasedParser.v2"
state_type = "ner"
extra_state_tokens = false
hidden_width = 64
maxout_pieces = 2
use_upper = false
nO = null

[components.ner.model.tok2vec]
@architectures = "spacy.Tok2VecListener.v1"
width = 256
upstream = "tok2vec"

[components.segmentation]
factory = "healthsea.segmentation.v1"

[components.sentencizer]
factory = "sentencizer"
overwrite = false
punct_chars = null
scorer = {"@scorers":"spacy.senter_scorer.v1"}

[components.tok2vec]
factory = "tok2vec"

[components.tok2vec.model]
@architectures = "spacy.Tok2Vec.v2"

[components.tok2vec.model.embed]
@architectures = "spacy.MultiHashEmbed.v2"
width = 256
attrs = ["ORTH","SHAPE"]
rows = [5000,2500]
include_static_vectors = true

[components.tok2vec.model.encode]
@architectures = "spacy.MaxoutWindowEncoder.v2"
width = 256
depth = 8
window_size = 1
maxout_pieces = 3

[corpora]

[corpora.dev]
@readers = "healthsea.clausecat_reader.v1"
path = ${paths.dev}

[corpora.train]
@readers = "healthsea.clausecat_reader.v1"
path = ${paths.train}

[training]
accumulate_gradient = 3
dev_corpus = "corpora.dev"
train_corpus = "corpora.train"
seed = ${system.seed}
gpu_allocator = ${system.gpu_allocator}
dropout = 0.1
patience = 1600
max_epochs = 0
max_steps = 20000
eval_frequency = 200
frozen_components = ["tok2vec","ner"]
annotating_components = []
before_to_disk = null

[training.batcher]
@batchers = "spacy.batch_by_words.v1"
discard_oversize = false
tolerance = 0.2
get_length = null

[training.batcher.size]
@schedules = "compounding.v1"
start = 100
stop = 1000
compound = 1.001
t = 0.0

[training.logger]
@loggers = "spacy.ConsoleLogger.v1"
progress_bar = true

[training.optimizer]
@optimizers = "Adam.v1"
beta1 = 0.9
beta2 = 0.999
L2_is_weight_decay = true
L2 = 0.01
grad_clip = 1.0
use_averages = false
eps = 0.00000001

[training.optimizer.learn_rate]
@schedules = "warmup_linear.v1"
warmup_steps = 250
total_steps = 20000
initial_rate = 0.00005

[training.score_weights]
sents_f = null
sents_p = null
sents_r = null
ents_f = null
ents_p = null
ents_r = null
ents_per_type = null
cats_score = 1.0
cats_score_desc = null
cats_micro_p = null
cats_micro_r = null
cats_micro_f = null
cats_macro_p = null
cats_macro_r = null
cats_macro_f = null
cats_macro_auc = null
cats_f_per_type = null
cats_macro_auc_per_type = null

[pretraining]
max_epochs = 100
dropout = 0.2
n_save_every = null
n_save_epoch = 1
component = "clausecat"
layer = "tok2vec"
corpus = "corpora.pretrain"

[pretraining.batcher]
@batchers = "spacy.batch_by_words.v1"
size = 10000
discard_oversize = false
tolerance = 0.2
get_length = null

[pretraining.objective]
@architectures = "spacy.PretrainVectors.v1"
maxout_pieces = 3
hidden_size = 300
loss = "cosine"

[pretraining.optimizer]
@optimizers = "Adam.v1"
beta1 = 0.9
beta2 = 0.999
L2_is_weight_decay = true
L2 = 0.01
grad_clip = 1.0
use_averages = true
eps = 0.00000001
learn_rate = 0.001

[initialize]
vectors = ${paths.vectors}
init_tok2vec = ${paths.init_tok2vec}
vocab_data = null
lookups = null
before_init = null
after_init = null

[initialize.components]

[initialize.tokenizer]