Rooba
/
spaCy
mirror of https://github.com/explosion/spaCy.git
1
1
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Update config

This commit is contained in:
Matthew Honnibal 2020-07-29 14:26:44 +02:00
parent 105cf29967
commit b5bbfec591
1 changed files with 44 additions and 39 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

83
examples/experiments/onto-joint/defaults.cfg
View File

@ -20,20 +20,20 @@ seed = 0
accumulate_gradient = 1 accumulate_gradient = 1
use_pytorch_for_gpu_memory = false use_pytorch_for_gpu_memory = false
# Control how scores are printed and checkpoints are evaluated. # Control how scores are printed and checkpoints are evaluated.
scores = ["speed", "tags_acc", "uas", "las", "ents_f"] eval_batch_size = 128
score_weights = {"las": 0.4, "ents_f": 0.4, "tags_acc": 0.2} score_weights = {"las": 0.4, "ents_f": 0.4, "tags_acc": 0.2}
# These settings are invalid for the transformer models.
init_tok2vec = null init_tok2vec = null
discard_oversize = false discard_oversize = false
omit_extra_lookups = false
batch_by = "words" batch_by = "words"
use_gpu = -1
raw_text = null raw_text = null
tag_map = null tag_map = null
vectors = null
base_model = null
morph_rules = null
[training.batch_size] [training.batch_size]
@schedules = "compounding.v1" @schedules = "compounding.v1"
start = 1000 start = 100
stop = 1000 stop = 1000
compound = 1.001 compound = 1.001
@ -46,74 +46,79 @@ L2 = 0.01
grad_clip = 1.0 grad_clip = 1.0
use_averages = false use_averages = false
eps = 1e-8 eps = 1e-8
#learn_rate = 0.001 learn_rate = 0.001
[training.optimizer.learn_rate]
@schedules = "warmup_linear.v1"
warmup_steps = 250
total_steps = 20000
initial_rate = 0.001
[nlp] [nlp]
lang = "en" lang = "en"
base_model = null load_vocab_data = false
vectors = null pipeline = ["tok2vec", "ner", "tagger", "parser"]
[nlp.pipeline] [nlp.tokenizer]
@tokenizers = "spacy.Tokenizer.v1"
[nlp.pipeline.tok2vec] [nlp.lemmatizer]
@lemmatizers = "spacy.Lemmatizer.v1"
[components]
[components.tok2vec]
factory = "tok2vec" factory = "tok2vec"
[components.ner]
[nlp.pipeline.ner]
factory = "ner" factory = "ner"
learn_tokens = false learn_tokens = false
min_action_freq = 1 min_action_freq = 1
[nlp.pipeline.tagger] [components.tagger]
factory = "tagger" factory = "tagger"
[nlp.pipeline.parser] [components.parser]
factory = "parser" factory = "parser"
learn_tokens = false learn_tokens = false
min_action_freq = 30 min_action_freq = 30
[nlp.pipeline.tagger.model] [components.tagger.model]
@architectures = "spacy.Tagger.v1" @architectures = "spacy.Tagger.v1"
[nlp.pipeline.tagger.model.tok2vec] [components.tagger.model.tok2vec]
@architectures = "spacy.Tok2VecTensors.v1" @architectures = "spacy.Tok2VecListener.v1"
width = ${nlp.pipeline.tok2vec.model:width} width = ${components.tok2vec.model.encode:width}
[nlp.pipeline.parser.model] [components.parser.model]
@architectures = "spacy.TransitionBasedParser.v1" @architectures = "spacy.TransitionBasedParser.v1"
nr_feature_tokens = 8 nr_feature_tokens = 8
hidden_width = 128 hidden_width = 128
maxout_pieces = 2 maxout_pieces = 2
use_upper = true use_upper = true
[nlp.pipeline.parser.model.tok2vec] [components.parser.model.tok2vec]
@architectures = "spacy.Tok2VecTensors.v1" @architectures = "spacy.Tok2VecListener.v1"
width = ${nlp.pipeline.tok2vec.model:width} width = ${components.tok2vec.model.encode:width}
[nlp.pipeline.ner.model] [components.ner.model]
@architectures = "spacy.TransitionBasedParser.v1" @architectures = "spacy.TransitionBasedParser.v1"
nr_feature_tokens = 3 nr_feature_tokens = 3
hidden_width = 128 hidden_width = 128
maxout_pieces = 2 maxout_pieces = 2
use_upper = true use_upper = true
[nlp.pipeline.ner.model.tok2vec] [components.ner.model.tok2vec]
@architectures = "spacy.Tok2VecTensors.v1" @architectures = "spacy.Tok2VecListener.v1"
width = ${nlp.pipeline.tok2vec.model:width} width = ${components.tok2vec.model.encode:width}
[nlp.pipeline.tok2vec.model] [components.tok2vec.model]
@architectures = "spacy.HashEmbedCNN.v1" @architectures = "spacy.Tok2Vec.v1"
pretrained_vectors = ${nlp:vectors}
width = 128 [components.tok2vec.model.embed]
@architectures = "spacy.MultiHashEmbed.v1"
width = ${components.tok2vec.model.encode:width}
rows = 2000
also_embed_subwords = true
also_use_static_vectors = false
[components.tok2vec.model.encode]
@architectures = "spacy.MaxoutWindowEncoder.v1"
width = 96
depth = 4 depth = 4
window_size = 1 window_size = 1
embed_size = 7000
maxout_pieces = 3 maxout_pieces = 3
subword_features = true
dropout = ${training:dropout}