IncrementalTraining.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "%%time\n",
    "import sagemaker\n",
    "from sagemaker import get_execution_role\n",
    "\n",
    "role = get_execution_role()\n",
    "print(role)\n",
    "sess = sagemaker.Session()\n",
    "\n",
    "bucket = 'sagemaker-bucket' # custom bucket name.\n",
    "prefix = 'my-sample'\n",
    "\n",
    "from sagemaker.amazon.amazon_estimator import get_image_uri\n",
    "\n",
    "training_image = get_image_uri(sess.boto_region_name, 'object-detection', repo_version=\"latest\")\n",
    "print (training_image)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# DataSet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import urllib.request\n",
    "\n",
    "# DataSet\n",
    "train_channel = prefix + '/train'\n",
    "validation_channel = prefix + '/validation'\n",
    "train_annotation_channel = prefix + '/train_annotation'\n",
    "validation_annotation_channel = prefix + '/validation_annotation'\n",
    "\n",
    "s3_train_data = 's3://{}/{}'.format(bucket, train_channel)\n",
    "s3_validation_data = 's3://{}/{}'.format(bucket, validation_channel)\n",
    "s3_train_annotation = 's3://{}/{}'.format(bucket, train_annotation_channel)\n",
    "s3_validation_annotation = 's3://{}/{}'.format(bucket, validation_annotation_channel)\n",
    "\n",
    "train_data = sagemaker.session.s3_input(s3_train_data, distribution='FullyReplicated', \n",
    "                        content_type='image/jpeg', s3_data_type='S3Prefix')\n",
    "validation_data = sagemaker.session.s3_input(s3_validation_data, distribution='FullyReplicated', \n",
    "                             content_type='image/jpeg', s3_data_type='S3Prefix')\n",
    "train_annotation = sagemaker.session.s3_input(s3_train_annotation, distribution='FullyReplicated', \n",
    "                             content_type='image/jpeg', s3_data_type='S3Prefix')\n",
    "validation_annotation = sagemaker.session.s3_input(s3_validation_annotation, distribution='FullyReplicated', \n",
    "                             content_type='image/jpeg', s3_data_type='S3Prefix')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Model\n",
    "s3_model_data = \"s3://sagemaker-bucket/my-sample/output/object-detection-2020-04-21-20-51-50-507/output/model.tar.gz\" #od_model.model_data\n",
    "model_data = sagemaker.session.s3_input(s3_model_data, distribution='FullyReplicated', \n",
    "                             content_type='application/x-sagemaker-model', s3_data_type='S3Prefix')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_channels = {'train': train_data, 'validation': validation_data, 'train_annotation': train_annotation, 'validation_annotation':validation_annotation,'model': model_data}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "s3_output_location = 's3://{}/{}/output'.format(bucket, prefix)\n",
    "\n",
    "od_model = sagemaker.estimator.Estimator(training_image,\n",
    "                                         role, \n",
    "                                         train_instance_count=1, \n",
    "                                         train_instance_type='ml.p3.2xlarge',\n",
    "                                         train_volume_size = 50,\n",
    "                                         train_max_run = 360000,\n",
    "                                         input_mode = 'File',\n",
    "                                         output_path=s3_output_location,\n",
    "                                         sagemaker_session=sess)\n",
    "\n",
    "od_model.set_hyperparameters(base_network='resnet-50',\n",
    "                             #use_pretrained_model=1,\n",
    "                             num_classes=3,  ### label count ###\n",
    "                             mini_batch_size=16,\n",
    "                             epochs=10,  ### epoch count ###\n",
    "                             learning_rate=0.001,\n",
    "                             lr_scheduler_step='10',\n",
    "                             lr_scheduler_factor=0.1,\n",
    "                             optimizer='sgd',\n",
    "                             momentum=0.9,\n",
    "                             weight_decay=0.0005,\n",
    "                             overlap_threshold=0.5,\n",
    "                             nms_threshold=0.45,\n",
    "                             image_shape=512,\n",
    "                             label_width=600,\n",
    "                             num_training_samples=1808)  ### data count ###\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "od_model.fit(inputs=data_channels, logs=True)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "conda_mxnet_p36",
   "language": "python",
   "name": "conda_mxnet_p36"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}