Skip to content

Commit 2155634

Browse files
nomadblnomadbl
committed
convert indentation to spaces
1 parent 2896bf5 commit 2155634

File tree

1 file changed

+94
-94
lines changed

1 file changed

+94
-94
lines changed

‎src/Optional/LogGraph.jl‎

Lines changed: 94 additions & 94 deletions
Original file line numberDiff line numberDiff line change
@@ -2,107 +2,107 @@ using .tensorboard: NodeDef, AttrValue, NameAttrList
22
using .tensorboard: var"AttrValue.ListValue" as AttrValue_ListValue
33

44
"""
5-
log_graph
5+
log_graph
66
"""
77
function log_graph(logger::TBLogger, g::AbstractGraph; step = nothing, nodelabel::Vector{String} = map(string, vertices(g)), nodeop::Vector{String} = map(string, vertices(g)), nodedevice::Vector{String} = fill("cpu", nv(g)), nodevalue::Vector{Any} = fill(nothing, nv(g)))
8-
nv(g) == length(nodelabel) || throw(ArgumentError("length of nodelable must be same as number of vertices"))
9-
nv(g) == length(nodeop) || throw(ArgumentError("length of nodeop must be same as number of vertices"))
10-
nv(g) == length(nodedevice) || throw(ArgumentError("length of nodedevice must be same as number of vertices"))
11-
nv(g) == length(nodevalue) || throw(ArgumentError("length of nodevalue must be same as number of vertices"))
12-
summ = SummaryCollection(graph_summary(g, nodelabel, nodeop, nodedevice, nodevalue))
8+
nv(g) == length(nodelabel) || throw(ArgumentError("length of nodelable must be same as number of vertices"))
9+
nv(g) == length(nodeop) || throw(ArgumentError("length of nodeop must be same as number of vertices"))
10+
nv(g) == length(nodedevice) || throw(ArgumentError("length of nodedevice must be same as number of vertices"))
11+
nv(g) == length(nodevalue) || throw(ArgumentError("length of nodevalue must be same as number of vertices"))
12+
summ = SummaryCollection(graph_summary(g, nodelabel, nodeop, nodedevice, nodevalue))
1313
write_event(logger.file, make_event(logger, summ, step=step))
1414
end
1515

1616
function graph_summary(g, nodelabel, nodeop, nodedevice, nodevalue)
17-
nodes = Vector{NodeDef}()
18-
for v in vertices(g)
19-
name = nodelabel[v]
20-
op = nodeop[v]
21-
input = [nodelabel[x] for x in inneighbors(g, v)]
22-
device = nodedevice[v]
23-
attr = Dict{String, AttrValue}()
24-
x = nodevalue[v]
25-
if isa(x, AbstractString)
26-
attr["value"] = AttrValue(OneOf(:s, Vector{UInt8}(x)))
27-
attr["dtype"] = AttrValue(OneOf(:_type,jltype2tf(typeof(x))))
28-
elseif isa(x, Integer)
29-
attr["value"] = AttrValue(OneOf(:i,Int64(x)))
30-
attr["dtype"] = AttrValue(OneOf(:_type, jltype2tf(typeof(x))))
31-
elseif isa(x, Real)
32-
attr["value"] = AttrValue(OneOf(:f, Float32(x)))
33-
attr["dtype"] = AttrValue(OneOf(:_type, jltype2tf(typeof(x))))
34-
elseif isa(x, Bool)
35-
attr["value"] = AttrValue(OneOf(:b, x))
36-
attr["dtype"] = AttrValue(OneOf(:_type, jltype2tf(typeof(x))))
37-
elseif isa(x, AbstractArray)
38-
shape = TensorShapeProto([TensorShapeProto_Dim(d, "") for d in (collect(size(x)))], false)
39-
t = TensorProto(dtype = jltype2tf(eltype(x)), tensor_shape = shape, tensor_content = serialize_proto(string(x)))
40-
attr["value"] = AttrValue(OneOf(:tensor, t))
41-
listvalue = AttrValue_ListValue(Vector{Vector{UInt8}}(),
42-
Vector{Int64}(),
43-
Vector{Float32}(),
44-
Vector{Bool}(),
45-
Vector{var"#DataType".T}(),
46-
[shape],
47-
Vector{TensorProto}(),
48-
Vector{NameAttrList}())
49-
attr["_output_shapes"] = AttrValue(OneOf(:list, listvalue))
50-
elseif isa(x, Tuple)
51-
listvalue = AttrValue_ListValue([Vector{UInt8}(repr(y)) for y in x],
52-
Vector{Int64}(),
53-
Vector{Float32}(),
54-
Vector{Bool}(),
55-
Vector{var"#DataType".T}(),
56-
Vector{TensorShapeProto}(),
57-
Vector{TensorProto}(),
58-
Vector{NameAttrList}())
59-
attr["value"] = AttrValue(OneOf(:list, listvalue))
60-
shape = TensorShapeProto([TensorShapeProto_Dim(length(x), "")], false)
61-
listvalue = AttrValue_ListValue(Vector{Vector{UInt8}}(),
62-
Vector{Int64}(),
63-
Vector{Float32}(),
64-
Vector{Bool}(),
65-
Vector{var"#DataType".T}(),
66-
[shape],
67-
Vector{TensorProto}(),
68-
Vector{NameAttrList}())
69-
attr["_output_shapes"] = AttrValue(OneOf(:list, listvalue))
70-
elseif isa(x, Function)
71-
attr["value"] = AttrValue(OneOf(:func, NameAttrList(name = repr(x))))
72-
else
73-
#donothing
74-
end
75-
node = NodeDef(name, op, input, device, attr, nothing, nothing)
76-
push!(nodes, node)
77-
end
78-
GraphDef(nodes, nothing, 0, nothing, nothing)
17+
nodes = Vector{NodeDef}()
18+
for v in vertices(g)
19+
name = nodelabel[v]
20+
op = nodeop[v]
21+
input = [nodelabel[x] for x in inneighbors(g, v)]
22+
device = nodedevice[v]
23+
attr = Dict{String, AttrValue}()
24+
x = nodevalue[v]
25+
if isa(x, AbstractString)
26+
attr["value"] = AttrValue(OneOf(:s, Vector{UInt8}(x)))
27+
attr["dtype"] = AttrValue(OneOf(:_type,jltype2tf(typeof(x))))
28+
elseif isa(x, Integer)
29+
attr["value"] = AttrValue(OneOf(:i,Int64(x)))
30+
attr["dtype"] = AttrValue(OneOf(:_type, jltype2tf(typeof(x))))
31+
elseif isa(x, Real)
32+
attr["value"] = AttrValue(OneOf(:f, Float32(x)))
33+
attr["dtype"] = AttrValue(OneOf(:_type, jltype2tf(typeof(x))))
34+
elseif isa(x, Bool)
35+
attr["value"] = AttrValue(OneOf(:b, x))
36+
attr["dtype"] = AttrValue(OneOf(:_type, jltype2tf(typeof(x))))
37+
elseif isa(x, AbstractArray)
38+
shape = TensorShapeProto([TensorShapeProto_Dim(d, "") for d in (collect(size(x)))], false)
39+
t = TensorProto(dtype = jltype2tf(eltype(x)), tensor_shape = shape, tensor_content = serialize_proto(string(x)))
40+
attr["value"] = AttrValue(OneOf(:tensor, t))
41+
listvalue = AttrValue_ListValue(Vector{Vector{UInt8}}(),
42+
Vector{Int64}(),
43+
Vector{Float32}(),
44+
Vector{Bool}(),
45+
Vector{var"#DataType".T}(),
46+
[shape],
47+
Vector{TensorProto}(),
48+
Vector{NameAttrList}())
49+
attr["_output_shapes"] = AttrValue(OneOf(:list, listvalue))
50+
elseif isa(x, Tuple)
51+
listvalue = AttrValue_ListValue([Vector{UInt8}(repr(y)) for y in x],
52+
Vector{Int64}(),
53+
Vector{Float32}(),
54+
Vector{Bool}(),
55+
Vector{var"#DataType".T}(),
56+
Vector{TensorShapeProto}(),
57+
Vector{TensorProto}(),
58+
Vector{NameAttrList}())
59+
attr["value"] = AttrValue(OneOf(:list, listvalue))
60+
shape = TensorShapeProto([TensorShapeProto_Dim(length(x), "")], false)
61+
listvalue = AttrValue_ListValue(Vector{Vector{UInt8}}(),
62+
Vector{Int64}(),
63+
Vector{Float32}(),
64+
Vector{Bool}(),
65+
Vector{var"#DataType".T}(),
66+
[shape],
67+
Vector{TensorProto}(),
68+
Vector{NameAttrList}())
69+
attr["_output_shapes"] = AttrValue(OneOf(:list, listvalue))
70+
elseif isa(x, Function)
71+
attr["value"] = AttrValue(OneOf(:func, NameAttrList(name = repr(x))))
72+
else
73+
#donothing
74+
end
75+
node = NodeDef(name, op, input, device, attr, nothing, nothing)
76+
push!(nodes, node)
77+
end
78+
GraphDef(nodes, nothing, 0, nothing, nothing)
7979
end
8080

8181
function jltype2tf(dtype::DataType)
82-
nodetype =
83-
dtype == UInt8 ? _DataType.DT_UINT8 :
84-
dtype == UInt16 ? _DataType.DT_UINT16 :
85-
dtype == UInt32 ? _DataType.DT_UINT32 :
86-
dtype == UInt64 ? _DataType.DT_UINT64 :
87-
dtype == Int8 ? _DataType.DT_INT8 :
88-
dtype == Int16 ? _DataType.DT_INT16 :
89-
dtype == Int32 ? _DataType.DT_INT32 :
90-
dtype == Int64 ? _DataType.DT_INT64 :
91-
dtype == Float16 ? _DataType.DT_BFLOAT16 :
92-
dtype == Float32 ? _DataType.DT_FLOAT :
93-
dtype == Float64 ? _DataType.DT_DOUBLE :
94-
dtype <: AbstractString ? _DataType.DT_STRING :
95-
dtype == Bool ? _DataType.DT_BOOL :
96-
dtype ∈ [Complex{Float32},
97-
Complex{Float16},
98-
Complex{UInt8},
99-
Complex{UInt16},
100-
Complex{UInt32},
101-
Complex{Int8},
102-
Complex{Int16},
103-
Complex{Int32}] ? _DataType.DT_COMPLEX64 :
104-
dtype ∈ [Complex{Float64},
105-
Complex{UInt64},
106-
Complex{Int64}] ? _DataType.DT_COMPLEX128 :
107-
_DataType.DT_INVALID
82+
nodetype =
83+
dtype == UInt8 ? _DataType.DT_UINT8 :
84+
dtype == UInt16 ? _DataType.DT_UINT16 :
85+
dtype == UInt32 ? _DataType.DT_UINT32 :
86+
dtype == UInt64 ? _DataType.DT_UINT64 :
87+
dtype == Int8 ? _DataType.DT_INT8 :
88+
dtype == Int16 ? _DataType.DT_INT16 :
89+
dtype == Int32 ? _DataType.DT_INT32 :
90+
dtype == Int64 ? _DataType.DT_INT64 :
91+
dtype == Float16 ? _DataType.DT_BFLOAT16 :
92+
dtype == Float32 ? _DataType.DT_FLOAT :
93+
dtype == Float64 ? _DataType.DT_DOUBLE :
94+
dtype <: AbstractString ? _DataType.DT_STRING :
95+
dtype == Bool ? _DataType.DT_BOOL :
96+
dtype ∈ [Complex{Float32},
97+
Complex{Float16},
98+
Complex{UInt8},
99+
Complex{UInt16},
100+
Complex{UInt32},
101+
Complex{Int8},
102+
Complex{Int16},
103+
Complex{Int32}] ? _DataType.DT_COMPLEX64 :
104+
dtype ∈ [Complex{Float64},
105+
Complex{UInt64},
106+
Complex{Int64}] ? _DataType.DT_COMPLEX128 :
107+
_DataType.DT_INVALID
108108
end

0 commit comments

Comments
 (0)