[pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
This commit is contained in:
pre-commit-ci[bot] 2025-03-17 17:25:03 +00:00
parent 624b5978bb
commit 6f61be451f
10 changed files with 15 additions and 15 deletions

View File

@ -81,11 +81,11 @@ with gr.Blocks(css=CSS) as demo:
) )
with gr.Row(): with gr.Row():
btn = gr.UploadButton("📁", file_types=["file"], file_count="multiple", size="sm") btn = gr.UploadButton("📁", file_types=["file"], file_count="multiple", size="sm")
restart_btn = gr.Button(str("\u21BB"), elem_id="restart-btn", scale=1) restart_btn = gr.Button(str("\u21bb"), elem_id="restart-btn", scale=1)
txt = gr.Textbox( txt = gr.Textbox(
scale=8, scale=8,
show_label=False, show_label=False,
placeholder="Enter text and press enter, or use 📁 to upload files, click \u21BB to clear loaded files and restart chat", placeholder="Enter text and press enter, or use 📁 to upload files, click \u21bb to clear loaded files and restart chat",
container=True, container=True,
autofocus=True, autofocus=True,
) )

View File

@ -34,8 +34,8 @@ class PreTrainingDataset:
self.do_whole_word_mask = do_whole_word_mask self.do_whole_word_mask = do_whole_word_mask
self.max_predictions_per_seq = max_predictions_per_seq self.max_predictions_per_seq = max_predictions_per_seq
self.vocab_words = list(tokenizer.vocab.keys()) self.vocab_words = list(tokenizer.vocab.keys())
self.rec = re.compile("[\u4E00-\u9FA5]") self.rec = re.compile("[\u4e00-\u9fa5]")
self.whole_rec = re.compile("##[\u4E00-\u9FA5]") self.whole_rec = re.compile("##[\u4e00-\u9fa5]")
self.mlm_p = 0.15 self.mlm_p = 0.15
self.mlm_mask_p = 0.8 self.mlm_mask_p = 0.8