RaspberryPiTensorFlow

Dockerfile

FROM tensorflow/tensorflow:1.5.0-gpu
RUN apt-get update && apt-get install -y --no-install-recommends \
    python-pip \
    protobuf-compiler \
    python-pil \
    python-lxml \
    python-numpy \
    python-scipy \
    python-dev \
    python-pip \
    python-nose \
    g++ \
    libopenblas-dev \
    git \
    python-tk \
    wget \
    nano
RUN pip --no-cache-dir install \
    lxml \
    jupyter \
    matplotlib \
    pillow \
    Cython \
    h5py
RUN ldconfig
WORKDIR /
RUN mkdir tensorflow
WORKDIR tensorflow
RUN git clone https://github.com/tensorflow/models.git
WORKDIR models/research/
RUN protoc object_detection/protos/*.proto --python_out=.
WORKDIR /
RUN curl https://gist.githubusercontent.com/sjain07/a2fc03cae9e9f061846c86d6e5a6af5b/raw/7c4ad9632b8780126e850573fe10afccea96555d/tf_object_detection.sh > run_object_detection_training.sh
ENTRYPOINT sh run_object_detection_training.sh

ImageClassificationRetraining.sh

git clone https://github.com/googlecodelabs/tensorflow-for-poets-2
cd tensorflow-for-poets-2
curl http://download.tensorflow.org/example_images/flower_photos.tgz | tar xz -C tf_files
ls tf_files/flower_photos
IMAGE_SIZE=224
ARCHITECTURE="mobilenet_0.50_${IMAGE_SIZE}"
tensorboard --logdir tf_files/training_summaries &
python -m scripts.retrain \
  --bottleneck_dir=tf_files/bottlenecks \
  --how_many_training_steps=500 \
  --model_dir=tf_files/models/ \
  --summaries_dir=tf_files/training_summaries/"${ARCHITECTURE}" \
  --output_graph=tf_files/retrained_graph.pb \
  --output_labels=tf_files/retrained_labels.txt \
  --architecture="${ARCHITECTURE}" \
  --image_dir=tf_files/flower_photos
python -m scripts.label_image \
  --graph=tf_files/retrained_graph.pb  \
  --image=tf_files/flower_photos/daisy/21652746_cc379e0eea_m.jpg

PI_README.md

1. run sh setup_tf_pi.sh (installs tensorflow)
2. run sh ImageClassificationRetraining.sh (runs the test image retraining code, you can check the progress at 0.0.0.0:6006)

run.sh

#!/bin/bash
sudo apt-get update
sudo apt-get install -y nvidia-384
sudo docker -v
if [ $? -ne 0 ]; then
  curl -fsSL get.docker.com -o get-docker.sh && sudo sh get-docker.sh
fi

sudo nvidia-docker -v
if [ $? -ne 0 ]; then
    echo "INSTALLING NVIDIA DOCKER"
    wget -P /tmp https://github.com/NVIDIA/nvidia-docker/releases/download/v1.0.1/nvidia-docker_1.0.1-1_amd64.deb
    sudo dpkg -i /tmp/nvidia-docker*.deb && rm /tmp/nvidia-docker*.deb
    sudo apt-get install -y nvidia-modprobe
fi

sudo service nvidia-docker start
curl https://gist.githubusercontent.com/sjain07/a2fc03cae9e9f061846c86d6e5a6af5b/raw/8328fdf017ab6fa3de568f5208e0ec2262011bad/Dockerfile > Dockerfile
sudo nvidia-docker build -t object_detection -f Dockerfile .
sudo nvidia-docker run object_detection

SetupTFPi.sh

sudo apt-get update
sudo apt-get install python-pip python-dev
wget https://github.com/samjabrahams/tensorflow-on-raspberry-pi/releases/download/v1.1.0/tensorflow-1.1.0-cp27-none-linux_armv7l.whl
sudo pip install tensorflow-1.1.0-cp27-none-linux_armv7l.whl
sudo pip uninstall mock
sudo pip install mock
echo 'import tensorflow as tf
hello = tf.constant("Hello, TensorFlow!")
sess = tf.Session()
print(sess.run(hello))' > test.py
python test.py

ssd_mobilenet_v1_pets.config

# SSD with Mobilenet v1, configured for Oxford-IIIT Pets Dataset.
# Users should configure the fine_tune_checkpoint field in the train config as
# well as the label_map_path and input_path fields in the train_input_reader and
# eval_input_reader. Search for "PATH_TO_BE_CONFIGURED" to find the fields that
# should be configured.

model {
  ssd {
    num_classes: 37
    box_coder {
      faster_rcnn_box_coder {
        y_scale: 10.0
        x_scale: 10.0
        height_scale: 5.0
        width_scale: 5.0
      }
    }
    matcher {
      argmax_matcher {
        matched_threshold: 0.5
        unmatched_threshold: 0.5
        ignore_thresholds: false
        negatives_lower_than_unmatched: true
        force_match_for_each_row: true
      }
    }
    similarity_calculator {
      iou_similarity {
      }
    }
    anchor_generator {
      ssd_anchor_generator {
        num_layers: 6
        min_scale: 0.2
        max_scale: 0.95
        aspect_ratios: 1.0
        aspect_ratios: 2.0
        aspect_ratios: 0.5
        aspect_ratios: 3.0
        aspect_ratios: 0.3333
      }
    }
    image_resizer {
      fixed_shape_resizer {
        height: 300
        width: 300
      }
    }
    box_predictor {
      convolutional_box_predictor {
        min_depth: 0
        max_depth: 0
        num_layers_before_predictor: 0
        use_dropout: false
        dropout_keep_probability: 0.8
        kernel_size: 1
        box_code_size: 4
        apply_sigmoid_to_scores: false
        conv_hyperparams {
          activation: RELU_6,
          regularizer {
            l2_regularizer {
              weight: 0.00004
            }
          }
          initializer {
            truncated_normal_initializer {
              stddev: 0.03
              mean: 0.0
            }
          }
          batch_norm {
            train: true,
            scale: true,
            center: true,
            decay: 0.9997,
            epsilon: 0.001,
          }
        }
      }
    }
    feature_extractor {
      type: 'ssd_mobilenet_v1'
      min_depth: 16
      depth_multiplier: 1.0
      conv_hyperparams {
        activation: RELU_6,
        regularizer {
          l2_regularizer {
            weight: 0.00004
          }
        }
        initializer {
          truncated_normal_initializer {
            stddev: 0.03
            mean: 0.0
          }
        }
        batch_norm {
          train: true,
          scale: true,
          center: true,
          decay: 0.9997,
          epsilon: 0.001,
        }
      }
    }
    loss {
      classification_loss {
        weighted_sigmoid {
          anchorwise_output: true
        }
      }
      localization_loss {
        weighted_smooth_l1 {
          anchorwise_output: true
        }
      }
      hard_example_miner {
        num_hard_examples: 3000
        iou_threshold: 0.99
        loss_type: CLASSIFICATION
        max_negatives_per_positive: 3
        min_negatives_per_image: 0
      }
      classification_weight: 1.0
      localization_weight: 1.0
    }
    normalize_loss_by_num_matches: true
    post_processing {
      batch_non_max_suppression {
        score_threshold: 1e-8
        iou_threshold: 0.6
        max_detections_per_class: 100
        max_total_detections: 100
      }
      score_converter: SIGMOID
    }
  }
}

train_config: {
  batch_size: 24
  optimizer {
    rms_prop_optimizer: {
      learning_rate: {
        exponential_decay_learning_rate {
          initial_learning_rate: 0.004
          decay_steps: 800720
          decay_factor: 0.95
        }
      }
      momentum_optimizer_value: 0.9
      decay: 0.9
      epsilon: 1.0
    }
  }
  #fine_tune_checkpoint: "object_detection/ace_venture/models/model/train/model.ckpt"
  #from_detection_checkpoint: false
  # Note: The below line limits the training process to 200K steps, which we
  # empirically found to be sufficient enough to train the pets dataset. This
  # effectively bypasses the learning rate schedule (the learning rate will
  # never decay). Remove the below line to train indefinitely.
  num_steps: 200000
  data_augmentation_options {
    random_horizontal_flip {
    }
  }
  data_augmentation_options {
    ssd_random_crop {
    }
  }
}

train_input_reader: {
  tf_record_input_reader {
    input_path: "object_detection/ace_ventura/data/pet_train_with_masks.record"
  }
  label_map_path: "object_detection/ace_ventura/data/pet_label_map.pbtxt"
}

eval_config: {
  num_examples: 2000
  # Note: The below line limits the evaluation process to 10 evaluations.
  # Remove the below line to evaluate indefinitely.
  max_evals: 10
}

eval_input_reader: {
  tf_record_input_reader {
    input_path: "object_detection/ace_ventura/data/pet_val_with_masks.record"
  }
  label_map_path: "object_detection/ace_ventura/data/pet_label_map.pbtxt"
  shuffle: false
  num_readers: 1
}

tf_object_detection.sh

set -e

cd /
git clone https://github.com/pdollar/coco.git
cd coco/PythonAPI
make
make install
python setup.py install

cd /tensorflow/models/research/
export PYTHONPATH=/tensorflow/models/research/slim:$PYTHONPATH
python object_detection/builders/model_builder_test.py
wget http://www.robots.ox.ac.uk/~vgg/data/pets/data/images.tar.gz
wget http://www.robots.ox.ac.uk/~vgg/data/pets/data/annotations.tar.gz
tar -xvf images.tar.gz
tar -xvf annotations.tar.gz

python object_detection/dataset_tools/create_pet_tf_record.py \
    --label_map_path=object_detection/data/pet_label_map.pbtxt \
    --data_dir=`pwd` \
    --output_dir=`pwd`

export MODEL_DIRECTORY=ace_ventura

cd object_detection
mkdir $MODEL_DIRECTORY

mkdir -p $MODEL_DIRECTORY/data/
mkdir -p $MODEL_DIRECTORY/models/model/train
mkdir -p $MODEL_DIRECTORY/models/model/eval

cp ./data/pet_label_map.pbtxt ./$MODEL_DIRECTORY/data/
cp ../pet_train_with_masks.record ./$MODEL_DIRECTORY/data/
cp ../pet_val_with_masks.record ./$MODEL_DIRECTORY/data/

cd ./$MODEL_DIRECTORY/models/model/
wget https://gist.githubusercontent.com/sjain07/a2fc03cae9e9f061846c86d6e5a6af5b/raw/aebc6033063fa56c2094c3d5a8a962dffbff9ee1/ssd_mobilenet_v1_pets.config

cd /tensorflow/models/research/
perl -i -pe 's/channels=3,\n/channels=3),\n/g' object_detection/data_decoders/tf_example_decoder.py
perl -i -pe 's/dct_method=dct_method/\#dct_method=dct_method/g' object_detection/data_decoders/tf_example_decoder.py
nohup & tensorboard --logdir=train:object_detection/$MODEL_DIRECTORY/models/model/train/,eval:object_detection/$MODEL_DIRECTORY/models/model/eval/ &
python object_detection/train.py \
    --logtostderr \
    --pipeline_config_path=/tensorflow/models/research/object_detection/$MODEL_DIRECTORY/models/model/ssd_mobilenet_v1_pets.config \
    --train_dir=/tensorflow/models/research/object_detection/$MODEL_DIRECTORY/models/model/train/

python object_detection/eval.py \
--logtostderr \
--pipeline_config_path=object_detection/$MODEL_DIRECTORY/models/model/ssd_mobilenet_v1_pets.config \
--eval_dir=object_detection/$MODEL_DIRECTORY/models/model/eval/ \
--checkpoint_dir=object_detection/$MODEL_DIRECTORY/models/model/train/