tmep

-# 내용 기반 유튜브 영상 추천 시스템 개선안
+# Video vector list를 이용한 내용 기반 유튜브 영상 추천 시스템 개선안
 
 
 ## Research field
 ## Research field
 * Video Understanding
 * Video Understanding

@@ -5,6 +5,7 @@ from numpy.linalg import norm
 import pandas as pd
 import pandas as pd
 import math
 import math
 import activation as ac
 import activation as ac
+import time
 
 
 def recommend_videos(tags, segments, tag_model_path, video_model_path, video_id_model, top_k, isPerSegment = True):
 def recommend_videos(tags, segments, tag_model_path, video_model_path, video_id_model, top_k, isPerSegment = True):
     # 이 함수에서 모든걸 다 함
     # 이 함수에서 모든걸 다 함
@@ -108,18 +109,15 @@ def recommend_videos(tags, segments, tag_model_path, video_model_path, video_id_
                         error_tags.append(tag)
                         error_tags.append(tag)
                     
                     
             kernel[k] = segment_vector
             kernel[k] = segment_vector
-        for t in tagKernel:
-            print(t)
         #여기에서 유사한 벡터들을 뽑아냄
         #여기에서 유사한 벡터들을 뽑아냄
         #현재는 비디오id로 영상을 얻을 수 없으므로 반환값으로 비디오 아이디와 태그들, 확률 사용
         #현재는 비디오id로 영상을 얻을 수 없으므로 반환값으로 비디오 아이디와 태그들, 확률 사용
         video_tags = pd.read_csv('/mnt/e/khuhub/2015104192/web/backend/yt8m/esot3ria/segment_tags.csv', encoding='utf8',error_bad_lines=False)
         video_tags = pd.read_csv('/mnt/e/khuhub/2015104192/web/backend/yt8m/esot3ria/segment_tags.csv', encoding='utf8',error_bad_lines=False)
-        video_tags_per_id = pd.read_csv('/mnt/e/khuhub/2015104192/web/backend/yt8m/esot3ria/videoIds.csv', encoding='utf8',error_bad_lines=False)
-        videoId_index = 0
         videoVectorList = []
         videoVectorList = []
         segmentTagList = []
         segmentTagList = []
         prevVideoId = ""
         prevVideoId = ""
         minimunVideoIds = [["",-1.0] for i in range(0,top_k)]
         minimunVideoIds = [["",-1.0] for i in range(0,top_k)]
 
 
+        prev_sev = time.time()
         for i, row in video_tags.iterrows():
         for i, row in video_tags.iterrows():
             video_id = row[0]
             video_id = row[0]
             if video_id == "vid_id":
             if video_id == "vid_id":
@@ -129,16 +127,6 @@ def recommend_videos(tags, segments, tag_model_path, video_model_path, video_id_
             
             
             if video_id[0:4] != prevVideoId[0:4]:
             if video_id[0:4] != prevVideoId[0:4]:
                 
                 
-                currentComparingVideoTags = video_tags_per_id.loc[videoId_index]
-                videoId_index += 1           
-                print(videoTagList)
-                print(currentComparingVideoTags)
-                if (currentComparingVideoTags['segment1'].split(':')[0] not in videoTagList and
-                    currentComparingVideoTags['segment2'].split(':')[0] not in videoTagList and
-                    currentComparingVideoTags['segment3'].split(':')[0] not in videoTagList and
-                    currentComparingVideoTags['segment4'].split(':')[0] not in videoTagList and
-                    currentComparingVideoTags['segment5'].split(':')[0] not in videoTagList):
-                    continue
                 #여기서 모다진걸로 컨볼루션 연산
                 #여기서 모다진걸로 컨볼루션 연산
                 #convmaxima, convidx = convolution(videoVectorList,kernel,prevVideoId)
                 #convmaxima, convidx = convolution(videoVectorList,kernel,prevVideoId)
                 maxima, idx = differenceMax(segmentTagList,tagKernel,tag_vectors,videoTagList)
                 maxima, idx = differenceMax(segmentTagList,tagKernel,tag_vectors,videoTagList)
@@ -167,11 +155,13 @@ def recommend_videos(tags, segments, tag_model_path, video_model_path, video_id_
             for i in range(1,top_k+1):
             for i in range(1,top_k+1):
                 tagList.append([row[i].split(":")[0],row[i].split(":")[1]])
                 tagList.append([row[i].split(":")[0],row[i].split(":")[1]])
             segmentTagList.append(tagList)
             segmentTagList.append(tagList)
-        
+        print(time.time() - prev_sev)
         for i in range(0,top_k):
         for i in range(0,top_k):
             similar_ids.append(minimunVideoIds[i][0])
             similar_ids.append(minimunVideoIds[i][0])
     else:
     else:
         similar_ids = [x[0] for x in video_ids.similar_by_vector(video_vector, top_k)]
         similar_ids = [x[0] for x in video_ids.similar_by_vector(video_vector, top_k)]
+        print(similar_ids)
+        
     return similar_ids
     return similar_ids
  
  
 
 
@@ -379,7 +369,7 @@ VIDEO_ID_MODEL_PATH = "/mnt/e/khuhub/2015104192/web/backend/yt8m/esot3ria/videoI
 
 
 TEST_TAGS_PATH = "/mnt/e/khuhub/2015104192/web/backend/yt8m/esot3ria/test_segement_tags.csv"
 TEST_TAGS_PATH = "/mnt/e/khuhub/2015104192/web/backend/yt8m/esot3ria/test_segement_tags.csv"
 
 
-
+'''
 test(TAG_VECTOR_MODEL_PATH,
 test(TAG_VECTOR_MODEL_PATH,
      VIDEO_VECTOR_MODEL_PATH,
      VIDEO_VECTOR_MODEL_PATH,
      VIDEO_ID_MODEL_PATH,
      VIDEO_ID_MODEL_PATH,
@@ -387,4 +377,4 @@ test(TAG_VECTOR_MODEL_PATH,
      VIDEO_TAGS_PATH,
      VIDEO_TAGS_PATH,
      TEST_TAGS_PATH,
      TEST_TAGS_PATH,
      5)
      5)
-     
\ No newline at end of file
+'''   
\ No newline at end of file