{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "## Train a model locally or remote" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "from sklearn import datasets, svm\n", "digits = datasets.load_digits()\n", "clf = svm.SVC(gamma=0.001, C=100.,probability=True)\n", "clf.fit(digits.data[:-1], digits.target[:-1])\n", "clf.predict(digits.data[-1:])" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "!pip show scikit-learn" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Save model file" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "!pip install joblib\n", "from joblib import dump\n", "dump(clf, 'model.joblib')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Step 1 : Write a model transform script\n", "\n", "#### Make sure you have a ...\n", "\n", "- \"load_model\" function\n", " - input args are model path\n", " - returns loaded model object\n", " - model name is the same as what you saved the model file as (see above step)\n", "

\n", "- \"predict\" function\n", " - input args are the loaded model object and a payload\n", " - returns the result of model.predict\n", " - make sure you format it as a single (or multiple) string return inside a list for real time (for mini batch)\n", " - from a client, a list or string or np.array that is sent for prediction is interpreted as bytes. Do what you have to for converting back to list or string or np.array\n", " - return the error for debugging\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "%%writefile modelscript_sklearn.py\n", "import sklearn\n", "from joblib import load\n", "import numpy as np\n", "import os\n", "\n", "#Return loaded model\n", "def load_model(modelpath):\n", " print(modelpath)\n", " clf = load(os.path.join(modelpath,'model.joblib'))\n", " print(\"loaded\")\n", " return clf\n", "\n", "# return prediction based on loaded model (from the step above) and an input payload\n", "def predict(model, payload):\n", " try:\n", " # locally, payload may come in as an np.ndarray\n", " if type(payload)==np.ndarray:\n", " out = [str(model.predict(np.frombuffer(payload).reshape((1,64))))]\n", " # in remote / container based deployment, payload comes in as a stream of bytes\n", " else:\n", " out = [str(model.predict(np.frombuffer(payload[0]['body']).reshape((1,64))))]\n", " except Exception as e:\n", " out = [type(payload),str(e)] #useful for debugging!\n", " \n", " return out" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Does this work locally? (not \"_in a container locally_\", but _actually_ in local)" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "from modelscript_sklearn import *\n", "model = load_model('.')" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "predict(model,digits.data[-1:])[0]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### ok great! Now let's install ezsmdeploy\n", "In some cases, installs fail due to an existing package installed called greenlet.\n", "This is not a direct dependency of ezsmdeploy but interferes with the installation. \n", "To fix this, either install in a virtualenv as seen above, or do:\n", "pip install ezsmdeploy[locust] --ignore-installed greenlet" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "!pip install ezsmdeploy" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "import ezsmdeploy" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### If you have been running other inference containers in local mode, stop existing containers to avoid conflict" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "!docker container stop $(docker container ls -aq) >/dev/null" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Deploy locally" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "ez = ezsmdeploy.Deploy(model = ['model.joblib'], # if you intend to add models later, pass model as list, otherwise str\n", " script = 'modelscript_sklearn.py',\n", " requirements = ['pyarrow','scikit-learn==0.22.1','numpy','joblib'], #or pass in the path to requirements.txt\n", " instance_type = 'local',\n", " autoscale = True,\n", " #framework = 'sklearn', # not required if you provide requirements\n", " wait = True)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Test containerized version locally" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "out = ez.predictor.predict(digits.data[-1:].tobytes()).decode()\n", "out" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "!docker container stop $(docker container ls -aq) >/dev/null" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Deploy on SageMaker" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "ezonsm = ezsmdeploy.Deploy(model = ['model.joblib','model.joblib'], # example of multimodel endpoint. \n", " script = 'modelscript_sklearn.py',\n", " requirements = ['pyarrow','scikit-learn==0.22.1','numpy','joblib'])" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "#!./src/build-docker.sh test" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "out = ezonsm.predictor.predict(digits.data[-1:].tobytes(),target_model='model1.tar.gz').decode() \n", "out" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Install the additional locust testing functionality to enable automated testing" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "!pip install ezsmdeploy[locust]" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "ezonsm.test(input_data=digits.data[-1:].tobytes(), target_model='model1.tar.gz',usercount=20,hatchrate=10,timeoutsecs=10)" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "\n", "pd.read_csv('src/locuststats_stats.csv')" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "ezonsm.predictor.delete_endpoint()" ] } ], "metadata": { "kernelspec": { "display_name": "conda_python3", "language": "python", "name": "conda_python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.6.10" } }, "nbformat": 4, "nbformat_minor": 4 }