Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/AIS-SNU/Smart-Infinity
/ tokenize
Method
tokenize
DeepSpeedExample/tools/preprocess_data.py:52–53 ·
view source on GitHub ↗
(self, *text)
Source
from the content-addressed store, hash-verified
50
51
class
IdentitySplitter(object):
52
def
tokenize(self, *text):
53
return
text
54
55
class
Encoder(object):
56
def
__init__(self, args):
Callers
5
encode
Method · 0.45
build_tokens_types_paddings_from_text
Function · 0.45
process_single_datapath
Function · 0.45
get_tokens
Method · 0.45
_build_wikitext103_dataset
Function · 0.45
Calls
no outgoing calls
Tested by
no test coverage detected