feat: 添加对比模型codet5

app.py

@@ -11,6 +11,7 @@ model2 = AutoModel.from_pretrained("jinaai/jina-embeddings-v2-base-en", trust_re
 model3 = AutoModel.from_pretrained("jinaai/jina-embeddings-v2-base-zh", trust_remote_code=True)
 model4 = SentenceTransformer("aspire/acge_text_embedding")
 model5 = SentenceTransformer("intfloat/multilingual-e5-large")
+model6 = SentenceTransformer("Salesforce/codet5p-110m-embedding")
 
 @spaces.GPU
 def generate(query1, query2, source_code):
@@ -22,8 +23,8 @@ def generate(query1, query2, source_code):
         source_code = "# Use the built-in enumerator\nfor idx, x in enumerate(xs):\n    print(idx, x)"
 
     results = []
-    model_names = ["jinaai/jina-embeddings-v2-base-code", "jinaai/jina-embeddings-v2-base-en", "jinaai/jina-embeddings-v2-base-zh", "aspire/acge_text_embedding", "intfloat/multilingual-e5-large"]
-    for model, name in zip([model1, model2, model3, model4, model5], model_names):
+    model_names = ["jinaai/jina-embeddings-v2-base-code", "jinaai/jina-embeddings-v2-base-en", "jinaai/jina-embeddings-v2-base-zh", "aspire/acge_text_embedding", "intfloat/multilingual-e5-large", "Salesforce/codet5p-110m-embedding"]
+    for model, name in zip([model1, model2, model3, model4, model5, model6], model_names):
         embeddings = model.encode([query1, query2, source_code])
         score1 = cos_sim(embeddings[0], embeddings[2])
         score2 = cos_sim(embeddings[1], embeddings[2])