### Install Evo from PyPI

Source: https://github.com/michaelgrupp/evo/blob/master/README.md

Use this command to install the latest release version of evo from PyPI. This is the easiest way to get started with the executables.

```bash
pip install evo
```

--------------------------------

### Run Rerun Example Script

Source: https://github.com/michaelgrupp/evo/blob/master/examples/rerun_example/README.md

Clone the evo repository and execute the provided Python script to start sending data to the Rerun viewer.

```bash
python rerun_example.py
```

--------------------------------

### Install Evo from Source

Source: https://github.com/michaelgrupp/evo/blob/master/README.md

Install evo directly from the source code repository. This is useful for development or when working with a specific branch.

```bash
pip install --editable .
```

--------------------------------

### Install Rerun SDK

Source: https://github.com/michaelgrupp/evo/blob/master/examples/rerun_example/README.md

Install the rerun SDK using pip. Ensure this is done in the same Python environment where evo is installed.

```bash
pip install rerun-sdk
```

--------------------------------

### Check evo installation

Source: https://github.com/michaelgrupp/evo/blob/master/doc/install_in_virtualenv.md

Verifies that evo is installed correctly by running the evo command-line tool.

```shell
evo
```

--------------------------------

### Install Jupyter and Widgets

Source: https://github.com/michaelgrupp/evo/blob/master/doc/jupyter_notebook.md

Installs Jupyter and enables the widgets extension for notebooks. Ensure pip is available.

```bash
pip install jupyter
jupyter nbextension enable --py --sys-prefix widgetsnbextension
```

--------------------------------

### Install evo from source

Source: https://github.com/michaelgrupp/evo/blob/master/doc/install_in_virtualenv.md

Installs evo in editable mode from its source code. Navigate to the evo base source folder first.

```shell
cd <evo>  # go to evo base source folder that contains pyproject.toml
pip install --editable .
```

--------------------------------

### Install evo package

Source: https://github.com/michaelgrupp/evo/blob/master/doc/install_in_virtualenv.md

Installs the evo package and its dependencies within the active virtual environment using pip.

```shell
pip install evo
```

--------------------------------

### Install virtualenvwrapper on Ubuntu

Source: https://github.com/michaelgrupp/evo/blob/master/doc/install_in_virtualenv.md

Installs the virtualenv and virtualenvwrapper packages using apt. Ensure you are on an Ubuntu or Debian-based system.

```shell
sudo apt install python3-virtualenvwrapper
```

--------------------------------

### Start Local Jupyter Notebook Server

Source: https://github.com/michaelgrupp/evo/blob/master/doc/jupyter_notebook.md

Starts a Jupyter notebook server in the current directory and automatically opens it in your default web browser.

```bash
jupyter notebook
```

--------------------------------

### Install Required Packages

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics_interactive.ipynb

Installs ipympl and ipywidgets for interactive plots and widgets in Jupyter environments.

```bash
pip install ipympl ipywidgets
```

--------------------------------

### Evo Configuration File Content

Source: https://github.com/michaelgrupp/evo/wiki/Configuration

Example content of a JSON configuration file generated by evo_config. Parameters from this file take precedence over command line options.

```json
{
    "delta": 1.0,
    "delta_unit": "m",
    "plot": true,
    "pose_relation": "angle_deg",
    "verbose": true
}
```

--------------------------------

### Upgrade Evo from PyPI

Source: https://github.com/michaelgrupp/evo/blob/master/README.md

Run this command to upgrade your existing evo installation to the latest version available on PyPI.

```bash
pip install --upgrade evo
```

--------------------------------

### KITTI Dataset Pose Format Example

Source: https://github.com/michaelgrupp/evo/wiki/Formats

Illustrates the KITTI dataset pose format, where each line represents the first three rows of a 4x4 homogeneous pose matrix flattened into a single line.

```text
a b c d e f g h i j k l
```

--------------------------------

### Absolute Pose Error (APE) Example

Source: https://github.com/michaelgrupp/evo/wiki/Metrics

Example of calculating Absolute Pose Error using evo_ape. Includes alignment option.

```bash
evo_ape tum reference.txt estimate.txt --align
```

--------------------------------

### Load Trajectory Files

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Loads two example trajectory files in TUM format using the file_interface module.

```python
from evo.tools import file_interface

ref_file = "../test/data/freiburg1_xyz-groundtruth.txt"
est_file = "../test/data/freiburg1_xyz-rgbdslam_drift.txt"

traj_ref = file_interface.read_tum_trajectory_file(ref_file)
traj_est = file_interface.read_tum_trajectory_file(est_file)
```

--------------------------------

### Start Remote Jupyter Notebook Server

Source: https://github.com/michaelgrupp/evo/blob/master/doc/jupyter_notebook.md

Starts a Jupyter notebook server on the remote machine without opening a browser, specifying a port for access. Ensure the specified port is not in use.

```bash
jupyter notebook --no-browser --port=8888
```

--------------------------------

### Activate the virtual environment

Source: https://github.com/michaelgrupp/evo/blob/master/doc/install_in_virtualenv.md

Activates the 'evaluation' virtual environment, making its isolated Python installation active.

```shell
workon evaluation
```

--------------------------------

### Relative Pose Error (RPE) Example

Source: https://github.com/michaelgrupp/evo/wiki/Metrics

Example of calculating Relative Pose Error using evo_rpe. Demonstrates pose relation and delta unit options.

```bash
evo_rpe tum reference.txt estimate.txt --pose_relation angle_deg --delta 1 --delta_unit m
```

--------------------------------

### Generate Jupyter Notebook Configuration

Source: https://github.com/michaelgrupp/evo/blob/master/doc/jupyter_notebook.md

Generates the Jupyter notebook configuration file. This is a one-time setup step for disabling token authentication.

```bash
jupyter notebook --generate-config
```

--------------------------------

### Compare APE with evo

Source: https://github.com/michaelgrupp/evo/blob/master/doc/performance.md

Evaluates Absolute Pose Error (APE) using the evo tool with SE(3) Umeyama alignment. This example highlights evo's improved efficiency.

```bash
$ time evo_ape tum fr2_desk_groundtruth.txt fr2_desk_ORB.txt --align

APE w.r.t. translation part (m)
(with SE(3) Umeyama alignment)

       max	0.024300
      mean	0.007492
    median	0.007415
       min	0.000350
      rmse	0.008119
       sse	0.143305
       std	0.003129


real	0m0.735s
user	0m0.764s
sys	0m0.272s
```

--------------------------------

### Enable Start and End Markers

Source: https://github.com/michaelgrupp/evo/wiki/Plotting

Activates markers to indicate the start (circle) and end (cross) points of trajectories in plots.

```bash
evo_config set plot_start_end_markers true
```

--------------------------------

### TUM RGB-D Dataset Trajectory Format Example

Source: https://github.com/michaelgrupp/evo/wiki/Formats

Shows the TUM RGB-D dataset trajectory format, where each line contains timestamp, position (x, y, z), and quaternion orientation (qx, qy, qz, qw).

```text
timestamp x y z q_x q_y q_z q_w
```

--------------------------------

### Delete virtual environment

Source: https://github.com/michaelgrupp/evo/blob/master/doc/install_in_virtualenv.md

Removes the 'evaluation' virtual environment and all its installed packages.

```shell
rmvirtualenv evaluation
```

--------------------------------

### Compare ATE with Python script

Source: https://github.com/michaelgrupp/evo/blob/master/doc/performance.md

Evaluates Absolute Trajectory Error (ATE) using a Python script. This example shows the execution time and resource usage for comparison.

```bash
$ time ./evaluate_ate.py fr2_desk_groundtruth.txt fr2_desk_ORB.txt --verbose

compared_pose_pairs 2223 pairs
absolute_translational_error.rmse 0.008144 m
absolute_translational_error.mean 0.007514 m
absolute_translational_error.median 0.007432 m
absolute_translational_error.std 0.003140 m
absolute_translational_error.min 0.000332 m
absolute_translational_error.max 0.024329 m

real	0m16.753s
user	0m16.824s
sys	0m0.204s
```

--------------------------------

### Get All APE Statistics

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Fetches all available statistics for the Absolute Pose Error as a dictionary. This is useful for a comprehensive analysis of the trajectory accuracy.

```python
ape_stats = ape_metric.get_all_statistics()
pprint.pprint(ape_stats)
```

--------------------------------

### Get All RPE Statistics

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Fetches all available statistics for the Relative Pose Error in a dictionary format. This provides a comprehensive overview of the RPE results.

```python
rpe_stats = rpe_metric.get_all_statistics()
pprint.pprint(rpe_stats)
```

--------------------------------

### Enable LaTeX Rendering for Plots

Source: https://github.com/michaelgrupp/evo/wiki/Plotting

Enables the use of the LaTeX renderer for plot fonts, ensuring consistency with LaTeX documents. May require setting `plot_texsystem` based on your local LaTeX installation.

```bash
evo_config set plot_usetex
```

--------------------------------

### Get Single APE Statistic

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Retrieves a specific statistic, like RMSE, for the Absolute Pose Error. Ensure the APE metric object is already initialized and processed.

```python
ape_stat = ape_metric.get_statistic(metrics.StatisticsType.rmse)
print(ape_stat)
```

--------------------------------

### Basic Command Line Syntax

Source: https://github.com/michaelgrupp/evo/wiki/Metrics

General format for using evo command-line tools with trajectory files. 'format' specifies the input file type.

```bash
command format reference-trajectory estimated-trajectory [options]
```

--------------------------------

### Configure Logging

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics_interactive.ipynb

Initializes logging for the evo tools. This should be called early in your script.

```python
from evo.tools import log
log.configure_logging()
```

--------------------------------

### Run Evo with Configuration File

Source: https://github.com/michaelgrupp/evo/wiki/Configuration

Executes an evo command using parameters loaded from a specified JSON configuration file via the `-c` argument.

```bash
evo_rpe groundtruth.txt estimate2.txt -c rpe_config.json
```

--------------------------------

### Create a virtual environment for evo

Source: https://github.com/michaelgrupp/evo/blob/master/doc/install_in_virtualenv.md

Creates a new virtual environment named 'evaluation'. The --system-site-packages flag is recommended for ROS compatibility.

```shell
mkvirtualenv evaluation --system-site-packages
```

--------------------------------

### Generate Custom Evo Configuration File

Source: https://github.com/michaelgrupp/evo/wiki/Configuration

Converts arbitrary command line parameters into a JSON configuration file. Use `--out` to specify the output file name.

```bash
evo_config generate --pose_relation angle_deg --delta 1 --delta_unit m --verbose --plot --out rpe_config.json
```

--------------------------------

### ROS Bagfile Command Line Syntax

Source: https://github.com/michaelgrupp/evo/wiki/Metrics

Syntax for using evo command-line tools with ROS bagfiles. Specify the bagfile path and relevant topics.

```bash
command bag2 bagfile-path reference-topic estimated-topic [options]
```

--------------------------------

### Show Current Evo Settings

Source: https://github.com/michaelgrupp/evo/wiki/Configuration

Displays the current global settings of evo. Use `--diff` to show only changed parameters from defaults.

```bash
evo_config show
```

```bash
evo_config show plot_mode_default map_tile_provider
```

```bash
evo_config show --diff
```

--------------------------------

### Configure Logging and Utilities

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Sets up logging and imports utility modules for plotting and numerical operations. It also configures plotting settings and enables matplotlib's widget backend.

```python
from evo.tools import log
log.configure_logging(verbose=True, debug=True, silent=False)

import pprint
import numpy as np

from evo.tools import plot
import matplotlib.pyplot as plt

# temporarily override some package settings
from evo.tools.settings import SETTINGS
SETTINGS.plot_usetex = False

plot.apply_settings(SETTINGS)
%matplotlib widget
```

--------------------------------

### Get Single RPE Statistic

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Retrieves a specific statistic, such as RMSE, for the calculated Relative Pose Error. The `rpe_metric` object must have been processed beforehand.

```python
rpe_stat = rpe_metric.get_statistic(metrics.StatisticsType.rmse)
print(rpe_stat)
```

--------------------------------

### Import Core Metrics and Units

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Imports the necessary metrics and unit classes from the evo library.

```python
from evo.core import metrics
from evo.core.units import Unit
```

--------------------------------

### Configure Interactive Widgets

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics_interactive.ipynb

Sets up various ipywidgets for controlling demo parameters like alignment, scale correction, plot mode, and delta units.

```python
import ipywidgets

check_opts_ape = {"align": False, "correct_scale": False, "show_plot": True}
check_boxes_ape=[ipywidgets.Checkbox(description=desc, value=val) for desc, val in check_opts_ape.items()]
check_opts_rpe = {"align": False, "correct_scale": False, "all_pairs": False, "show_plot": True}
check_boxes_rpe=[ipywidgets.Checkbox(description=desc, value=val) for desc, val in check_opts_rpe.items()]
delta_input = ipywidgets.FloatText(value=1.0, description='delta', disabled=False, color='black')
delta_unit_selector=ipywidgets.Dropdown(
    options={u.value: u for u in Unit if u is not Unit.seconds},
    value=Unit.frames, description='delta_unit'
)
plotmode_selector=ipywidgets.Dropdown(
    options={p.value: p for p in PlotMode},
    value=PlotMode.xy, description='plot_mode'
)
pose_relation_selector=ipywidgets.Dropdown(
    options={p.value: p for p in PoseRelation},
    value=PoseRelation.translation_part, description='pose_relation'
)
```

--------------------------------

### Configure virtualenvwrapper

Source: https://github.com/michaelgrupp/evo/blob/master/doc/install_in_virtualenv.md

Adds the necessary configuration to your .bashrc file to enable virtualenvwrapper. Source the file to apply changes immediately.

```shell
echo "export WORKON_HOME=$HOME/.virtualenvs && source /usr/share/virtualenvwrapper/virtualenvwrapper.sh" >> ~/.bashrc
source ~/.bashrc
```

--------------------------------

### Loading ROS Bagfile Trajectories

Source: https://github.com/michaelgrupp/evo/wiki/evo_traj

Load trajectories from a ROS bagfile, specifying topics. Use --all_topics to load all trajectories.

```bash
evo_traj bag2 data /groundtruth /odom /tf:map.base_link
```

--------------------------------

### Interactive APE Demo

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics_interactive.ipynb

Runs an interactive demo for Absolute Pose Error (APE). Configures parameters via GUI widgets and visualizes results with plots. Requires trajectories to be loaded beforehand.

```python
import evo.main_ape as main_ape

count = 0
results = []

def callback_ape(pose_relation, align, correct_scale, plot_mode, show_plot):
    global results, count
    est_name="APE Test #{}".format(count)
    
    result = main_ape.ape(traj_ref, traj_est, est_name=est_name,
                          pose_relation=pose_relation, align=align, correct_scale=correct_scale)
    count += 1
    results.append(result)
    
    if show_plot:
        fig = plt.figure()
        ax = plot.prepare_axis(fig, plot_mode)
        plot.traj(ax, plot_mode, traj_ref, style="--", alpha=0.5)
        plot.traj_colormap(
            ax, result.trajectories[est_name], result.np_arrays["error_array"], plot_mode,
            min_map=result.stats["min"], max_map=result.stats["max"])
        plt.show()
    
_ = ipywidgets.interact_manual(callback_ape, pose_relation=pose_relation_selector, plot_mode=plotmode_selector,
                               **{c.description: c.value for c in check_boxes_ape})
```

--------------------------------

### Set Evo Global Settings

Source: https://github.com/michaelgrupp/evo/wiki/Configuration

Sets global evo settings. Supports string, list, and boolean values. Multiple settings can be set at once.

```bash
evo_config set plot_mode_default xy
```

```bash
evo_config set plot_statistics mean std max
```

```bash
evo_config set plot_show_legend false
```

```bash
evo_config set ros_map_enable_masking
```

```bash
evo_config set plot_mode_default xy plot_statistics mean std max plot_show_legend false ros_map_enable_masking
```

--------------------------------

### Aligning First N Poses

Source: https://github.com/michaelgrupp/evo/wiki/evo_traj

Align trajectories based on the first N poses using the --n_to_align option.

```bash
evo_traj tum traj_1.txt traj_2.txt --ref traj_ref.txt --n_to_align 100
```

--------------------------------

### Sim(3) Umeyama Alignment

Source: https://github.com/michaelgrupp/evo/wiki/evo_traj

Align trajectories using Sim(3) Umeyama method (rotation, translation, and scale) with --align --correct_scale or -as.

```bash
evo_traj tum traj_1.txt traj_2.txt --ref traj_ref.txt --align --correct_scale
```

--------------------------------

### Configure Plotting Settings

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics_interactive.ipynb

Sets up plotting configurations, including figure size, splitting plots, and LaTeX usage. Enables the interactive matplotlib widget backend.

```python
from evo.tools import plot
from evo.tools.plot import PlotMode
from evo.core.metrics import PoseRelation
from evo.core.units import Unit
from evo.tools.settings import SETTINGS

# temporarily override some package settings
SETTINGS.plot_figsize = [6, 6]
SETTINGS.plot_split = True
SETTINGS.plot_usetex = False

# magic plot configuration
import matplotlib.pyplot as plt
%matplotlib widget
```

--------------------------------

### Establish SSH Port Forwarding

Source: https://github.com/michaelgrupp/evo/blob/master/doc/jupyter_notebook.md

Forwards a port from the remote server to your local machine, enabling access to the remote notebook server. Replace 'username' and 'remotehost' with your actual credentials and hostname.

```bash
ssh username@remotehost -L 8889:localhost:8888
```

--------------------------------

### Generate Result Files using evo_ape

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/pandas_bridge.ipynb

Generates result files (zip archives) by running the evo_ape command-line tool. This requires subprocess and specifies input files and output paths.

```python
# generate some result files
import subprocess as sp
cmd_1 = "evo_ape kitti ../test/data/KITTI_00_gt.txt ../test/data/KITTI_00_ORB.txt --save_results ../test/data/res1.zip --no_warnings"
cmd_2 = "evo_ape kitti ../test/data/KITTI_00_gt.txt ../test/data/KITTI_00_SPTAM.txt --save_results ../test/data/res2.zip --no_warnings"
sp.call(cmd_1.split(" "))
sp.call(cmd_2.split(" "))
```

--------------------------------

### Load Trajectory from KITTI File

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/pandas_bridge.ipynb

Loads a trajectory from a KITTI format pose file. Specify the file path.

```python
# ...or from a KITTI file...
traj = file_interface.read_kitti_poses_file("../test/data/KITTI_00_gt.txt")
```

--------------------------------

### Run RPE Calculation with GUI Parameters

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics_interactive.ipynb

This snippet demonstrates how to run the RPE calculation using the `main_rpe` module. It defines a callback function that is triggered by an interactive GUI, allowing users to configure parameters like pose relation, delta, and plot mode before executing the RPE analysis. The results are stored globally and can be optionally plotted.

```python
import evo.main_rpe as main_rpe

count = 0
results = []

def callback_rpe(pose_relation, delta, delta_unit, all_pairs, align, correct_scale, plot_mode, show_plot):
    global results, count
    est_name="RPE Test #{}".format(count)
    result = main_rpe.rpe(traj_ref, traj_est, est_name=est_name,
                          pose_relation=pose_relation, delta=delta, delta_unit=delta_unit, 
                          all_pairs=all_pairs, align=align, correct_scale=correct_scale, 
                          support_loop=True)
    count += 1
    results.append(result)
    
    if show_plot:
        fig = plt.figure()
        ax = plot.prepare_axis(fig, plot_mode)
        plot.traj(ax, plot_mode, traj_ref, style="--", alpha=0.5)
        plot.traj_colormap(
            ax, result.trajectories[est_name], result.np_arrays["error_array"], plot_mode,
            min_map=result.stats["min"], max_map=result.stats["max"])
        plt.show()

_ = ipywidgets.interact_manual(callback_rpe, pose_relation=pose_relation_selector, plot_mode=plotmode_selector, 
                               delta=delta_input, delta_unit=delta_unit_selector, 
                               **{c.description: c.value for c in check_boxes_rpe})
```

--------------------------------

### Process Data for APE Calculation

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Initializes the APE metric with the chosen pose relation and processes the selected trajectory data.

```python
ape_metric = metrics.APE(pose_relation)
ape_metric.process_data(data)
```

--------------------------------

### Loading Text Trajectories

Source: https://github.com/michaelgrupp/evo/wiki/evo_traj

Load multiple trajectories from text files. Supports glob notation for multiple files.

```bash
evo_traj tum traj_1.txt traj_2.txt traj_3.txt
```

--------------------------------

### Plot Multiple Trajectories with Evo

Source: https://github.com/michaelgrupp/evo/blob/master/README.md

This command plots multiple KITTI pose files and the ground truth using the evo_traj tool. It specifies the input files, the reference file, enables plotting, and sets the plot mode to xz.

```bash
cd test/data
evo_traj kitti KITTI_00_ORB.txt KITTI_00_SPTAM.txt --ref=KITTI_00_gt.txt -p --plot_mode=xz
```

--------------------------------

### Set Geographic Map Tile Provider with API Token

Source: https://github.com/michaelgrupp/evo/wiki/Plotting

Set a map tile provider that requires an API token and provide the token. This is necessary for certain map tile services.

```bash
evo_config set map_tile_provider HEREv3.basicMap map_tile_api_token <your token>
```

--------------------------------

### Loading ROS Bagfile with Reference Trajectory

Source: https://github.com/michaelgrupp/evo/wiki/evo_traj

Load a ROS bagfile and designate a reference trajectory using the --ref flag for alignment.

```bash
evo_traj bag2 data.mcap /odom /tf:map.base_link --ref /groundtruth
```

--------------------------------

### Configure RPE Settings

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Sets up parameters for Relative Pose Error calculation, including the pose relation (e.g., rotation angle in degrees) and delta settings for comparing pose pairs.

```python
pose_relation = metrics.PoseRelation.rotation_angle_deg

# normal mode
delta = 1
delta_unit = Unit.frames

# all pairs mode
all_pairs = False  # activate
```

--------------------------------

### Import File Interface and Sync

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics_interactive.ipynb

Imports necessary modules for reading trajectory files and synchronizing them.

```python
from evo.tools import file_interface
from evo.core import sync
```

--------------------------------

### Prepare Trajectory Data for Plotting

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Prepares reference and estimated trajectories by reducing them to specific delta IDs for plotting. Ensures data is ready for visualization functions.

```python
import copy
traj_ref_plot = copy.deepcopy(traj_ref)
traj_est_plot = copy.deepcopy(traj_est)
traj_ref_plot.reduce_to_ids(rpe_metric.delta_ids)
traj_est_plot.reduce_to_ids(rpe_metric.delta_ids)
seconds_from_start = [t - traj_est.timestamps[0] for t in traj_est.timestamps[1:]]
```

--------------------------------

### Basic Trajectory Info Output

Source: https://github.com/michaelgrupp/evo/wiki/evo_traj

Default output format for trajectory information, showing name, pose count, and path length.

```text
name:	groundtruth
infos:	12765 poses, 304.207m path length, 889.019s duration
```

--------------------------------

### Execute TUM RPE Benchmark Script

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Executes the official TUM RGB-D benchmark script 'evaluate_rpe.py' using subprocess. This is useful for comparing results directly with the benchmark's output. Requires the script and input files.

```python
import subprocess as sp
import sys

delta = 15
cmd = cmd = ["python", "../test/tum_benchmark_tools/evaluate_rpe.py", ref_file, est_file, "--delta", str(delta), "--delta_unit", 'f', '--fixed_delta']
out = sp.check_output(cmd)
print(out.decode(sys.stdout.encoding))
```

--------------------------------

### Process and compare multiple trajectory results

Source: https://github.com/michaelgrupp/evo/blob/master/README.md

Processes multiple result files (e.g., from `evo_ape`) to generate a comparative plot and save statistics to a CSV table. Use this to analyze and compare the performance of different algorithms or configurations.

```bash
evo_res results/*.zip -p --save_table results/table.csv
```

--------------------------------

### Load KITTI Trajectories

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics_interactive.ipynb

Loads trajectory data from KITTI format files, which contain SE(3) matrices per line without timestamps.

```python
traj_ref = file_interface.read_kitti_poses_file("../test/data/KITTI_00_gt.txt")
traj_est = file_interface.read_kitti_poses_file("../test/data/KITTI_00_ORB.txt")
```

--------------------------------

### Import Evo and Pandas Modules

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/pandas_bridge.ipynb

Imports the pandas library and specific modules from the evo.tools package for data handling and bridge functionality.

```python
import pandas as pd
from evo.tools import pandas_bridge
from evo.tools import file_interface
```

--------------------------------

### BibTeX citation for evo package

Source: https://github.com/michaelgrupp/evo/blob/master/README.md

Provides the BibTeX entry for citing the evo package in academic publications. Include this in your LaTeX documents when referencing the package.

```bibtex
@misc{grupp2017evo,
  title={evo: Python package for the evaluation of odometry and SLAM.},
  author={Grupp, Michael},
  howpublished={\url{https://github.com/MichaelGrupp/evo}},
  year={2017}
}
```

--------------------------------

### Import Plotting and Display Modules

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/pandas_bridge.ipynb

Imports necessary libraries for plotting and displaying output in a Jupyter environment.

```python
import matplotlib.pyplot as plt
import seaborn as sns
%matplotlib widget

from IPython.display import display
```

--------------------------------

### SE(3) Umeyama Alignment

Source: https://github.com/michaelgrupp/evo/wiki/evo_traj

Align trajectories using SE(3) Umeyama method (rotation and translation) with --align or -a.

```bash
evo_traj tum traj_1.txt traj_2.txt --ref traj_ref.txt --align
```

--------------------------------

### Print Trajectories

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics_interactive.ipynb

Prints the loaded reference and estimated trajectories to the console.

```python
print(traj_ref)
print(traj_est)
```

--------------------------------

### Enable Pose Correspondence Markers

Source: https://github.com/michaelgrupp/evo/wiki/Plotting

Activates markers that visually connect corresponding poses between the reference trajectory and other trajectories in plots. Useful for `evo_ape`, `evo_rpe`, and `evo_traj`.

```bash
evo_config set plot_pose_correspondences true
```

--------------------------------

### Reset Evo Global Settings

Source: https://github.com/michaelgrupp/evo/wiki/Configuration

Resets evo settings to their default values. Can reset all settings or specific listed parameters.

```bash
evo_config reset
```

```bash
evo_config reset plot_figsize tf_cache_debug
```

--------------------------------

### Enable global tab completion

Source: https://github.com/michaelgrupp/evo/blob/master/doc/install_in_virtualenv.md

Enables tab completion for Python commands using argcomplete within your virtual environment. Requires opening a new terminal afterwards.

```shell
activate-global-python-argcomplete --user
```

--------------------------------

### Run TUM ATE Equivalent Script

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Executes the official TUM RGB-D benchmark script `evaluate_ate.py` to compute the RMSE of the xyz difference of aligned trajectories. This requires the script to be available and provides a way to compare results.

```python
import subprocess as sp
import sys
cmd = ["python", "../test/tum_benchmark_tools/evaluate_ate.py", ref_file, est_file, "--max_difference", str(max_diff)]
out = sp.check_output(cmd)
print(out.decode(sys.stdout.encoding))
```

--------------------------------

### Load TF Trajectory from ROS 1 Bagfile

Source: https://github.com/michaelgrupp/evo/wiki/Formats

Loads a specific TF transform chain's trajectory from a ROS 1 bagfile. This command works for both ROS 1 and ROS 2 bagfiles.

```bash
evo_traj bag my_data.bag /tf:map.base_link
```

--------------------------------

### Load TUM Trajectories

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics_interactive.ipynb

Loads trajectory data from TUM format files, which include 3D position and orientation quaternion per line. Trajectories are then synchronized.

```python
traj_ref = file_interface.read_tum_trajectory_file("../test/data/fr2_desk_groundtruth.txt")
traj_est = file_interface.read_tum_trajectory_file("../test/data/fr2_desk_ORB_kf_mono.txt")
traj_ref, traj_est = sync.associate_trajectories(traj_ref, traj_est)
```

--------------------------------

### Export TUM to KITTI format

Source: https://github.com/michaelgrupp/evo/wiki/Formats

Exports multiple TUM trajectory files to the KITTI format. The output files will be saved with a .kitti extension.

```bash
evo_traj tum traj_1.txt traj_2.txt traj_3.txt --save_as_kitti
# (will be saved as *.kitti)
```

--------------------------------

### Reset Plot Settings to Default

Source: https://github.com/michaelgrupp/evo/wiki/Plotting

Restores all Evo configuration settings, including plot parameters, to their default values.

```bash
evo_config reset
```

--------------------------------

### Initialize and Process RPE Metric

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Creates an instance of the RPE metric with specified settings (pose relation, delta, all_pairs) and processes the trajectory data. This prepares the metric for statistic extraction.

```python
rpe_metric = metrics.RPE(pose_relation=pose_relation, delta=delta, delta_unit=delta_unit, all_pairs=all_pairs)
rpe_metric.process_data(data)
```

--------------------------------

### Load ROS 1 Bagfile Trajectory

Source: https://github.com/michaelgrupp/evo/wiki/Formats

Loads trajectory data from a ROS 1 bagfile using the 'bag' command. Specify the bag file and the topic containing pose or transform messages.

```bash
evo_traj bag my_data.bag /odom /pose
```

--------------------------------

### Load Result Files

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/pandas_bridge.ipynb

Loads previously generated result files (zip archives) using the file_interface.load_res_file function.

```python
result_1 = file_interface.load_res_file("../test/data/res1.zip")
result_2 = file_interface.load_res_file("../test/data/res2.zip")
```

--------------------------------

### Configure saving trajectory backups in zip files

Source: https://github.com/michaelgrupp/evo/wiki/Formats

Enables the automatic saving of trajectory backups within the zip files generated by the --save_results option in evo_res. Note that this increases file size.

```bash
evo_config set save_traj_in_zip true
```

--------------------------------

### Specify Custom Static TF Topic

Source: https://github.com/michaelgrupp/evo/wiki/Formats

Loads a TF trajectory from a custom static TF topic. This allows specifying a topic other than the default '/tf_static'.

```bash
/tf:map.base_link:/custom_static_tf
```

--------------------------------

### Load TF Trajectory from ROS 2 Bagfile

Source: https://github.com/michaelgrupp/evo/wiki/Formats

Loads a specific TF transform chain's trajectory from a ROS 2 bagfile. Specify the TF topic and the parent-child frame pair.

```bash
evo_traj bag2 my_data.mcap /tf:map.base_link /tf:odom.base_link
```

--------------------------------

### Load ROS 2 Bagfile Trajectory

Source: https://github.com/michaelgrupp/evo/wiki/Formats

Loads trajectory data from a ROS 2 bagfile using the 'bag2' command. Specify the bag directory or file and the topic containing pose or transform messages.

```bash
evo_traj bag2 my_data /odom /pose
```

--------------------------------

### Full Trajectory Check Output

Source: https://github.com/michaelgrupp/evo/wiki/evo_traj

Most detailed output with --full_check, including data validation checks (SE(3) conform, quaternions, timestamps) and speed statistics.

```text
name:	groundtruth
infos
        duration (s)    889.01894474
        nr. of poses    12765
        path length (m) 304.206897009
        pos_end (m)     [ -3.32159757  -4.64051651  32.7839329 ]
        pos_start (m)   [-0.00489994 -0.01775981 -0.01375532]
        t_end (s)       1502793459.3
        t_start (s)     1502792570.28
checks
        SE(3) conform   yes
        array shapes    ok
        nr. of stamps   ok
        quaternions     ok
        timestamps      ok
stats
        v_avg (km/h)    1.411572
        v_avg (m/s)     0.392103
        v_max (km/h)    3.038775
        v_max (m/s)     0.844104
        v_min (km/h)    0.001567
        v_min (m/s)     0.000435
```

--------------------------------

### Verbose Trajectory Info Output

Source: https://github.com/michaelgrupp/evo/wiki/evo_traj

Detailed trajectory information in verbose mode (-v/--verbose), including duration, pose count, path length, start/end positions, and timestamps.

```text
name:	groundtruth
infos
        duration (s)    889.01894474
        nr. of poses    12765
        path length (m) 304.206897009
        pos_end (m)     [ -3.32159757  -4.64051651  32.7839329 ]
        pos_start (m)   [-0.00489994 -0.01775981 -0.01375532]
        t_end (s)       1502793459.3
        t_start (s)     1502792570.28
```

--------------------------------

### Load Trajectory from TUM File

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/pandas_bridge.ipynb

Loads a trajectory from a TUM format pose file. Specify the file path.

```python
# ...or from a TUM file...
traj = file_interface.read_tum_trajectory_file("../test/data/fr2_desk_ORB.txt")
```

--------------------------------

### Set Geographic Map Tile Provider

Source: https://github.com/michaelgrupp/evo/wiki/Plotting

Configure the map tile provider for geographic maps. The default is 'OpenStreetMap.Mapnik'. You can choose from various providers supported by contextily/xyzservices.

```bash
evo_config set map_tile_provider CartoDB.Positron
```

--------------------------------

### Plotting Trajectories

Source: https://github.com/michaelgrupp/evo/wiki/evo_traj

Plot trajectories using the -p or --plot flag. Specify view with --plot_mode (e.g., xz, xyz).

```bash
evo_traj tum traj_1.txt --plot --plot_mode xz
```

--------------------------------

### Origin Alignment

Source: https://github.com/michaelgrupp/evo/wiki/evo_traj

Perform simple origin alignment for drift/loop closure evaluation using --align_origin. Not based on Umeyama.

```bash
evo_traj tum traj_1.txt --ref traj_ref.txt --align_origin
```

--------------------------------

### Prepare RPE Data

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Defines the reference and estimated trajectories to be used for RPE calculation. Ensure `traj_ref` and `traj_est` are properly loaded and formatted.

```python
data = (traj_ref, traj_est)
```

--------------------------------

### Analyze and Plot Trajectory DataFrame

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/pandas_bridge.ipynb

Demonstrates basic analysis and plotting of a trajectory DataFrame. Displays the head, statistics, and a line plot of the x, y, and z coordinates.

```python
print("First entries of the dataframe:")
display(traj_df.head())

print("Some statistics of the dataframe:")
display(traj_df.describe())

print("A plot:")
traj_df[["x", "y", "z"]].plot(kind="line", subplots=True)
```

--------------------------------

### Load ROS Bag Trajectories

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics_interactive.ipynb

Loads trajectory data from a ROS bag file. Supports various message types like PoseStamped and Odometry. Trajectories are then synchronized.

```python
from rosbags.rosbag1 import Reader as Rosbag1Reader
with Rosbag1Reader("../test/data/ROS_example.bag") as reader:
    traj_ref = file_interface.read_bag_trajectory(reader, "groundtruth")
    traj_est = file_interface.read_bag_trajectory(reader, "ORB-SLAM")
traj_ref, traj_est = sync.associate_trajectories(traj_ref, traj_est)
```

--------------------------------

### Load Trajectory from ROS Bag

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/pandas_bridge.ipynb

Loads a trajectory from a ROS bag file using the file_interface. Requires rosbags library. Specify the bag file path and the topic name.

```python
# load a trajectory from a ROS bag...
from rosbags.rosbag1 import Reader as Rosbag1Reader
with Rosbag1Reader("../test/data/ROS_example.bag") as reader:
    traj = file_interface.read_bag_trajectory(reader, "S-PTAM")
```

--------------------------------

### Plot Trajectories with ROS Map

Source: https://github.com/michaelgrupp/evo/wiki/Plotting

Insert 2D ROS maps into Evo plots by specifying the YAML file. The map will be displayed alongside trajectories in xy/yx plot modes, transformed to the x-y plane origin.

```bash
evo_traj tum file.txt --plot_mode xy --plot --ros_map_yaml map.yaml
```

--------------------------------

### Export EuRoC to TUM format

Source: https://github.com/michaelgrupp/evo/wiki/Formats

Exports a EuRoC ground truth file to the TUM trajectory format. The output file will be saved with a .tum extension.

```bash
evo_traj euroc data.csv --save_as_tum
# (will be saved as data.tum)
```

--------------------------------

### Associate Trajectories by Timestamp

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Associates two trajectories based on matching timestamps, with a maximum allowed time difference. This is crucial for comparing poses at corresponding moments.

```python
from evo.core import sync

max_diff = 0.01

traj_ref, traj_est = sync.associate_trajectories(traj_ref, traj_est, max_diff)
```

--------------------------------

### Calculate TUM RPE Equivalent Metric

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Calculates the RPE metric equivalent to the TUM benchmark script's output programmatically. This allows for direct comparison of metrics calculated by the script and by the evo library.

```python
import evo.metrics as metrics
from evo.units import Unit

tum_rpe_equivalent = metrics.RPE(metrics.PoseRelation.translation_part, delta, Unit.frames, all_pairs=True)
tum_rpe_equivalent.process_data((traj_ref, traj_est))
print(tum_rpe_equivalent.get_statistic(metrics.StatisticsType.mean))
```

--------------------------------

### Plot Trajectories with Geographic Map Tiles

Source: https://github.com/michaelgrupp/evo/wiki/Plotting

Add a map tile to the plot for geo-referenced data. Map tiles are downloaded at runtime using contextily. Specify the EPSG code for the map projection.

```bash
evo_traj tum georeferenced.tum --plot_mode xy --plot --map_tile epsg:32632
```

--------------------------------

### Export TUM to ROS bag format

Source: https://github.com/michaelgrupp/evo/wiki/Formats

Exports multiple TUM trajectory files to a ROS bag file. The output file will be named with a timestamp and contain topics for each input trajectory.

```bash
evo_traj tum traj_1.txt traj_2.txt traj_3.txt --save_as_bag
# (will be saved as <timestamp>.bag with topics traj_1, traj_2 and traj_3)
```

--------------------------------

### Select Trajectory Data for APE

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Determines which trajectory data (original or aligned) to use for APE calculation based on the 'use_aligned_trajectories' flag.

```python
if use_aligned_trajectories:
    data = (traj_ref, traj_est_aligned) 
else:
    data = (traj_ref, traj_est)
```

--------------------------------

### Convert Results to DataFrame and Concatenate

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/pandas_bridge.ipynb

Converts loaded evo result objects into Pandas DataFrames and then concatenates them side-by-side.

```python
result_df_1 = pandas_bridge.result_to_df(result_1)
result_df_2 = pandas_bridge.result_to_df(result_2)

result_df = pd.concat([result_df_1, result_df_2], axis="columns")
display(result_df)
```

--------------------------------

### Set Reference Trajectory Line Style

Source: https://github.com/michaelgrupp/evo/wiki/Plotting

Changes the line style of the reference trajectory to a dashed line ('-').

```bash
evo_config set plot_reference_linestyle -
```

--------------------------------

### Custom 3D Transformation (JSON Format)

Source: https://github.com/michaelgrupp/evo/wiki/evo_traj

Defines a custom 3D transformation using a JSON file with translation and rotation quaternion. This can be applied to trajectory poses.

```json
{
  "x": 0.0,
  "y": 0.0,
  "z": 0.0,
  "qx": 0.0,
  "qy": 0.0,
  "qz": 0.0,
  "qw": 1.0
}
```

--------------------------------

### Plot Trajectories

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Visualizes the reference and estimated trajectories, including an aligned version, using matplotlib.

```python
fig = plt.figure()
traj_by_label = {
    "estimate (not aligned)": traj_est,
    "estimate (aligned)": traj_est_aligned,
    "reference": traj_ref
}
plot.trajectories(fig, traj_by_label, plot.PlotMode.xyz)
plt.show()
```

--------------------------------

### Plot APE Values and Statistics

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Visualizes the Absolute Pose Error over time, along with key statistics. Requires matplotlib and the evo plotting utilities. Ensure `ape_metric` and `ape_stats` are computed.

```python
seconds_from_start = [t - traj_est.timestamps[0] for t in traj_est.timestamps]
fig = plt.figure()
plot.error_array(fig.gca(), ape_metric.error, x_array=seconds_from_start,
                 statistics={s:v for s,v in ape_stats.items() if s != "sse"},
                 name="APE", title="APE w.r.t. " + ape_metric.pose_relation.value, xlabel="$t$ (s)")
plt.show()
```

--------------------------------

### Plot Trajectory with APE Colormapping

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Renders the trajectory with its path colored according to the Absolute Pose Error magnitude. This helps in identifying regions with high error. Requires `traj_ref`, `traj_est_aligned` (or `traj_est`), and computed `ape_metric` and `ape_stats`.

```python
plot_mode = plot.PlotMode.xy
fig = plt.figure()
ax = plot.prepare_axis(fig, plot_mode)
plot.traj(ax, plot_mode, traj_ref, '--', "gray", "reference")
plot.traj_colormap(ax, traj_est_aligned if use_aligned_trajectories else traj_est, ape_metric.error,
                   plot_mode, min_map=ape_stats["min"], max_map=ape_stats["max"])
ax.legend()
plt.show()
```

--------------------------------

### Set Coordinate Axis Marker Scale

Source: https://github.com/michaelgrupp/evo/wiki/Plotting

Activates coordinate axis markers by setting their scale to 0.1. Adjust this value based on trajectory size.

```bash
evo_config set plot_axis_marker_scale 0.1
```

--------------------------------

### Plot Trajectory with RPE Colormapping

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Visualizes the reference and estimated trajectories, with the estimated trajectory colormapped according to the RPE error. Uses the plot utility and requires matplotlib.

```python
import matplotlib.pyplot as plt
import evo.tools.plot as plot

plot_mode = plot.PlotMode.xy
fig = plt.figure()
ax = plot.prepare_axis(fig, plot_mode)
plot.traj(ax, plot_mode, traj_ref_plot, '--', "gray", "reference")
plot.traj_colormap(ax, traj_est_plot, rpe_metric.error, plot_mode, min_map=rpe_stats["min"], max_map=rpe_stats["max"])
ax.legend()
plt.show()
```

--------------------------------

### Deactivate virtual environment

Source: https://github.com/michaelgrupp/evo/blob/master/doc/install_in_virtualenv.md

Exits the currently active virtual environment.

```shell
deactivate
```

--------------------------------

### Calculate TUM ATE Equivalent Metric

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Calculates the RMSE of the translation part of the pose error, mimicking the TUM ATE script's output. This is done using the evo library's APE metric.

```python
tum_ate_equivalent = metrics.APE(metrics.PoseRelation.translation_part)
tum_ate_equivalent.process_data((traj_ref, traj_est_aligned))
print(tum_ate_equivalent.get_statistic(metrics.StatisticsType.rmse))
```

--------------------------------

### Configure ROS Map Viewport Behavior

Source: https://github.com/michaelgrupp/evo/wiki/Plotting

Define how the plot viewport changes when a ROS map is displayed. Options include 'keep_unchanged', 'update', and 'zoom_to_map'. This is particularly useful for automated plot saving.

```bash
evo_config set ros_map_viewport {keep_unchanged, update, zoom_to_map}
```

--------------------------------

### Set Font Family and Scale

Source: https://github.com/michaelgrupp/evo/wiki/Plotting

Switches the plot font to 'serif' and increases the font scale to 1.2 for better readability. This is useful when the default font does not match the document's style.

```bash
evo_config set plot_fontfamily serif plot_fontscale 1.2
```

--------------------------------

### Set Plot Line Width

Source: https://github.com/michaelgrupp/evo/wiki/Plotting

Reduces the plot line width to 1.0 to match a smaller font size, improving overall visual balance.

```bash
evo_config set plot_linewidth 1.0
```

--------------------------------

### Analyze and Plot Result DataFrame

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/pandas_bridge.ipynb

Demonstrates analysis of a concatenated results DataFrame. It displays the 'stats' row and generates a bar plot of the statistics, excluding the 'sse' value.

```python
display(result_df.loc["stats"])

exclude = result_df.loc["stats"].index.isin(["sse"])  # don't plot sse
result_df.loc["stats"][~exclude].plot(kind="bar")
```

--------------------------------

### Filter Trajectory by Motion

Source: https://github.com/michaelgrupp/evo/wiki/evo_traj

Applies a motion filter to remove poses with insufficient translation or rotation. Useful for keeping only significant movements in the data.

```bash
evo_traj tum data.txt -v --motion_filter 0.5 5
```

--------------------------------

### Configure APE Pose Relation

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Sets the pose relation for calculating Absolute Pose Error (APE). Options include translation part, rotation angle, rotation part, or the full transformation.

```python
pose_relation = metrics.PoseRelation.translation_part
use_aligned_trajectories = False
```

--------------------------------

### Calculate Absolute Pose Error (APE) for S-PTAM trajectory

Source: https://github.com/michaelgrupp/evo/blob/master/README.md

Calculates the absolute pose error for a given trajectory against a ground truth, plots the results, and saves them to a zip file. Use this to evaluate the accuracy of a single trajectory.

```bash
evo_ape kitti KITTI_00_gt.txt KITTI_00_SPTAM.txt -va --plot --plot_mode xz --save_results results/SPTAM.zip
```

--------------------------------

### Calculate Absolute Pose Error (APE) for ORB-SLAM trajectory

Source: https://github.com/michaelgrupp/evo/blob/master/README.md

Calculates the absolute pose error for a given trajectory against a ground truth, plots the results, and saves them to a zip file. Use this to evaluate the accuracy of a single trajectory.

```bash
mkdir results
evo_ape kitti KITTI_00_gt.txt KITTI_00_ORB.txt -va --plot --plot_mode xz --save_results results/ORB.zip
```

--------------------------------

### Convert Trajectory to Pandas DataFrame

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/pandas_bridge.ipynb

Converts an evo trajectory object into a Pandas DataFrame using the pandas_bridge.trajectory_to_df function.

```python
traj_df = pandas_bridge.trajectory_to_df(traj)
```

--------------------------------

### Scale Correction Only

Source: https://github.com/michaelgrupp/evo/wiki/evo_traj

Apply only scale correction to trajectories using --correct_scale or -s.

```bash
evo_traj tum traj_1.txt traj_2.txt --ref traj_ref.txt --correct_scale
```

--------------------------------

### Align Trajectories and Correct Scale

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics.py_API_Documentation.ipynb

Aligns an estimated trajectory to a reference trajectory using Umeyama's method. Optionally, it can correct only the scale factor.

```python
import copy

traj_est_aligned = copy.deepcopy(traj_est)
traj_est_aligned.align(traj_ref, correct_scale=False, correct_only_scale=False)
```

--------------------------------

### Set Reference Trajectory Axis Marker Scale

Source: https://github.com/michaelgrupp/evo/wiki/Plotting

Sets the scale for coordinate axis markers specifically on the reference trajectory to 0.1.

```bash
evo_config set plot_reference_axis_marker_scale 0.1
```

--------------------------------

### Process RPE Results with Pandas

Source: https://github.com/michaelgrupp/evo/blob/master/notebooks/metrics_interactive.ipynb

This snippet shows how to process the results obtained from the RPE calculation. It uses the pandas library and the `pandas_bridge` module to convert the RPE results into a DataFrame for easier analysis and manipulation. This is useful for aggregating and comparing multiple RPE runs.

```python
import pandas as pd
from evo.tools import pandas_bridge

df = pd.DataFrame()
for result in results:
    df = pd.concat((df, pandas_bridge.result_to_df(result)), axis="columns")
df
```

--------------------------------

### Set Seaborn Plot Style

Source: https://github.com/michaelgrupp/evo/wiki/Plotting

Changes the Seaborn plot style to 'whitegrid' for a white background and grid, suitable for printing.

```bash
evo_config set plot_seaborn_style whitegrid
```

--------------------------------

### Configure ROS Map Unknown Cell Value

Source: https://github.com/michaelgrupp/evo/wiki/Plotting

Set the uint8 value representing unknown cells in a ROS map image. This is useful for correctly masking unknown areas in the plot. The default is 205, but some tools like Cartographer use 128.

```bash
evo_config set ros_map_unknown_cell_value 128
```