This repository is forked and modified from tensorflow object detection model.
To run baseline code for the hackathon, you may follow the instruction below:
-
!!!important: make it as your own private repository and assign right to YITU admin (refer to section 5 in "Hackathon_Baseline_User_Guider"!!!
fork/mirror https://github.com/Dawdleryang/object_detection.git
# cd hackathon_sg git clone https://github.com/your_own_private_repository
-
# activate AWS virtual environment source activate tensorflow_p36
# Setup PYTHONPATH # cd hackathon_sg/ export PYTHONPATH=$PYTHONPATH:`pwd`:`pwd`/slim
-
python script/generate_tfrecord_from_csv.py --image_dir ../input/training/images/ --output_path ../input/yitu --csv_file ../input/training/train_label.csv --validation_set_size 500
-
python script/train.py --logtostderr --train_dir=training/baseline/ --pipeline_config_path=training/hackathon_baseline.config
#to visualize the training results, can only use 8000 for port no. tensorboard --logdir=training/baseline --port 8000
-
python script/eval.py --logtostderr --pipeline_config_path=training/hackathon_baseline.config --checkpoint_dir=training/baseline --eval_dir=eval/baseline
#To visualize the eval results, can only use 8000 for port no. tensorboard --logdir=eval/baseline --port 8000
-
python script/export_inference_graph.py --input_type image_tensor --pipeline_config_path training/hackathon_baseline.config --trained_checkpoint_prefix training/baseline/model.ckpt-20000 --output_directory output/
-
python script/output_csv_results.py threshold=0.5 data_dir=../input/testing/images/ model_path=output/frozen_inference_graph.pb output_path=output/submission.csv label_map=../input/label_map.pbtxt
-
please follow the instruction from ”Hackathon Infra User Manual“ to submit your detection results.
git push your modified codes to your private repository whenever you make a submission.
!!!DO PUSH YOUR CODES FOR EVERY SUBMISSION YOU MADE FOR CODE VERIFICATION PURPOSE!!!