experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Fix format and add literal explanation in metadata doc.

Browse Source 
PiperOrigin-RevId: 300289479
Change-Id: I51bf686324af2c1582bc10ede0951bcd4153976e
This commit is contained in:
Xunkai Zhang 2020-03-11 03:32:48 -07:00 committed by TensorFlower Gardener
parent 59d97e3414
commit 0a0a204598
1 changed files with 13 additions and 18 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

31
tensorflow/lite/g3doc/convert/metadata.md
View File

@ -83,21 +83,19 @@ Metadata starts by creating a new model info:
```python
from tflite_support import flatbuffers
from tflite_support import metadata as
_metadata
from tflite_support import metadata_schema_py_generated as
_metadata_fb
from tflite_support import metadata as _metadata
from tflite_support import metadata_schema_py_generated as _metadata_fb
""" ... """
"""Creates the metadata for an image classifier."""
# Creates model info.
model_meta = _metadata_fb.ModelMetadataT()
model_meta.name = self.model_info.name
model_meta.name = "MobileNetV1 image classifier"
model_meta.description = ("Identify the most prominent object in the "
"image from a set of 1,001 categories such as "
"trees, animals, food, vehicles, person etc.")
model_meta.version = self.model_info.version
model_meta.version = "v1"
model_meta.author = "TensorFlow"
model_meta.license = ("Apache License. Version 2.0 "
"http://www.apache.org/licenses/LICENSE-2.0.")
@ -130,8 +128,7 @@ input_meta.name = "image"
input_meta.description = (
"Input image to be classified. The expected image is {0} x {1}, with "
"three channels (red, blue, and green) per pixel. Each value in the "
"tensor is a single byte between 0 and 255.".format(
self.model_info.image_width, self.model_info.image_height))
"tensor is a single byte between 0 and 255.".format(160, 160))
input_meta.content = _metadata_fb.ContentT()
input_meta.content.contentProperties = _metadata_fb.ImagePropertiesT()
input_meta.content.contentProperties.colorSpace = (
@ -146,8 +143,8 @@ input_normalization.options.mean = [127.5]
input_normalization.options.std = [127.5]
input_meta.processUnits = [input_normalization]
input_stats = _metadata_fb.StatsT()
input_stats.max = [self.model_info.image_max]
input_stats.min = [self.model_info.image_min]
input_stats.max = [255]
input_stats.min = [0]
input_meta.stats = input_stats
```
@ -170,7 +167,7 @@ output_stats.max = [1.0]
output_stats.min = [0.0]
output_meta.stats = output_stats
label_file = _metadata_fb.AssociatedFileT()
label_file.name = os.path.basename(self.label_file_path)
label_file.name = os.path.basename("your_path_to_label_file")
label_file.description = "Labels for objects that the model can recognize."
label_file.type = _metadata_fb.AssociatedFileType.TENSOR_AXIS_LABELS
output_meta.associatedFiles = [label_file]
@ -192,19 +189,17 @@ b = flatbuffers.Builder(0)
b.Finish(
model_meta.Pack(b),
_metadata.MetadataPopulator.METADATA_FILE_IDENTIFIER)
self.metadata_buf = b.Output()
metadata_buf = b.Output()
```
Once the data structure is ready, the writing of the metadata into the tflite
file is done via the `populate` method:
```python
def _populate_metadata(self):
"""Populates metadata and label file to the model file."""
populator = _metadata.MetadataPopulator.with_model_file(self.model_file)
populator.load_metadata_buffer(self.metadata_buf)
populator.load_associated_files([self.label_file_path])
populator.populate()
populator = _metadata.MetadataPopulator.with_model_file(model_file)
populator.load_metadata_buffer(metadata_buf)
populator.load_associated_files(["your_path_to_label_file"])
populator.populate()
```
#### Verify the metadata