---
title: Obtain Field Values from a Model
slug: manufacturing-connect-edge/obtain-field-values-from-a-model
docTags: 
createdAt: 2022-10-20T18:36:36.000Z
---

When defining Analytics processors to interact with a Machine Learning model [TensorFlow](docId\:gJi_eMTYQQ-k21OVM_Xas)  or [TensorFlow Images](docId\:gJi_eMTYQQ-k21OVM_Xas)) , you need to fill out a number of fields using values from that model.

This procedure describes how you can obtain the relevant values from your model.

**To obtain the relevant field values:**

1. With TensorFlow installed on your local machine, enter this command in shell:
   `saved_model_cli show --dir {PATH_TO_SAVE_MODEL}/{NAME_OF_SAVED_MODEL} --all`
   This gives an output similar to the screenshot below.
   ::Image[]{src="https://api.archbee.com/api/optimize/SSUUxKZUk9bFTEPNn_6Zo/ID-pxsqmxG5tUV2LB6i5r_form-fields-shell-command.png" size="80" width="782" height="275" caption="TensorFLow output" position="center" showCaption="true"}
2. Find the `The given savedModel SignatureDef contains the following input(s)` section.
3. Copy aside:
   - The input tensor name (input operation) from the `name` parameter (`serving_default_conv1d_input` in the screenshot above).
   - The window size and number of inputs expected from the `shape` parameter. `(-1,48,19)` in the screenshot above means that the window size is `48`, and the number of inputs is `19`.
4. Find the `The given savedModel SignatureDef contains the following output(s)` section.
5. Copy aside the output tensor name (output operation) from the `name `parameter (`StatefulPartitionedCall` in the screenshot above).