Scribd
Upload
28 views1 page

Purpose of Using Pretrained

The document discusses two files that are used in an object detection model. The ssd_mobilenet_v3_large_coco_2020_01_14.pbtxt file contains the model architecture and design. The frozen_inference_graph.pb file holds the pre-trained weights that were learned during training and contains the model's knowledge.

Uploaded by

asha.py81
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
28 views1 page

Purpose of Using Pretrained

The document discusses two files that are used in an object detection model. The ssd_mobilenet_v3_large_coco_2020_01_14.pbtxt file contains the model architecture and design. The frozen_inference_graph.pb file holds the pre-trained weights that were learned during training and contains the model's knowledge.

Uploaded by

asha.py81
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
You are on page 1/ 1

"The 'ssd_mobilenet_v3_large_coco_2020_01_14.

pbtxt' file is like a blueprint for


our model. It's a configuration file that tells us how the model is structured,
kind of like a set of instructions for building it. It specifies the architecture
of the model, helping us understand its design and how different parts are
connected.

On the other hand, the 'frozen_inference_graph.pb' file contains the actual brains
of our model. It holds the pre-trained weights for the MobileNetV3 model. These
weights are learned during the training process, where the model learns to
recognize patterns in images. So, you can think of this file as the knowledge our
model gained from being trained on a large dataset.

In simpler terms, one file tells us how the model is built


('ssd_mobilenet_v3_large_coco_2020_01_14.pbtxt'), and the other file carries the
knowledge it gained from training ('frozen_inference_graph.pb'). Together, they
make our model ready to identify objects in new pictures based on what it learned
during training."

You might also like