執行create_pet_tf_record.py 找不到object_detection.protos
不好意思, 不太熟悉Python module path 怎樣運作,請指教...
object_detection.protos/string_int_label_map_pb2.py 是建在的, 但不知為何找不到。
!pwd
/content/gdrive/My Drive/models/research
for path in sys.path:
print(path)
/env/python /usr/lib/python36.zip /usr/lib/python3.6 /usr/lib/python3.6/lib-dynload /usr/local/lib/python3.6/dist-packages /usr/lib/python3/dist-packages /usr/local/lib/python3.6/dist-packages/IPython/extensions /root/.ipython
!python object_detection/dataset_tools/create_pet_tf_record.py \
--label_map_path=object_detection/data/pet_label_map.pbtxt \
--data_dir=/content/gdrive/My\ Drive/models/research \
--output_dir=/content/gdrive/My\ Drive/models/research
!ls *.record*
2020-04-13 11:21:40.452298: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.1 Traceback (most recent call last): File "object_detection/dataset_tools/create_pet_tf_record.py", line 44, in <module> from object_detection.utils import label_map_util File "/usr/local/lib/python3.6/dist-packages/object_detection-0.1-py3.6.egg/object_detection/utils/label_map_util.py", line 27, in <module> from object_detection.protos import string_int_label_map_pb2 ImportError: cannot import name 'string_int_label_map_pb2' ls: cannot access '*.record*': No such file or directory
回答列表
-
2020/04/14 上午 07:06WP贊同數:0不贊同數:0留言數:1
重新執行前面的setup後,終於成功繼續, 但最後又卡了, 無論用 tf1.x 還是2.2, 都出現兩處問題:
1. 'tensorflow' has no attribute 'app'
2. 'tensorflow' has no attribute 'contrib'
Traceback (most recent call last): File "object_detection/dataset_tools/create_pet_tf_record.py", line 46, in <module> flags = tf.app.flags AttributeError: module 'tensorflow' has no attribute 'app'
pet_faces_train.record-00000-of-00010 pet_faces_val.record-00000-of-00010 pet_faces_train.record-00001-of-00010 pet_faces_val.record-00001-of-00010 pet_faces_train.record-00002-of-00010 pet_faces_val.record-00002-of-00010 pet_faces_train.record-00003-of-00010 pet_faces_val.record-00003-of-00010 pet_faces_train.record-00004-of-00010 pet_faces_val.record-00004-of-00010 pet_faces_train.record-00005-of-00010 pet_faces_val.record-00005-of-00010 pet_faces_train.record-00006-of-00010 pet_faces_val.record-00006-of-00010 pet_faces_train.record-00007-of-00010 pet_faces_val.record-00007-of-00010 pet_faces_train.record-00008-of-00010 pet_faces_val.record-00008-of-00010 pet_faces_train.record-00009-of-00010 pet_faces_val.record-00009-of-00010
2020-04-13 22:59:32.738838: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.1 Traceback (most recent call last): File "object_detection/model_main.py", line 26, in <module> from object_detection import model_lib File "/usr/local/lib/python3.6/dist-packages/object_detection-0.1-py3.6.egg/object_detection/model_lib.py", line 27, in <module> from object_detection import eval_util File "/usr/local/lib/python3.6/dist-packages/object_detection-0.1-py3.6.egg/object_detection/eval_util.py", line 40, in <module> slim = tf.contrib.slim AttributeError: module 'tensorflow' has no attribute 'contrib'
-
2020/04/16 下午 04:02Jeffrey贊同數:0不贊同數:0留言數:1
請確認一下, 環境安裝的套件:
1. Windows/Linux? 還是 Windows+Anaconda?
2. 安裝 python3.6? tensorflow 使用 CPU or GPU 的版本?
libcudart.so --> 應該是 GPU LIB.
Linux 參考:
1. sudo apt install protobuf-compiler #來安裝protobuf,
2. 將object-detection API中的models/research/object_detection/protos/string_int_label_map.proto
下載下来,
3. 用protoc your/path/to/protos/string_int_label_map.proto --python_out=.命令生成string_int_label_map_pb2.py文件。