{"library":"pykalman","title":"PyKalman","description":"PyKalman is an active Python library providing robust implementations of the Kalman Filter, Kalman Smoother, and the Expectation-Maximization (EM) algorithm. It is designed for state estimation in linear dynamical systems and parameter learning, with recent updates ensuring compatibility with modern Python and NumPy versions. The library maintains a steady release cadence with regular bug fixes and maintenance updates.","language":"python","status":"active","last_verified":"Mon May 18","install":{"commands":["pip install pykalman"],"cli":null},"imports":["from pykalman import KalmanFilter","from pykalman import UnscentedKalmanFilter"],"auth":{"required":false,"env_vars":[]},"quickstart":{"code":"import numpy as np\nfrom pykalman import KalmanFilter\n\n# Define the parameters of a simple linear system\ntransition_matrices = [[1, 1], [0, 1]] # State transition matrix\nobservation_matrices = [[1, 0]]       # Observation matrix (we observe position only)\ntransition_covariance = np.eye(2) * 0.1 # Small process noise\nobservation_covariance = np.eye(1) * 1.0 # Larger observation noise\ninitial_state_mean = [0, 1]           # Initial state (position=0, velocity=1)\ninitial_state_covariance = np.eye(2) * 1.0 # Initial uncertainty\n\n# Create a Kalman Filter instance\nkf = KalmanFilter(\n    transition_matrices=transition_matrices,\n    observation_matrices=observation_matrices,\n    transition_covariance=transition_covariance,\n    observation_covariance=observation_covariance,\n    initial_state_mean=initial_state_mean,\n    initial_state_covariance=initial_state_covariance\n)\n\n# Generate some noisy 'measurements' for a simple linear motion\nn_timesteps = 50\ntrue_states = np.zeros((n_timesteps, 2))\nmeasurements = np.zeros((n_timesteps, 1))\ntrue_states[0] = initial_state_mean\nfor t in range(1, n_timesteps):\n    true_states[t] = np.dot(transition_matrices, true_states[t-1]) + np.random.multivariate_normal([0, 0], transition_covariance)\n    measurements[t] = np.dot(observation_matrices, true_states[t]) + np.random.normal(0, np.sqrt(observation_covariance[0,0]))\n\n# Use the Kalman Filter to estimate the states from measurements\n(filtered_state_means, filtered_state_covariances) = kf.filter(measurements)\n(smoothed_state_means, smoothed_state_covariances) = kf.smooth(measurements)\n\nprint(\"First 5 filtered state means (position, velocity):\")\nprint(filtered_state_means[:5])\nprint(\"\\nFirst 5 smoothed state means (position, velocity):\")\nprint(smoothed_state_means[:5])","lang":"python","description":"This quickstart demonstrates the basic usage of `KalmanFilter` to estimate states from noisy measurements. It defines a simple linear system, generates synthetic data, and then applies the `filter` and `smooth` methods to estimate the underlying states.","tag":null,"tag_description":null,"last_tested":null,"results":[]},"compatibility":{"tag":null,"tag_description":null,"last_tested":"2026-05-18","installed_version":"0.10.2","pypi_latest":"0.11.2","is_stale":true,"summary":{"python_range":"3.10–3.9","success_rate":100,"avg_install_s":7.6,"avg_import_s":1.38,"wheel_type":"wheel"},"results":[{"runtime":"python:3.10-alpine","python_version":"3.10","os_libc":"alpine (musl)","variant":"pykalman","exit_code":0,"wheel_type":"wheel","failure_reason":null,"import_side_effects":"clean","install_time_s":null,"import_time_s":1.21,"mem_mb":26,"disk_size":"233.8M"},{"runtime":"python:3.10-slim","python_version":"3.10","os_libc":"slim (glibc)","variant":"pykalman","exit_code":0,"wheel_type":"wheel","failure_reason":null,"import_side_effects":"clean","install_time_s":7.2,"import_time_s":0.98,"mem_mb":26,"disk_size":"225M"},{"runtime":"python:3.11-alpine","python_version":"3.11","os_libc":"alpine (musl)","variant":"pykalman","exit_code":0,"wheel_type":"wheel","failure_reason":null,"import_side_effects":"clean","install_time_s":null,"import_time_s":1.53,"mem_mb":26.9,"disk_size":"249.3M"},{"runtime":"python:3.11-slim","python_version":"3.11","os_libc":"slim (glibc)","variant":"pykalman","exit_code":0,"wheel_type":"wheel","failure_reason":null,"import_side_effects":"clean","install_time_s":7.1,"import_time_s":1.45,"mem_mb":26.9,"disk_size":"238M"},{"runtime":"python:3.12-alpine","python_version":"3.12","os_libc":"alpine (musl)","variant":"pykalman","exit_code":0,"wheel_type":"wheel","failure_reason":null,"import_side_effects":"clean","install_time_s":null,"import_time_s":1.62,"mem_mb":26.6,"disk_size":"235.3M"},{"runtime":"python:3.12-slim","python_version":"3.12","os_libc":"slim (glibc)","variant":"pykalman","exit_code":0,"wheel_type":"wheel","failure_reason":null,"import_side_effects":"clean","install_time_s":7.8,"import_time_s":1.82,"mem_mb":26.6,"disk_size":"224M"},{"runtime":"python:3.13-alpine","python_version":"3.13","os_libc":"alpine (musl)","variant":"pykalman","exit_code":0,"wheel_type":"wheel","failure_reason":null,"import_side_effects":"clean","install_time_s":null,"import_time_s":1.49,"mem_mb":26.4,"disk_size":"234.0M"},{"runtime":"python:3.13-slim","python_version":"3.13","os_libc":"slim (glibc)","variant":"pykalman","exit_code":0,"wheel_type":"wheel","failure_reason":null,"import_side_effects":"clean","install_time_s":7.5,"import_time_s":1.64,"mem_mb":26.4,"disk_size":"223M"},{"runtime":"python:3.9-alpine","python_version":"3.9","os_libc":"alpine (musl)","variant":"pykalman","exit_code":0,"wheel_type":"wheel","failure_reason":null,"import_side_effects":"clean","install_time_s":null,"import_time_s":1.08,"mem_mb":22.5,"disk_size":"235.9M"},{"runtime":"python:3.9-slim","python_version":"3.9","os_libc":"slim (glibc)","variant":"pykalman","exit_code":0,"wheel_type":"wheel","failure_reason":null,"import_side_effects":"clean","install_time_s":8.2,"import_time_s":0.98,"mem_mb":23.6,"disk_size":"232M"}]}}