Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in
Toggle navigation
Menu
Open sidebar
chenpangpang
transformers
Commits
3e20c2e8
Commit
3e20c2e8
authored
Nov 12, 2019
by
Louis MARTIN
Committed by
Julien Chaumond
Nov 16, 2019
Browse files
Update demo_camembert.py with new classes
parent
f12e4d8d
Changes
1
Show whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
3 additions
and
14 deletions
+3
-14
examples/contrib/demo_camembert.py
examples/contrib/demo_camembert.py
+3
-14
No files found.
examples/contrib/demo_camembert.py
View file @
3e20c2e8
...
@@ -5,7 +5,7 @@ import urllib.request
...
@@ -5,7 +5,7 @@ import urllib.request
import
torch
import
torch
from
transformers.tokenization_camembert
import
CamembertTokenizer
from
transformers.tokenization_camembert
import
CamembertTokenizer
from
transformers.modeling_
ro
bert
a
import
Ro
bert
a
ForMaskedLM
from
transformers.modeling_
camem
bert
import
Camem
bertForMaskedLM
def
fill_mask
(
masked_input
,
model
,
tokenizer
,
topk
=
5
):
def
fill_mask
(
masked_input
,
model
,
tokenizer
,
topk
=
5
):
...
@@ -40,19 +40,8 @@ def fill_mask(masked_input, model, tokenizer, topk=5):
...
@@ -40,19 +40,8 @@ def fill_mask(masked_input, model, tokenizer, topk=5):
return
topk_filled_outputs
return
topk_filled_outputs
model_path
=
Path
(
'camembert.v0.pytorch'
)
tokenizer
=
CamembertTokenizer
.
from_pretrained
(
'camembert-base'
)
if
not
model_path
.
exists
():
model
=
CamembertForMaskedLM
.
from_pretrained
(
'camembert-base'
)
compressed_path
=
model_path
.
with_suffix
(
'.tar.gz'
)
url
=
'http://dl.fbaipublicfiles.com/camembert/camembert.v0.pytorch.tar.gz'
print
(
'Downloading model...'
)
urllib
.
request
.
urlretrieve
(
url
,
compressed_path
)
print
(
'Extracting model...'
)
with
tarfile
.
open
(
compressed_path
)
as
f
:
f
.
extractall
(
model_path
.
parent
)
assert
model_path
.
exists
()
tokenizer_path
=
model_path
/
'sentencepiece.bpe.model'
tokenizer
=
CamembertTokenizer
.
from_pretrained
(
tokenizer_path
)
model
=
RobertaForMaskedLM
.
from_pretrained
(
model_path
)
model
.
eval
()
model
.
eval
()
masked_input
=
"Le camembert est <mask> :)"
masked_input
=
"Le camembert est <mask> :)"
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment