Skip to content

Add role token to context_tokens and response #8

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
krta2 wants to merge 4 commits into
base: add-print-every-batch
Choose a base branch
from
Open

Add role token to context_tokens and response #8

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
ea5e3f6
tokens added to context_tokens and response
gml9812 Nov 21, 2021
098379e
Add role tokens
krta2 Nov 23, 2021
41ffc49
Refactor adding items to dictionary
krta2 Nov 23, 2021
7f6b5ac
Update role token
krta2 Nov 23, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
17 changes: 17 additions & 0 deletions crslab/data/dataset/redial/redial.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -108,6 +108,14 @@ def _load_raw_data(self):

def _load_vocab(self):
self.tok2ind = json.load(open(os.path.join(self.dpath, 'token2id.json'), 'r', encoding='utf-8'))

# Add role tokens
# last_index = len(self.tok2ind)
# self.role_seeker_token_idx = last_index
# self.role_recommender_token_idx = last_index + 1
# self.tok2ind["__Seeker__"] = self.role_seeker_token_idx
# self.tok2ind["__Recommender__"] = self.role_recommender_token_idx

self.ind2tok = {idx: word for word, idx in self.tok2ind.items()}

logger.debug(f"[Load vocab from {os.path.join(self.dpath, 'token2id.json')}]")
Expand Down Expand Up @@ -185,6 +193,15 @@ def _augment_and_add(self, raw_conv_dict):
entity_set, word_set = set(), set()
for i, conv in enumerate(raw_conv_dict):
text_tokens, entities, movies, words = conv["text"], conv["entity"], conv["movie"], conv["word"]

# Add role token in front of text_tokens
self.role_seeker_token_idx = 2459 # "seeker"
self.role_recommender_token_idx = 1755 # "recommender"
if conv['role'] == 'Seeker':
text_tokens.insert(0, self.role_seeker_token_idx)
else:
text_tokens.insert(0, self.role_recommender_token_idx)

if len(context_tokens) > 0:
conv_dict = {
"role": conv['role'],
Expand Down