File size: 10,000 Bytes
2d13eb2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
[
  {
    "function_name": "Transpose.__init__",
    "file_path": "models/PatchTST.py",
    "function_calls": [
      {
        "function_call_name": "super().__init__",
        "functionality": "Initializes the parent class (nn.Module).",
        "input": "",
        "output": "",
        "location_or_namespace": "torch.nn.Module"
      }
    ]
  },
  {
    "function_name": "Transpose.forward",
    "file_path": "models/PatchTST.py",
    "function_calls": [
      {
        "function_call_name": "x.transpose",
        "functionality": "Transposes a tensor.",
        "input": "*self.dims",
        "output": "Transposed tensor",
        "location_or_namespace": "torch.Tensor"
      },
      {
        "function_call_name": "x.transpose().contiguous",
        "functionality": "Returns a contiguous tensor containing the same data as self tensor.",
        "input": "",
        "output": "Contiguous tensor",
        "location_or_namespace": "torch.Tensor"
      }
    ]
  },
  {
    "function_name": "FlattenHead.__init__",
    "file_path": "models/PatchTST.py",
    "function_calls": [
      {
        "function_call_name": "super().__init__",
        "functionality": "Initializes the parent class (nn.Module).",
        "input": "",
        "output": "",
        "location_or_namespace": "torch.nn.Module"
      },
      {
        "function_call_name": "nn.Flatten",
        "functionality": "Flattens a contiguous range of dims into a tensor.",
        "input": "start_dim=-2",
        "output": "Flatten layer object",
        "location_or_namespace": "torch.nn"
      },
      {
        "function_call_name": "nn.Linear",
        "functionality": "Applies a linear transformation to the incoming data.",
        "input": "nf, target_window",
        "output": "Linear layer object",
        "location_or_namespace": "torch.nn"
      },
      {
        "function_call_name": "nn.Dropout",
        "functionality": "During training, randomly zeroes some of the elements of the input tensor with probability p.",
        "input": "head_dropout",
        "output": "Dropout layer object",
        "location_or_namespace": "torch.nn"
      }
    ]
  },
  {
    "function_name": "FlattenHead.forward",
    "file_path": "models/PatchTST.py",
    "function_calls": []
  },
  {
    "function_name": "Model.__init__",
    "file_path": "models/PatchTST.py",
    "function_calls": [
      {
        "function_call_name": "super().__init__",
        "functionality": "Initializes the parent class (nn.Module).",
        "input": "",
        "output": "",
        "location_or_namespace": "torch.nn.Module"
      },
      {
        "function_call_name": "PatchEmbedding",
        "functionality": "",
        "input": "configs.d_model, patch_len, stride, padding, configs.dropout",
        "output": "PatchEmbedding object",
        "location_or_namespace": "layers.Embed.PatchEmbedding"
      },
      {
        "function_call_name": "FullAttention",
        "functionality": "",
        "input": "False, configs.factor, attention_dropout=configs.dropout, output_attention=False",
        "output": "FullAttention object",
        "location_or_namespace": "layers.SelfAttention_Family.FullAttention"
      },
      {
        "function_call_name": "AttentionLayer",
        "functionality": "",
        "input": "FullAttention_object, configs.d_model, configs.n_heads",
        "output": "AttentionLayer object",
        "location_or_namespace": "layers.SelfAttention_Family.AttentionLayer"
      },
      {
        "function_call_name": "EncoderLayer",
        "functionality": "",
        "input": "AttentionLayer_object, configs.d_model, configs.d_ff, dropout=configs.dropout, activation=configs.activation",
        "output": "EncoderLayer object",
        "location_or_namespace": "layers.Transformer_EncDec.EncoderLayer"
      },
      {
        "function_call_name": "range",
        "functionality": "Generates a sequence of numbers.",
        "input": "configs.e_layers",
        "output": "Range object",
        "location_or_namespace": "builtins"
      },
      {
        "function_call_name": "Transpose",
        "functionality": "Transposes input tensor.",
        "input": "1, 2",
        "output": "Transpose layer object",
        "location_or_namespace": "main.Transpose"
      },
      {
        "function_call_name": "nn.BatchNorm1d",
        "functionality": "Applies Batch Normalization over a 2D or 3D input.",
        "input": "configs.d_model",
        "output": "BatchNorm1d layer object",
        "location_or_namespace": "torch.nn"
      },
      {
        "function_call_name": "nn.Sequential",
        "functionality": "A sequential container.",
        "input": "Transpose_object, nn.BatchNorm1d_object, Transpose_object",
        "output": "Sequential container object",
        "location_or_namespace": "torch.nn"
      },
      {
        "function_call_name": "Encoder",
        "functionality": "",
        "input": "list_of_EncoderLayer_objects, norm_layer=nn.Sequential_object",
        "output": "Encoder object",
        "location_or_namespace": "layers.Transformer_EncDec.Encoder"
      },
      {
        "function_call_name": "int",
        "functionality": "Converts a number or string to an integer.",
        "input": "(configs.seq_len - patch_len) / stride + 2",
        "output": "Integer value",
        "location_or_namespace": "builtins"
      },
      {
        "function_call_name": "FlattenHead",
        "functionality": "Flattens the head of the network and applies linear and dropout layers.",
        "input": "configs.enc_in, self.head_nf, configs.pred_len, head_dropout=configs.dropout",
        "output": "FlattenHead object",
        "location_or_namespace": "main.FlattenHead"
      },
      {
        "function_call_name": "nn.Flatten",
        "functionality": "Flattens a contiguous range of dims into a tensor.",
        "input": "start_dim=-2",
        "output": "Flatten layer object",
        "location_or_namespace": "torch.nn"
      },
      {
        "function_call_name": "nn.Dropout",
        "functionality": "During training, randomly zeroes some of the elements of the input tensor with probability p.",
        "input": "configs.dropout",
        "output": "Dropout layer object",
        "location_or_namespace": "torch.nn"
      },
      {
        "function_call_name": "nn.Linear",
        "functionality": "Applies a linear transformation to the incoming data.",
        "input": "self.head_nf * configs.enc_in, configs.num_class",
        "output": "Linear layer object",
        "location_or_namespace": "torch.nn"
      }
    ]
  },
  {
    "function_name": "Model.forecast",
    "file_path": "models/PatchTST.py",
    "function_calls": [
      {
        "function_call_name": "x_enc.mean",
        "functionality": "Computes the mean of elements across a dimension.",
        "input": "1, keepdim=True",
        "output": "Tensor with mean values",
        "location_or_namespace": "torch.Tensor"
      },
      {
        "function_call_name": "means.detach",
        "functionality": "Returns a new Tensor, detached from the current graph.",
        "input": "",
        "output": "Detached tensor",
        "location_or_namespace": "torch.Tensor"
      },
      {
        "function_call_name": "torch.var",
        "functionality": "Calculates the variance of all elements in the input tensor.",
        "input": "x_enc, dim=1, keepdim=True, unbiased=False",
        "output": "Tensor with variance values",
        "location_or_namespace": "torch"
      },
      {
        "function_call_name": "torch.sqrt",
        "functionality": "Returns a new tensor with the square-root of the elements of input.",
        "input": "torch.var_result + 1e-5",
        "output": "Tensor with square root values",
        "location_or_namespace": "torch"
      },
      {
        "function_call_name": "x_enc.permute",
        "functionality": "Returns a view of the original tensor input with its dimensions permuted.",
        "input": "0, 2, 1",
        "output": "Permuted tensor",
        "location_or_namespace": "torch.Tensor"
      },
      {
        "function_call_name": "torch.reshape",
        "functionality": "Returns a tensor with the same data and number of elements as input, but with the specified shape.",
        "input": "enc_out, (-1, n_vars, enc_out.shape[-2], enc_out.shape[-1])",
        "output": "Reshaped tensor",
        "location_or_namespace": "torch"
      },
      {
        "function_call_name": "stdev.unsqueeze",
        "functionality": "Returns a new tensor with a dimension of size one inserted at the specified position.",
        "input": "1",
        "output": "Unsqueezed tensor",
        "location_or_namespace": "torch.Tensor"
      },
      {
        "function_call_name": "stdev_unsqueezed.repeat",
        "functionality": "Repeats this tensor along the specified dimensions.",
        "input": "1, self.pred_len, 1",
        "output": "Repeated tensor",
        "location_or_namespace": "torch.Tensor"
      },
      {
        "function_call_name": "means.unsqueeze",
        "functionality": "Returns a new tensor with a dimension of size one inserted at the specified position.",
        "input": "1",
        "output": "Unsqueezed tensor",
        "location_or_namespace": "torch.Tensor"
      },
      {
        "function_call_name": "means_unsqueezed.repeat",
        "functionality": "Repeats this tensor along the specified dimensions.",
        "input": "1, self.pred_len, 1",
        "output": "Repeated tensor",
        "location_or_namespace": "torch.Tensor"
      }
    ]
  },
  {
    "function_name": "Model.forward",
    "file_path": "models/PatchTST.py",
    "function_calls": [
      {
        "function_call_name": "self.forecast",
        "functionality": "Performs forecasting using the model.",
        "input": "x_enc, x_mark_enc, x_dec, x_mark_dec",
        "output": "Forecasted tensor",
        "location_or_namespace": "main.Model"
      }
    ]
  }
]