How to get train input data on training Tensorflow Object Detection API?

Question

When model of faster_rcnn_resnet101 trains, losses are shown on terminal each step.

I want to know which data is input each step. when loss increases, i don't know why loss increases.

Is there someone knowing how to see input data each step?

Are you trying to view the labelled classification after each epoch, that should not be a problem, you can save y_pred as soon as you complete epoch, exact implementation will be based on your code, and how you want to store the data, but this should give you an overview? — anand_v.singh
@anand_v.singh i've trained model using tensorflow object detection api. i want to save "image file name, ground truth, predict data". I do not know where script to edit on object detection api. — DongHwan Shin
github.com/tensorflow/models/blob/master/research/… In this link at the end of the function run_inference_for_single_image(image, graph) You get the output predicted, you already have the ground truth, save them from there. — anand_v.singh

Prashant Palve Prashant Palve · Accepted Answer · 2019-01-30T05:11:47

You can't check each step result ,but in your object_detection/training directory trained checkpoint prefix is created which updating on after specific no of step. you can check object detection using current trained model.

Eg:

python3 export_inference_graph.py \
    --input_type image_tensor \
    --pipeline_config_path training/ssd_mobilenet_v1_pets.config \
    --trained_checkpoint_prefix training/model.ckpt-25000 \
    --output_directory latest_dataset

here model.ckpt-25000 is the no of steps(25000) trained till now.

How to get train input data on training Tensorflow Object Detection API?

1 Answers