rerun

src/diary_cover_similarity.py

@@ -55,21 +55,23 @@ def save_diary_image_info(save_file, face_to_vec_f):
     }
     count = 0
     with open(save_file, "w") as f:
-        res_dict = es_query("diary", q, 0, 500000)
-
-        for item in res_dict["hits"]["hits"]:
-            diary_id = item["_source"]["id"]
-            # before_cover_url = item["_source"]["before_cover_url"] + "-w"
-            after_cover_url = item["_source"]["after_cover_url"] + "-w"
-            img = url_to_ndarray(after_cover_url)
-            if img.any():
-                count += 1
-                print("count: " + str(count))
-                faces = face_to_vec_f(img)
-                for face in faces:
-                    line = str(diary_id) + "\t" + face["feature"] + "\n"
-                    # print(line)
-                    f.write(line)
+        steps = [(0, 200000), (200000, 400000), (400000, 600000)]
+        for (start, stop) in steps:
+            res_dict = es_query("diary", q, start, stop)
+
+            for item in res_dict["hits"]["hits"]:
+                diary_id = item["_source"]["id"]
+                # before_cover_url = item["_source"]["before_cover_url"] + "-w"
+                after_cover_url = item["_source"]["after_cover_url"] + "-w"
+                img = url_to_ndarray(after_cover_url)
+                if img.any():
+                    count += 1
+                    print("count: " + str(count))
+                    faces = face_to_vec_f(img)
+                    for face in faces:
+                        line = str(diary_id) + "\t" + face["feature"] + "\n"
+                        # print(line)
+                        f.write(line)
 
 
 def save_faiss_index(load_file, save_path):