Skip to content

Instantly share code, notes, and snippets.

Embed
What would you like to do?
export DATASET_DIR=/tmp/data/flowers
export TRAIN_DIR=/tmp/data/tfrun
export CHECKPOINT_PATH=/tmp/data/checkpoints
mkdir -p $DATASET_DIR $TRAIN_DIR $CHECKPOINT_PATH
CWD=`pwd`
cd $CHECKPOINT_PATH && wget http://download.tensorflow.org/models/inception_v4_2016_09_09.tar.gz && tar -xvzf inception_v4_2016_09_09.tar.gz
cd $CWD
python download_and_convert_data.py \
--dataset_name=flowers \
--dataset_dir="${DATASET_DIR}"
python train_image_classifier.py \
--train_dir=${TRAIN_DIR} \
--dataset_dir=${DATASET_DIR} \
--dataset_name=flowers \
--dataset_split_name=train \
--model_name=inception_v4 \
--checkpoint_path=${CHECKPOINT_PATH}/inception_v4.ckpt \
--checkpoint_exclude_scopes=InceptionV4/Logits,InceptionV4/AuxLogits \
--save_summaries_secs=120 \
--batch_size=32 \
--save_interval_secs=1800
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment