tf_model_desc.json overview

Before importing your externally trained TensorFlow models, you must:.

Before importing your externally trained TensorFlow models, you must:

  • save the model in frozen graph (.pb) format

  • create tf_model_desc.json, which describes to your Vertica database how to map its inputs and outputs to input/output tables

Conveniently, the script freeze_tf2_model.py included in the TensorFlow directory of the Machine-Learning-Examples repository (and in opt/vertica/packages/TFIntegration/examples) will do both of these automatically. In most cases, the generated tf_model_desc.json can be used as-is, but for more complex datasets and use cases, you might need to edit it.

tf_model_desc.json for an example dataset

The following tf_model_desc.json is generated from the MNIST handwriting dataset used by the TensorFlow example.

{
    "frozen_graph": "mnist_keras.pb",
    "input_desc": [
        {
            "op_name": "image_input",
            "tensor_map": [
                {
                    "idx": 0,
                    "dim": [
                        1,
                        28,
                        28,
                        1
                    ],
                    "col_start": 0
                }
            ]
        }
    ],
    "output_desc": [
        {
            "op_name": "OUTPUT/Softmax",
            "tensor_map": [
                {
                    "idx": 0,
                    "dim": [
                        1,
                        10
                    ],
                    "col_start": 0
                }
            ]
        }
    ]
}

This file describes the structure of the model's inputs and outputs. It must contain a frozen_graph field that matches the filename of the .pb model, an input_desc field, and an output_desc field.

  • input_desc and output_desc: the descriptions of the input and output nodes in the TensorFlow graph. Each of these include the following fields:

    • op_name: the name of the operation node which is set when creating and training the model. You can typically retrieve the names of these parameters from tfmodel.inputs and tfmodel.outputs. For example:

      
$ print({t.name:t for t in tfmodel.inputs})
{'image_input:0': <tf.Tensor 'image_input:0' shape=(?, 28, 28, 1) dtype=float32>}

      $ print({t.name:t for t in tfmodel.outputs})
{'OUTPUT/Softmax:0': <tf.Tensor 'OUTPUT/Softmax:0' shape=(?, 10) dtype=float32>}

      In this case, the respective values for op_name would be the following.

      • input_desc: image_input

      • output_desc: OUTPUT/Softmax

      For a more detailed example of this process, review the code for freeze_tf2_model.py.

    • tensor_map: how to map the tensor to Vertica columns, which can be specified with the following:

      • idx: the index of the tensor (should be 0 for the first input/output, 1 for the second input/output, etc.)

      • dim: the vector holding the dimensions of the tensor; it provides the number of columns

      • col_start (only used if col_idx is not specified): the starting column index. When used with dim, it specifies a range of indices of Vertica columns starting at col_start and ending at col_start+dim; Vertica starts at the column specified by the index col_start and gets the next dim columns

      • col_idx: the indices in the Vertica columns corresponding to the flattened tensors. This allows you explicitly specify the indices of the Vertica columns that couldn't otherwise be specified as a simple range with col_start and dim (e.g. 1, 3, 5, 7)

      • data_type (not shown): the data type of the input or output, one of the following:

        • TF_FLOAT (default)

        • TF_DOUBLE

        • TF_INT8

        • TF_INT16

        • TF_INT32

        • TF_INT64

Complex tf_model_desc.json example

Below is a more complex example that includes multiple inputs and outputs:

{
    "input_desc": [
        {
            "op_name": "input1",
            "tensor_map": [
                {
                    "idx": 0,
                    "dim": [
                        4
                    ],
                    "col_idx": [
                        0,
                        1,
                        2,
                        3
                    ]
                },
                {
                    "idx": 1,
                    "dim": [
                        2,
                        2
                    ],
                    "col_start": 4
                }
            ]
        },
        {
            "op_name": "input2",
            "tensor_map": [
                {
                    "idx": 0,
                    "dim": [],
                    "col_idx": [
                        8
                    ]
                },
                {
                    "idx": 1,
                    "dim": [
                        2
                    ],
                    "col_start": 9
                }
            ]
        }
    ],
    "output_desc": [
        {
            "op_name": "output",
            "tensor_map": [
                {
                    "idx": 0,
                    "dim": [
                        2
                    ],
                    "col_start": 0
                }
            ]
        }
    ]
}

See also