{
"cells": [
{
"cell_type": "markdown",
"metadata": {
"pycharm": {
"name": "#%% md\n"
}
},
"source": [
"## Netdata Anomaly Detection Deepdive"
]
},
{
"cell_type": "markdown",
"metadata": {
"pycharm": {
"name": "#%% md\n"
}
},
"source": [
"[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/netdata/netdata/blob/master/ml/notebooks/netdata_anomaly_detection_deepdive.ipynb)"
]
},
{
"cell_type": "markdown",
"metadata": {
"pycharm": {
"name": "#%% md\n"
}
},
"source": [
"This notebook will walk through a simplified python based implementation of the C & C++ code in [`netdata/netdata/ml/`](https://github.com/netdata/netdata/tree/master/ml) used to power the [anomaly detection capabilities](https://github.com/netdata/netdata/blob/master/ml/README.md) of the Netdata agent.\n",
"\n",
"The main goal here is to help interested users learn more about how the machine learning works under the hood. If you just want to get started by enabling ml on your agent you can check out these [simple configuration steps](https://learn.netdata.cloud/docs/agent/ml#configuration). \n",
"\n",
"🚧 **Note**: This functionality is still under active development and considered experimental. Changes might cause the feature to break. We dogfood it internally and among early adopters within the Netdata community to build the feature. If you would like to get involved and help us with some feedback, email us at analytics-ml-team@netdata.cloud or come join us in the [🤖-ml-powered-monitoring](https://discord.gg/4eRSEUpJnc) channel of the Netdata discord. Alternativley, if GitHub is more of your thing, feel free to create a [GitHub discussion](https://github.com/netdata/netdata/discussions?discussions_q=label%3Aarea%2Fml).\n",
"\n",
"In this notebook we will:\n",
"\n",
"1. [**Get raw data**](#get-raw-data): Pull some recent data from one of our demo agents.\n",
"2. [**Add some anomalous data**](#add-some-anomalous-data): Be evil and mess up the tail end of the data to make it obviously \"anomalous\".\n",
"3. [**Lets do some ML!**](#lets-do-some-ml): Implement an unsupervised clustering based approach to anomaly detection.\n",
"4. [**Lets visualize all this!**](#lets-visualize-all-this): Plot and explore all this visually.\n",
"5. [**So, how does it _actually_ work?**](#so-how-does-it-actually-work): Dig a little deeper on what's going on under the hood."
]
},
{
"cell_type": "markdown",
"metadata": {
"pycharm": {
"name": "#%% md\n"
}
},
"source": [
"### Imports & Helper Functions"
]
},
{
"cell_type": "markdown",
"metadata": {
"pycharm": {
"name": "#%% md\n"
}
},
"source": [
"Uncomment and run the next cell to install [netdata-pandas](https://github.com/netdata/netdata-pandas) which we will use to easily pull data from the [Netdata agent REST API](https://learn.netdata.cloud/docs/agent/web/api) into a nice clean [Pandas](https://pandas.pydata.org/) [DataFrame](https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.html) where it will be easier to work with. \n",
"\n",
"Once you have [netdata-pandas](https://github.com/netdata/netdata-pandas) installed you can comment it back out and rerun the cell to clear the output."
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"id": "aL4gm-jUffEx",
"pycharm": {
"is_executing": true
}
},
"outputs": [],
"source": [
"# uncomment the line below (when running in google colab) to install the netdata-pandas library, comment it again when done.\n",
"#!pip install netdata-pandas"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"id": "EMZBHjG4mOQh",
"pycharm": {
"is_executing": true
}
},
"outputs": [],
"source": [
"from datetime import datetime, timedelta\n",
"import itertools\n",
"import random\n",
"import pandas as pd\n",
"import numpy as np\n",
"import matplotlib.pyplot as plt\n",
"import seaborn as sns\n",
"import matplotlib.patches as mpatches\n",
"from sklearn.cluster import KMeans\n",
"from scipy.spatial.distance import cdist\n",
"from netdata_pandas.data import get_data\n",
"\n",
"# helper functions\n",
"\n",
"\n",
"def preprocess_df(df, lags_n, diffs_n, smooth_n):\n",
" \"\"\"Given a pandas dataframe preprocess it to take differences, add smoothing, lags and abs values. \n",
" \"\"\"\n",
" if diffs_n >= 1:\n",
" # take differences\n",
" df = df.diff(diffs_n).dropna()\n",
" if smooth_n >= 2:\n",
" # apply a rolling average to smooth out the data a bit\n",
" df = df.rolling(smooth_n).mean().dropna()\n",
" if lags_n >= 1:\n",
" # for each dimension add a new columns for each of lags_n lags of the differenced and smoothed values for that dimension\n",
" df_columns_new = [f'{col}_lag{n}' for n in range(lags_n+1) for col in df.columns]\n",
" df = pd.concat([df.shift(n) for n in range(lags_n + 1)], axis=1).dropna()\n",
" df.columns = df_columns_new\n",
" # sort columns to have lagged values next to each other for clarity when looking at the feature vectors\n",
" df = df.reindex(sorted(df.columns), axis=1)\n",
" \n",
" # take absolute values as last step\n",
" df = abs(df)\n",
" \n",
" return df\n",
"\n",
"\n",
"def add_shading_to_plot(ax, a, b, t, c='y', alpha=0.2):\n",
" \"\"\"Helper function to add shading to plot and add legend item.\n",
" \"\"\"\n",
" plt.axvspan(a, b, color=c, alpha=alpha, lw=0)\n",
" handles, labels = ax.get_legend_handles_labels()\n",
" patch = mpatches.Patch(color=c, label=t, alpha=alpha)\n",
" handles.append(patch) \n",
" plt.legend(handles=handles)\n",
"\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Inputs & Parameters"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"A full list of all the anomaly detection configuration parameters, and descriptions of each, can be found in the [configuration](https://github.com/netdata/netdata/blob/master/ml/README.md#configuration) section of the [ml readme](https://github.com/netdata/netdata/blob/master/ml/README.md).\n",
"\n",
"Below we will focus on some basic params to decide what data to pull and the main ml params of importance in understanding how it all works.\n",
"\n",
"#### training size/scheduling parameters:\n",
"- `train_every`: How often to train or retrain each model.\n",
"- `num_samples_to_train`: How much of the recent data to train on, for example 3600 would mean training on the last 1 hour of raw data. The default in the netdata agent currently is 14400, so last 4 hours.\n",
"\n",
"#### feature preprocessing related parameters:\n",
"- `num_samples_to_diff`: This is really just a 1 or 0 flag to turn on or off differencing in the feature preprocessing. It defaults to 1 (to take differences) and generally should be left alone.\n",
"- `num_samples_to_smooth`: The extent of smoothing (averaging) applied as part of feature preprocessing.\n",
"- `num_samples_to_lag`: The number of previous values to also include in our feature vector.\n",
"\n",
"#### anomaly score related parameters:\n",
"- `dimension_anomaly_score_threshold`: The threshold on the anomaly score, above which the data it considered anomalous and the [anomaly bit](https://github.com/netdata/netdata/blob/master/ml/README.md#anomaly-bit) is set to 1 (its actually set to 100 in reality but this just to make it behave more like a rate when aggregated in the netdata agent api). By default this is `0.99` which means anything with an anomaly score above 99% is considered anomalous. Decreasing this threshold makes the model more sensitive and will leave to more anomaly bits, increasing it does the opposite.\n",
"\n",
"#### model parameters:\n",
"- `n_clusters_per_dimension`: This is the number of clusters to fit for each model, by default it is set to 2 such that 2 cluster [centroids](https://en.wikipedia.org/wiki/Centroid) will be fit for each model.\n",
"- `max_iterations`: The maximum number of iterations the fitting of the clusters is allowed to take. In reality the clustering will converge a lot sooner than this.\n",
"\n",
"**Note**: There is much more detailed discussion of all there configuration parameters in the [\"Configuration\"](https://github.com/netdata/netdata/blob/master/ml/README.md#configuration) section of the ml readme."
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"id": "tBUVUpR3fohX"
},
"outputs": [],
"source": [
"# data params\n",
"hosts = ['london.my-netdata.io']\n",
"charts = ['system.cpu']\n",
"# if want to just focus on a subset of dims, in this case lets just pick one for simplicity\n",
"dims = ['system.cpu|user'] \n",
"last_n_hours = 2\n",
"# based on last_n_hours define the relevant 'before' and 'after' params for the netdata rest api on the agent\n",
"before = int(datetime.utcnow().timestamp())\n",
"after = int((datetime.utcnow() - timedelta(hours=last_n_hours)).timestamp())\n",
"\n",
"# ml params\n",
"train_every = 3600\n",
"num_samples_to_train = 3600\n",
"num_samples_to_diff = 1\n",
"num_samples_to_smooth = 3\n",
"num_samples_to_lag = 5\n",
"dimension_anomaly_score_threshold = 0.99\n",
"n_clusters_per_dimension = 2\n",
"max_iterations = 1000"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 1. Get raw data"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Next we will use the `get_data()` function from the [netdata-pandas](https://github.com/netdata/netdata-pandas) library to just pull down our raw data from the agent into a Pandas dataframe."
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 660
},
"id": "Ypudrfu-fpje",
"outputId": "b25c7322-03b4-4475-c416-37c3abbe78a4"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"(7200, 1)\n",
"1647978087 1647985286\n"
]
},
{
"data": {
"text/html": [
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"# get raw data\n",
"df = get_data(hosts=hosts, charts=charts, after=after, before=before)\n",
"\n",
"# filter df for just the dims if set\n",
"if len(dims):\n",
" df = df[[dim for dim in dims]]\n",
"\n",
"# set some variables based on our data\n",
"df_timestamp_min = df.index.min()\n",
"df_timestamp_max = df.index.max()\n",
"\n",
"# print some info\n",
"print(df.shape)\n",
"print(df_timestamp_min, df_timestamp_max)\n",
"display(df.head())\n",
"\n",
"# lets just plot each dimension to have a look at it\n",
"for col in df.columns: \n",
"\n",
" # plot dimension, setting index to datetime so its more readable on the plot\n",
" df[[col]].set_index(pd.to_datetime(df.index, unit='s')).plot(title=f'Raw Data - {col}', figsize=(16,6))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 2. Add some anomalous data"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Below we will pick the last `n_tail_anomalous` observations and mess them up in some random but noticable way. In this case we randomly shuffle the data and then multiply each observation by some integer randomly chosen from `integers_to_pick_randomly`"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 405
},
"id": "RDuB5PdjOaAX",
"outputId": "d686cea5-d0a8-4ed4-aa58-64770a063fbb"
},
"outputs": [],
"source": [
"# size of anomalous data\n",
"n_tail_anomalous = 500\n",
"integers_to_pick_randomly = [0,1,5,10]\n",
"\n",
"# randomly scramble data and multiply randomly by some numbers to make it anomalous looking\n",
"anomalous_shape = (n_tail_anomalous, len(df.columns))\n",
"randomly_scrambled_data = np.random.choice(df.tail(n_tail_anomalous).values.reshape(-1,), anomalous_shape)\n",
"random_integers = np.random.choice(integers_to_pick_randomly, anomalous_shape)\n",
"data_anomalous = randomly_scrambled_data * random_integers\n",
"\n",
"# create anomalous dataframe\n",
"df_anomalous = pd.DataFrame(data = data_anomalous, columns = df.columns)\n",
"# make sure it has the expected index since we don't want to shuffle that\n",
"df_anomalous.index = df.tail(n_tail_anomalous).index\n",
"\n",
"# overwrite last n_tail observations with anomalous data\n",
"df.update(df_anomalous)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"In the plot below it should be clear that the light yellow section of the data has been messed with and is now \"anomalous\" or \"strange looking\" in comparison to all the data that comes before it. \n",
"\n",
"Our goal now is to create some sort of [anomaly score](https://github.com/netdata/netdata/blob/master/ml/README.md#anomaly-score) that can easily capture this."
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"scrolled": true
},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"# let's just plot each dimension now that we have added some anomalous data\n",
"for col in df.columns:\n",
" \n",
" ax = df.set_index(pd.to_datetime(df.index, unit='s')).plot(title=f'Anomalous Data Appended - {col}', figsize=(16,6))\n",
" add_shading_to_plot(ax, df_timestamp_max - n_tail_anomalous, df_timestamp_max, 'anomalous data')\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 3. Lets do some ML!"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"\n",
"In this notebook we will just use good old [kmeans](https://scikit-learn.org/stable/modules/generated/sklearn.cluster.KMeans.html) from [scikit-learn](https://scikit-learn.org/stable/index.html). \n",
"\n",
"In reality the Netdata Agent uses the awesome [dlib](https://github.com/davisking/dlib) c++ library and the [`find_clusters_using_kmeans`](http://dlib.net/ml.html#find_clusters_using_kmeans) function along with a few others. You can see the Netdata KMeans code [here](https://github.com/netdata/netdata/blob/master/ml/kmeans/KMeans.cc).\n",
"\n",
"The code below:\n",
"\n",
"1. Will initialize some empty objects to use during model training and inference.\n",
"2. Will loop over every observation and run training and inference in a similar way to how the Agent would process each observation.\n",
"\n",
"Of course the Agent implemtation is a lot more efficient and uses more efficient streaming and buffer based approaches as opposed to the fairly naive implementation below. \n",
"\n",
"The idea in this notebook is to make the general approach as readable and understandable as possible."
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"id": "W6UL8U04ppmM"
},
"outputs": [],
"source": [
"# initialize an empty kmeans model for each dimension\n",
"models = {\n",
" dim: {\n",
" 'model' : KMeans(n_clusters=n_clusters_per_dimension, max_iter=max_iterations),\n",
" 'fitted': False\n",
" } for dim in df.columns\n",
"}\n",
"\n",
"# initialize dictionary for storing anomaly scores for each dim\n",
"anomaly_scores = {\n",
" dim: {\n",
" 't' : [],\n",
" 'anomaly_score': []\n",
" } for dim in df.columns\n",
"}\n",
"\n",
"# initialize dictionary for storing anomaly bits for each dim\n",
"anomaly_bits = {\n",
" dim: {\n",
" 't' : [],\n",
" 'anomaly_bit': []\n",
" }\n",
" for dim in df.columns\n",
"}"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now we are ready to just loop over each row of data and produce anomaly scores once we have some trained models and train or retrain periodically as defined by `train_every`. \n",
"\n",
"**Note**: The Netdata Agent implementation spreads out the training across each `train_every` window as opposed to trying to train all models in one go like the below implementation. It also avoids some obvious edges cases where there is no need to retrain, for example when the data have not changed at all since last model was trained."
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "_wxIeEhGiWYv",
"outputId": "8fdfad43-917d-42d1-8997-a49daac25b3d"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"train at t=1647981687, (n=3600, train_after=1647981687, train_before=1647978087)\n"
]
}
],
"source": [
"# loop over each row of data in dataframe\n",
"for t, row in df.iterrows():\n",
"\n",
" # get n based on timestamp\n",
" n = t - df_timestamp_min\n",
"\n",
" # for each dimension, if we have a fitted model then make predictions\n",
" for dim in df.columns:\n",
"\n",
" # if we have a fitted model, get anomaly score\n",
" if models[dim]['fitted']:\n",
" \n",
" #################################\n",
" # Inference / Scoring\n",
" #################################\n",
"\n",
" # get a buffer of recent data\n",
" buffer_size = num_samples_to_diff + num_samples_to_smooth + num_samples_to_lag * 2\n",
" df_dim_recent = df[[dim]].loc[(t-buffer_size):t]\n",
"\n",
" # preprocess/featurize recent data\n",
" df_dim_recent_preprocessed = preprocess_df(\n",
" df_dim_recent,\n",
" num_samples_to_lag,\n",
" num_samples_to_diff,\n",
" num_samples_to_smooth\n",
" )\n",
"\n",
" # take most recent feature vector\n",
" X = df_dim_recent_preprocessed.tail(1).values\n",
" \n",
" # get the existing trained cluster centers\n",
" cluster_centers = models[dim]['model'].cluster_centers_\n",
"\n",
" # get anomaly score based on the sum of the euclidian distances between the \n",
" # feature vector and each cluster centroid\n",
" raw_anomaly_score = np.sum(cdist(X, cluster_centers, metric='euclidean'), axis=1)[0]\n",
"\n",
" # normalize anomaly score based on min-max normalization\n",
" # https://en.wikipedia.org/wiki/Feature_scaling#Rescaling_(min-max_normalization)\n",
" # the idea here is to convert the raw_anomaly_score we just computed into a number on a\n",
" # [0, 1] scale such that it behaves more like a percentage. We use the min and max raw scores\n",
" # observed during training to achieve this. This would mean that a normalized score of 1 would\n",
" # correspond to a distance as big as the biggest distance (most anomalous) observed on the \n",
" # training data. So scores that are 99% or higher will tend to be as strange or more strange\n",
" # as the most strange 1% observed during training.\n",
" \n",
" # normalize based on scores observed during training the model\n",
" train_raw_anomaly_score_min = models[dim]['train_raw_anomaly_score_min']\n",
" train_raw_anomaly_score_max = models[dim]['train_raw_anomaly_score_max']\n",
" train_raw_anomaly_score_range = train_raw_anomaly_score_max - train_raw_anomaly_score_min\n",
" \n",
" # normalize\n",
" anomaly_score = (raw_anomaly_score - train_raw_anomaly_score_min) / train_raw_anomaly_score_range\n",
" \n",
" # The Netdata Agent does not actually store the normalized_anomaly_score since doing so would require more storage\n",
" # space for each metric, essentially doubling the amount of metrics that need to be stored. Instead, the Netdata Agent\n",
" # makes use of an existing bit (the anomaly bit) in the internal storage representation used by netdata. So if the \n",
" # normalized_anomaly_score passed the dimension_anomaly_score_threshold netdata will flip the corresponding anomaly_bit\n",
" # from 0 to 1 to signify that the observation the scored feature vector is considered \"anomalous\". \n",
" # All without any extra storage overhead required for the Netdata Agent database! Yes it's almost magic :)\n",
"\n",
" # get anomaly bit\n",
" anomaly_bit = 100 if anomaly_score >= dimension_anomaly_score_threshold else 0\n",
" \n",
" # save anomaly score\n",
" anomaly_scores[dim]['t'].append(t)\n",
" anomaly_scores[dim]['anomaly_score'].append(anomaly_score)\n",
"\n",
" # save anomaly bit\n",
" anomaly_bits[dim]['t'].append(t)\n",
" anomaly_bits[dim]['anomaly_bit'].append(anomaly_bit)\n",
" \n",
" # check if the model needs (re)training\n",
" if (n >= num_samples_to_train) & (n % train_every == 0):\n",
" \n",
" #################################\n",
" # Train / Re-Train\n",
" #################################\n",
"\n",
" train_before = t - num_samples_to_train\n",
" train_after = t\n",
" print(f'train at t={t}, (n={n}, train_after={train_after}, train_before={train_before})')\n",
"\n",
" # loop over each dimension/model\n",
" for dim in df.columns:\n",
" \n",
" # get training data based on most recent num_samples_to_train\n",
" df_dim_train = df[[dim]].loc[(t-num_samples_to_train):t]\n",
" \n",
" # preprocess/featurize training data\n",
" df_dim_train_preprocessed = preprocess_df(\n",
" df_dim_train,\n",
" num_samples_to_lag,\n",
" num_samples_to_diff,\n",
" num_samples_to_smooth\n",
" )\n",
"\n",
" # fit model using the fit method of kmeans\n",
" models[dim]['model'].fit(df_dim_train_preprocessed.values) \n",
" models[dim]['fitted'] = True # mark model as fitted\n",
" \n",
" # get cluster centers of model we just trained\n",
" cluster_centers = models[dim]['model'].cluster_centers_\n",
"\n",
" # get training scores, needed to get min and max scores for normalization at inference time\n",
" train_raw_anomaly_scores = np.sum(cdist(df_dim_train_preprocessed.values, cluster_centers, metric='euclidean'), axis=1)\n",
" # save min and max anomaly score during training, used to normalize all scores to be 0,1 scale\n",
" models[dim]['train_raw_anomaly_score_min'] = min(train_raw_anomaly_scores)\n",
" models[dim]['train_raw_anomaly_score_max'] = max(train_raw_anomaly_scores)\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The hard work is now all done. The above cell has processed all the data, trained or retrained models as defined by the inital config, and saved all anomaly scores and anomaly bits.\n",
"\n",
"The rest of the notebook will try to help make more sense of all this."
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"id": "0iN0PCPGiWBx"
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
system.cpu|user
\n",
"
system.cpu|user__anomaly_score
\n",
"
system.cpu|user__anomaly_bit
\n",
"
\n",
"
\n",
"
time_idx
\n",
"
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
1647981888
\n",
"
0.753769
\n",
"
0.228337
\n",
"
0.0
\n",
"
\n",
"
\n",
"
1647984190
\n",
"
0.757576
\n",
"
0.144231
\n",
"
0.0
\n",
"
\n",
"
\n",
"
1647983651
\n",
"
0.753769
\n",
"
0.198606
\n",
"
0.0
\n",
"
\n",
"
\n",
"
1647982084
\n",
"
0.757576
\n",
"
0.189867
\n",
"
0.0
\n",
"
\n",
"
\n",
"
1647983422
\n",
"
1.002506
\n",
"
0.333199
\n",
"
0.0
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" system.cpu|user system.cpu|user__anomaly_score \\\n",
"time_idx \n",
"1647981888 0.753769 0.228337 \n",
"1647984190 0.757576 0.144231 \n",
"1647983651 0.753769 0.198606 \n",
"1647982084 0.757576 0.189867 \n",
"1647983422 1.002506 0.333199 \n",
"\n",
" system.cpu|user__anomaly_bit \n",
"time_idx \n",
"1647981888 0.0 \n",
"1647984190 0.0 \n",
"1647983651 0.0 \n",
"1647982084 0.0 \n",
"1647983422 0.0 "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# create dataframe of anomaly scores\n",
"df_anomaly_scores = pd.DataFrame()\n",
"for dim in anomaly_scores:\n",
" df_anomaly_scores_dim = pd.DataFrame(data=zip(anomaly_scores[dim]['t'],anomaly_scores[dim]['anomaly_score']),columns=['time_idx',f'{dim}__anomaly_score']).set_index('time_idx')\n",
" df_anomaly_scores = df_anomaly_scores.join(df_anomaly_scores_dim, how='outer')\n",
"\n",
"# create dataframe of anomaly bits\n",
"df_anomaly_bits = pd.DataFrame()\n",
"for dim in anomaly_bits:\n",
" df_anomaly_bits_dim = pd.DataFrame(data=zip(anomaly_bits[dim]['t'],anomaly_bits[dim]['anomaly_bit']),columns=['time_idx',f'{dim}__anomaly_bit']).set_index('time_idx')\n",
" df_anomaly_bits = df_anomaly_bits.join(df_anomaly_bits_dim, how='outer')\n",
"\n",
"# join anomaly scores to raw df\n",
"df_final = df.join(df_anomaly_scores, how='outer')\n",
"\n",
"# join anomaly bits to raw df\n",
"df_final = df_final.join(df_anomaly_bits, how='outer')\n",
"\n",
"# let's look at a sample of some scored observations\n",
"display(df_final.tail(num_samples_to_train).sample(5))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"In the dataframe above we see that each observation now also has a column with the `__anomaly_score` and one with the `__anomaly_bit`."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 4. Lets visualize all this!"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now that we have our raw data, our anomaly scores, and our anomaly bits - we can plot this all side by side to get a clear picture of how it all works together.\n",
"\n",
"In the plots below we see that during the light yellow \"anomalous\" period the \"[anomaly scores](https://github.com/netdata/netdata/blob/master/ml/README.md#anomaly-score)\" get elevated to such an extend that many \"[anomaly bits](https://github.com/netdata/netdata/blob/master/ml/README.md#anomaly-bit)\" start flipping from 0 to 1 and essentially \"turning on\" to signal potentially anomalous data."
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 1000
},
"id": "zVoR1BJ5nCGv",
"outputId": "ffcc7765-ea39-47c1-da99-ec79647d0871",
"scrolled": true
},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"
"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"figsize = (20,4)\n",
"\n",
"for dim in models:\n",
"\n",
" # create a dim with the raw data, anomaly score and anomaly bit for the dim\n",
" df_final_dim = df_final[[dim,f'{dim}__anomaly_score',f'{dim}__anomaly_bit']]\n",
" \n",
" # plot raw data, including the anomalous data\n",
" ax = df_final_dim[[dim]].set_index(pd.to_datetime(df_final_dim.index, unit='s')).plot(\n",
" title=f'Raw Data (Anomalous Appended) - {dim}', figsize=figsize\n",
" )\n",
" add_shading_to_plot(ax, df_timestamp_max - n_tail_anomalous, df_timestamp_max, 'Anomalous Data')\n",
" \n",
" # plat the corresponding anomaly scores\n",
" ax = df_final_dim[[f'{dim}__anomaly_score']].set_index(pd.to_datetime(df_final_dim.index, unit='s')).plot(\n",
" title=f'Anomaly Score - {dim}', figsize=figsize\n",
" )\n",
" add_shading_to_plot(ax, df_timestamp_max - n_tail_anomalous, df_timestamp_max, 'Anomalous Data')\n",
" \n",
" # plot the corresponding anomaly bits\n",
" ax = df_final_dim[[f'{dim}__anomaly_bit']].set_index(pd.to_datetime(df_final_dim.index, unit='s')).plot(\n",
" title=f'Anomaly Bit - {dim}', figsize=figsize\n",
" )\n",
" add_shading_to_plot(ax, df_timestamp_max - n_tail_anomalous, df_timestamp_max, 'Anomalous Data')\n",
"\n",
" # finally, plot it all on the same plot (which might not be so easy or clear to read)\n",
" df_final_dim_normalized = (df_final_dim-df_final_dim.min())/(df_final_dim.max()-df_final_dim.min())\n",
" ax = df_final_dim_normalized.set_index(pd.to_datetime(df_final_dim_normalized.index, unit='s')).plot(\n",
" title=f'Combined (Raw, Score, Bit) - {dim}', figsize=figsize\n",
" )\n",
" add_shading_to_plot(ax, df_timestamp_max - n_tail_anomalous, df_timestamp_max, 'Anomalous Data')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The last concept to introduce now is the \"[anomaly rate](https://github.com/netdata/netdata/blob/master/ml/README.md#anomaly-rate)\" which is really just an average over \"anomaly bits\".\n",
"\n",
"For example, in the next cell we will just average all the anomaly bits across the light yellow window of time to find the anomaly rate for the metric within this window. "
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"n_tail_anomalous_anomaly_rate = 96.6%\n",
"\n",
"This means the \"anomaly rate\" within the yellow period of anomalous data was 96.6%\n",
"\n",
"Another way to think of this is that 96.6% of the observations during the yellow \n",
"window were considered anomalous based on the latest trained model.\n"
]
}
],
"source": [
"# average the anomaly bits within the n_tail_anomalous period of the data\n",
"n_tail_anomalous_anomaly_rate = df_final_dim[[f'{dim}__anomaly_bit']].tail(n_tail_anomalous).mean()[0]\n",
"\n",
"print(f'n_tail_anomalous_anomaly_rate = {n_tail_anomalous_anomaly_rate}%')\n",
"print(f'\\nThis means the \"anomaly rate\" within the yellow period of anomalous data was {n_tail_anomalous_anomaly_rate}%')\n",
"print(f'\\nAnother way to think of this is that {n_tail_anomalous_anomaly_rate}% of the observations during the yellow \\nwindow were considered anomalous based on the latest trained model.')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 5. So, how does it _actually_ work?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"In this final section of the notebook below we will dig in to try understand this a bit more intuitivley.\n",
"\n",
"First we will \"[featureize](https://brilliant.org/wiki/feature-vector/)\" or \"preprocess\" all the data. Then we will explore what these feature vectors actually are, how they look, and how we derive anomaly scores based on thier distance to the models cluster centroids."
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
"# preprocess/featurize all data\n",
"df_preprocessed = preprocess_df(\n",
" df,\n",
" num_samples_to_lag,\n",
" num_samples_to_diff,\n",
" num_samples_to_smooth\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now that we have preprocessed all of our data, lets just take a look at it.\n",
"\n",
"You will see that we have essentially just added `num_samples_to_lag` additional columns to the dataframe, one for each lag. The numbers themselve also are now longer the original raw metric values, instead they have first been differenced (just take difference of latest value with pervious value so that we are working with delta's as opposed to original raw metric) and also smoothed (in this case by just averaging the previous `num_samples_to_smooth` previous differenced values).\n",
"\n",
"The idea here is to define the representation that the model will work in. In this case the model will decide if a recent observation is anomalous based on it's corresponding feature vector which is a differenced, smoothed, and lagged array or list of recent values."
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"(7192, 6)\n"
]
},
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
system.cpu|user_lag0
\n",
"
system.cpu|user_lag1
\n",
"
system.cpu|user_lag2
\n",
"
system.cpu|user_lag3
\n",
"
system.cpu|user_lag4
\n",
"
system.cpu|user_lag5
\n",
"
\n",
"
\n",
"
time_idx
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
1647983445
\n",
"
3.330669e-16
\n",
"
0.167293
\n",
"
0.499561
\n",
"
0.167504
\n",
"
0.000633
\n",
"
0.253165
\n",
"
\n",
"
\n",
"
1647980613
\n",
"
5.967300e-03
\n",
"
0.000422
\n",
"
0.166665
\n",
"
0.335848
\n",
"
0.083963
\n",
"
0.083542
\n",
"
\n",
"
\n",
"
1647984383
\n",
"
2.531518e-01
\n",
"
0.083327
\n",
"
0.001899
\n",
"
0.251886
\n",
"
0.083963
\n",
"
0.082700
\n",
"
\n",
"
\n",
"
1647984447
\n",
"
1.696266e-01
\n",
"
0.083542
\n",
"
0.081459
\n",
"
0.082074
\n",
"
0.082280
\n",
"
0.168344
\n",
"
\n",
"
\n",
"
1647983270
\n",
"
5.101800e-03
\n",
"
0.082498
\n",
"
0.082703
\n",
"
0.004262
\n",
"
0.174051
\n",
"
0.001050
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" system.cpu|user_lag0 system.cpu|user_lag1 system.cpu|user_lag2 \\\n",
"time_idx \n",
"1647983445 3.330669e-16 0.167293 0.499561 \n",
"1647980613 5.967300e-03 0.000422 0.166665 \n",
"1647984383 2.531518e-01 0.083327 0.001899 \n",
"1647984447 1.696266e-01 0.083542 0.081459 \n",
"1647983270 5.101800e-03 0.082498 0.082703 \n",
"\n",
" system.cpu|user_lag3 system.cpu|user_lag4 system.cpu|user_lag5 \n",
"time_idx \n",
"1647983445 0.167504 0.000633 0.253165 \n",
"1647980613 0.335848 0.083963 0.083542 \n",
"1647984383 0.251886 0.083963 0.082700 \n",
"1647984447 0.082074 0.082280 0.168344 \n",
"1647983270 0.004262 0.174051 0.001050 "
]
},
"execution_count": 13,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"print(df_preprocessed.shape)\n",
"df_preprocessed.sample(5)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The model works based on these feature vectors. A lot of ML is about training a model to define some [\"compressed representation\"](https://en.wikipedia.org/wiki/Data_compression#Machine_learning) of the training data that can then be useful for new data in some way.\n",
"\n",
"This is exactly what our cluster models are trying to do. They process a big bag of preprocessed feature vectors, covering `num_samples_to_train` raw observations, during training to come up with the best, synthetic, `n_clusters_per_dimension` feature vectors as a useful compressed representation of the training data.\n",
"\n",
"The cell below will just show you what those `n_clusters_per_dimension` (in this case 2) synthetic (made up by the kemans algo) feature vectors are."
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
system.cpu|user_lag0
\n",
"
system.cpu|user_lag1
\n",
"
system.cpu|user_lag2
\n",
"
system.cpu|user_lag3
\n",
"
system.cpu|user_lag4
\n",
"
system.cpu|user_lag5
\n",
"
\n",
" \n",
" \n",
"
\n",
"
centroid 0
\n",
"
0.182626
\n",
"
0.169506
\n",
"
0.100484
\n",
"
0.178778
\n",
"
0.177843
\n",
"
0.100711
\n",
"
\n",
"
\n",
"
centroid 1
\n",
"
0.115532
\n",
"
0.141029
\n",
"
0.276627
\n",
"
0.122611
\n",
"
0.124448
\n",
"
0.276112
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" system.cpu|user_lag0 system.cpu|user_lag1 system.cpu|user_lag2 \\\n",
"centroid 0 0.182626 0.169506 0.100484 \n",
"centroid 1 0.115532 0.141029 0.276627 \n",
"\n",
" system.cpu|user_lag3 system.cpu|user_lag4 system.cpu|user_lag5 \n",
"centroid 0 0.178778 0.177843 0.100711 \n",
"centroid 1 0.122611 0.124448 0.276112 "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# lets pick the first model to look at\n",
"model = list(models.keys())[0]\n",
"\n",
"# get the cluster centroids and put them in a dataframe similar to above\n",
"df_cluster_centers = pd.DataFrame(models[model]['model'].cluster_centers_, columns=df_preprocessed.columns)\n",
"df_cluster_centers.index = [f'centroid {i}' for i in df_cluster_centers.index.values]\n",
"display(df_cluster_centers)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"At inference time we can now use our `n_clusters_per_dimension` cluster centers as a sort of set of \"reference\" feature vectors we can compare against. \n",
"\n",
"When we see a new feature vector that is very far away from these \"reference\" feature vectors, we can take that as a signal that the recent data the feature vecotr was derived from may look significantly different than most of the data the clusters where initially train on. And as such it may be \"anomalous\" or \"strange\" in some way that might be meaningful to you are a user trying to monitor and troubleshoot systems based on these metrics.\n",
"\n",
"To try make this visually clearer we will take 10 random feature vectors from the first half of our data where things were generally normal and we will also take 10 random feature vectors from the yellow anomalous period of time. Lastly we will also include the cluster centroids themselves to see how they compare to both sets of 10 feature vectors. \n",
"\n",
"Basically this is represented in the heatmap below where each row is a processed feature vectors corresponding to some timestamp `t`."
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {
"pycharm": {
"name": "#%%\n"
}
},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 15,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"