similar-text

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+jwiki_w2v.kv filter=lfs diff=lfs merge=lfs -text

app.py

@@ -0,0 +1,35 @@
+import gradio as gr
+import gensim
+
+def process_text(input_text):
+	if input_text is None or input_text == '':
+		yield [None, None]
+		return
+
+	max_count = 20
+	text = []
+	label = {}
+	index = 0
+	model = gensim.models.keyedvectors.load_word2vec_format('jwiki_w2v.kv', binary=True)
+	for result in model.most_similar(positive=[input_text]):
+		if index >= max_count:
+			break
+		print(result)
+		text.append(result[0])
+		label[result[0]] = result[1]
+		index += 1
+		yield [','.join(text), label]
+
+app = gr.Interface(
+	title='similar-text',
+	description='similar-text',
+	fn=process_text,
+	inputs=gr.Textbox(label='text'),
+	outputs=[
+		gr.Textbox(label='text'),
+		gr.Label(label='scores'),
+	],
+	allow_flagging='never',
+)
+app.queue(concurrency_count=5)
+app.launch()

jwiki_w2v.kv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6147a4b9e96346f3829d0384965a26f3108581eadabc8f598c51a545fcd638a
+size 401236292

requirements.txt

@@ -0,0 +1,2 @@
+gradio
+gensim