Coverage for /pythoncovmergedfiles/medio/medio/usr/local/lib/python3.8/site-packages/tensorflow/python/debug/lib/common.py: 42%

1# Copyright 2017 The TensorFlow Authors. All Rights Reserved. 

2# 

3# Licensed under the Apache License, Version 2.0 (the "License"); 

4# you may not use this file except in compliance with the License. 

5# You may obtain a copy of the License at 

6# 

7# http://www.apache.org/licenses/LICENSE-2.0 

8# 

9# Unless required by applicable law or agreed to in writing, software 

10# distributed under the License is distributed on an "AS IS" BASIS, 

11# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 

12# See the License for the specific language governing permissions and 

13# limitations under the License. 

14# ============================================================================== 

15"""Common values and methods for TensorFlow Debugger.""" 

16import collections 

17import json 

18 

19GRPC_URL_PREFIX = "grpc://" 

20 

21# A key for a Session.run() call. 

22RunKey = collections.namedtuple("RunKey", ["feed_names", "fetch_names"]) 

23 

24 

25def get_graph_element_name(elem): 

26 """Obtain the name or string representation of a graph element. 

27 

28 If the graph element has the attribute "name", return name. Otherwise, return 

29 a __str__ representation of the graph element. Certain graph elements, such as 

30 `SparseTensor`s, do not have the attribute "name". 

31 

32 Args: 

33 elem: The graph element in question. 

34 

35 Returns: 

36 If the attribute 'name' is available, return the name. Otherwise, return 

37 str(fetch). 

38 """ 

39 

40 return elem.name if hasattr(elem, "name") else str(elem) 

41 

42 

43def get_flattened_names(feeds_or_fetches): 

44 """Get a flattened list of the names in run() call feeds or fetches. 

45 

46 Args: 

47 feeds_or_fetches: Feeds or fetches of the `Session.run()` call. It maybe 

48 a Tensor, an Operation or a Variable. It may also be nested lists, tuples 

49 or dicts. See doc of `Session.run()` for more details. 

50 

51 Returns: 

52 (list of str) A flattened list of fetch names from `feeds_or_fetches`. 

53 """ 

54 

55 lines = [] 

56 if isinstance(feeds_or_fetches, (list, tuple)): 

57 for item in feeds_or_fetches: 

58 lines.extend(get_flattened_names(item)) 

59 elif isinstance(feeds_or_fetches, dict): 

60 for key in feeds_or_fetches: 

61 lines.extend(get_flattened_names(feeds_or_fetches[key])) 

62 else: 

63 # This ought to be a Tensor, an Operation or a Variable, for which the name 

64 # attribute should be available. (Bottom-out condition of the recursion.) 

65 lines.append(get_graph_element_name(feeds_or_fetches)) 

66 

67 return lines 

68 

69 

70def get_run_key(feed_dict, fetches): 

71 """Summarize the names of feeds and fetches as a RunKey JSON string. 

72 

73 Args: 

74 feed_dict: The feed_dict given to the `Session.run()` call. 

75 fetches: The fetches from the `Session.run()` call. 

76 

77 Returns: 

78 A JSON Array consisting of two items. They first items is a flattened 

79 Array of the names of the feeds. The second item is a flattened Array of 

80 the names of the fetches. 

81 """ 

82 return json.dumps(RunKey(get_flattened_names(feed_dict), 

83 get_flattened_names(fetches)))