update model

config.py

@@ -14,10 +14,10 @@ SUBJECT_DATA_FILE = os.path.join(DATA_DIR, "subjectData.csv")
 SAMPLE_DATA_FILE = os.path.join(DATA_DIR, "sampleData.csv")
 STANDARD_NAME_MAP_DATA_FILE = os.path.join(DATA_DIR, "standardData.csv")
 # Model Names
-MODEL_NAME = "Detomo/cl-nagoya-sup-simcse-ja-nss-v_1_0_3"
+MODEL_NAME = "Detomo/cl-nagoya-sup-simcse-ja-nss-v_1_0_5"
 SENTENCE_EMBEDDING_FILE = os.path.join(
     DATA_DIR,
-    "anchor_name_sentence_embeddings_cl-nagoya-sup-simcse-ja-nss-v_1_0_3.pkl",
+    "anchor_name_sentence_embeddings_cl-nagoya-sup-simcse-ja-nss-v_1_0_5.pkl",
 )
 # SENTENCE_EMBEDDING_FILE = None
 MODEL_TYPE = "openvino"

data/{anchor_name_sentence_embeddings(cl-nagoya-sup-simcse-ja-for-standard-name-v0_9_11).pkl → anchor_name_sentence_embeddings_cl-nagoya-sup-simcse-ja-nss-v_1_0_5.pkl}

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4e3d15a8b600fc46392eef317d45fc2fda5b539b05b207b373fa18c498521ae1
-size 12251300
+oid sha256:c82797adfac6a639b7566fdd424f876273177e4ebb4a5279889d73e41630b7bd
+size 21209252

data/anchor_name_sentence_sentence_embeddings(cl-nagoya-sup-simcse-ja-for-standard-name-v0_9_10).pkl

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:062effa64e2fdd7b5b7253b596d777dea75a892b0b011592657ea51c3861ce62
-size 21565604

data/anchor_name_sentence_sentence_embeddings(cl-nagoya-sup-simcse-ja-for-standard-name-v0_9_11).pkl

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:a4ab694e49dc81208646d88e29eb5b593a0a5c9da8236396d4d8d9f1e28a9786
-size 12260516

prepare.py

@@ -4,7 +4,7 @@ from sentence_transformers import (
     export_dynamic_quantized_onnx_model,
 )
 
-MODEL_NAME = "Detomo/cl-nagoya-sup-simcse-ja-nss-v_1_0_3"
+MODEL_NAME = "Detomo/cl-nagoya-sup-simcse-ja-nss-v_1_0_5"
 
 
 def export_model(backend="onnx", use_qint8=False):
@@ -16,9 +16,10 @@ def export_model(backend="onnx", use_qint8=False):
                 quantization_config=None,
                 model_name_or_path=MODEL_NAME,
                 push_to_hub=True,
+                create_pr=True,
             )
         else:
-            model.push_to_hub(MODEL_NAME)
+            model.push_to_hub(MODEL_NAME, create_pr=True)
     elif backend == "onnx":
         model = SentenceTransformer(MODEL_NAME, backend="onnx")
         if use_qint8:
@@ -32,7 +33,7 @@ def export_model(backend="onnx", use_qint8=False):
 
 
 # Export all combinations
-for backend in ["openvino", "onnx"]:
-    for use_qint8 in [True, False]:
-        print(f"Exporting {backend} model with QINT8={use_qint8}")
-        export_model(backend=backend, use_qint8=use_qint8)
+backend = "openvino"
+use_qint8 = False
+print(f"Exporting {backend} model with QINT8={use_qint8}")
+export_model(backend=backend, use_qint8=use_qint8)

services/sentence_transformer_service.py

@@ -53,7 +53,7 @@ class SentenceTransformerService:
             )
             with open(SENTENCE_EMBEDDING_FILE, "wb") as f:
                 pickle.dump(self.anchor_name_sentence_embeddings, f)
-            print(f"Saved anchor name sentence embeddings to {SENTENCE_EMBEDDING_FILE}")
+            print(f"Saved anchor name sentence embeddings to {SENTENCE_EMBEDDING_FILE} with shape {self.anchor_name_sentence_embeddings.shape}")
 
         # Update embeddings in StandardNameMapData
         self.standardNameMapData.update_embeddings(self.anchor_name_sentence_embeddings)