explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2024-11-15 06:09:01 +03:00
Code Issues Packages Projects Releases Wiki Activity

Add more spacy pretrain options

Browse Source
This commit is contained in:
Matthw Honnibal 2019-10-18 17:24:13 +02:00
parent f3e2aaea1e
commit 36de9bf72a
1 changed files with 7 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
spacy/cli/pretrain.py
View File

@ -35,6 +35,8 @@ from .train import _load_pretrained_tok2vec
output_dir=("Directory to write models to on each epoch", "positional", None, str), output_dir=("Directory to write models to on each epoch", "positional", None, str),
width=("Width of CNN layers", "option", "cw", int), width=("Width of CNN layers", "option", "cw", int),
depth=("Depth of CNN layers", "option", "cd", int), depth=("Depth of CNN layers", "option", "cd", int),
use_chars=("Whether to use character-based embedding", "flag", "chr", bool),
sa_depth=("Depth of self-attention layers", "option", "sa", int),
bilstm_depth=("Depth of BiLSTM layers (requires PyTorch)", "option", "lstm", int), bilstm_depth=("Depth of BiLSTM layers (requires PyTorch)", "option", "lstm", int),
embed_rows=("Number of embedding rows", "option", "er", int), embed_rows=("Number of embedding rows", "option", "er", int),
loss_func=( loss_func=(
@ -82,6 +84,8 @@ def pretrain(
width=96, width=96,
depth=4, depth=4,
bilstm_depth=0, bilstm_depth=0,
sa_depth=0,
use_chars=False,
embed_rows=2000, embed_rows=2000,
loss_func="cosine", loss_func="cosine",
use_vectors=False, use_vectors=False,
@ -157,9 +161,11 @@ def pretrain(
embed_rows, embed_rows,
conv_depth=depth, conv_depth=depth,
pretrained_vectors=pretrained_vectors, pretrained_vectors=pretrained_vectors,
char_embed=use_chars,
self_attn_depth=sa_depth, # Experimental.
bilstm_depth=bilstm_depth, # Requires PyTorch. Experimental. bilstm_depth=bilstm_depth, # Requires PyTorch. Experimental.
cnn_maxout_pieces=3, # You can try setting this higher cnn_maxout_pieces=3, # You can try setting this higher
subword_features=True, # Set to False for Chinese etc subword_features=not use_chars, # Set to False for Chinese etc
), ),
) )
# Load in pretrained weights # Load in pretrained weights