Export training data sample codes in python

JadedEarth — Thu, 13 Oct 2022 13:44:17 GMT

I'm using ArcGIS Jupyter Notebook interface for the first time in ArcGIS Pro version 2.9. I can't find specific guidelines on how to run the Export Training Data for Deep Learning tool in arcgis.learn when you're using local data. Here's what I'm trying to do in Notebook:

Sample code from Land Cover Classification using Satellite Imagery and Deep Learning | ArcGIS API for Python:

import arcgis
from arcgis import learn
arcgis.env.verbose = True

export = learn.export_training_data(input_raster = world_imagery_layer,
output_location = samplefolder,
input_class_data = label_layer.url,
chip_format = "PNG",
tile_size = {"x":400,"y":400},
stride_size = {"x":0,"y":0},
metadata_format = "Classified_Tiles",
context = {"startIndex": 0, "exportAllTiles": False, "cellSize": 2},
context = context,
gis = ent_gis)

Questions:

1. For input raster, my data is located on a file geodatabase. How do I code the path to the geodatabase?

2. Output location: what is the format for path to the destination folder?

3. Input class data: my class data is again on a file geodatabase on my computer. How do I code the path?

4. On context parameter:

- Why does the cell size = 2? My data has 1-meter resolution. What's the appropriate cell size I should specify?

- What does "exportAllTiles" mean?

5. On "gis = ent_gis": What does this line mean?

Appreciate any help.

Re: Export training data sample codes in python

JadedEarth — Fri, 14 Oct 2022 12:57:09 GMT

Hi,

Never mind. Found the answers:

1. On paths: Just use double backslash to format them.

2. On cell size: Use the input raster's cell size (in my case was 0.6)

3. "exportAllTiles": Set the value to False so it would only create data that matches the classes.

4. "gis = ent_gis": I think this is for ArcGIS Online; not needed if your data are local.

topic Export training data sample codes in python in ArcGIS Pro Questions

Export training data sample codes in python

Re: Export training data sample codes in python