Skip to main content

Fernsicht (Python)

Remote progress tracking over WebRTC. Wrap your loop, get a shareable URL, watch from anywhere.

Install

pip install fernsicht

Quick Start

from fernsicht import blick

# Wrap any iterable
for item in blick(range(10000), desc="Training"):
    process(item)

# Allow up to 4 concurrent viewers
for item in blick(range(10000), desc="Training", max_viewers=4):
    process(item)

# Manual updates
from fernsicht import manual

bar = manual(total=100, desc="Uploading")
for chunk in stream:
    upload(chunk)
    bar.update(len(chunk))
bar.close()

# Context manager
with blick(total=500, desc="Epochs") as bar:
    for batch in data_loader:
        train(batch)
        bar.update(len(batch))

A URL like https://.../#room=<id>&role=viewer is printed when the bar starts. Open it on your phone or another machine to watch live progress including elapsed time, ETA, rate, and item count.

Configuration

No configuration is required for the default hosted node — the SDK points at https://signal.fernsicht.space out of the box.

Self-hosting

To use your own signaling server:

export FERNSICHT_SERVER_URL="https://your-signal-domain"

Authenticated session endpoint

If your server requires an API key on POST /session:

export FERNSICHT_SESSION_API_KEY="your-api-key"

Viewer limits

  • Default is max_viewers=8 (matches the R SDK and CLI defaults).
  • Pass max_viewers=1 for a single-viewer-only room, or any value up to the server's MAX_VIEWERS_PER_ROOM cap.
  • The server enforces an upper bound via MAX_VIEWERS_PER_ROOM.

How it works

  1. blick() calls POST /session on the signaling server, which creates a new room and returns a sender_secret.
  2. A background thread polls GET /poll/{room}?secret=… every ~25 seconds to check for viewer connections.
  3. When a viewer opens the link, it posts a WebRTC offer; the sender thread picks it up, creates an answer, and completes the handshake.
  4. Progress data flows directly over the WebRTC DataChannel — the server is never in the data path.

No persistent connection is held by the SDK. Between polls, the background thread is idle.

Other ways to use Fernsicht

Don't want to add a blick() import? Two alternatives:

  • CLI — fernsicht run -- python train.py wraps any command and auto-detects tqdm/[N/M]/percent-style progress with no code change. See cli/README.md.
  • R — remotes::install_github("MuteJester/Fernsicht", subdir="publishers/r") for the R sibling SDK.

All three speak the same wire protocol; rooms / viewer URLs / hosted defaults are identical.

Metadata

Release files for fernsicht 0.1.2

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for fernsicht 0.1.2
File Size Uploaded
fernsicht-0.1.2.tar.gz 28.4 kB Details

Built distribution (wheel)

Table of built distributions (wheels) for fernsicht 0.1.2
File Interpreter ABI Platform
fernsicht-0.1.2-py3-none-any.whl Python 3 none any Details

Total release size: 54.0 kB

Release files / fernsicht-0.1.2.tar.gz

Download URL fernsicht-0.1.2.tar.gz
Size 28.4 kB
Tags Source
SHA-256 checksum
How to use checksums
55554f91d8e5b87940e9aefd7ae6de0d4b4617a28cf5182c71547a8f3b88841b
BLAKE2b-256 checksum
How to use checksums
adaf69b40c057d13ccd8692c13957385ae2061267513c85bfa8b5decd4699551
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.12.9

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Apr 23, 2026.

Transparency log

Release files / fernsicht-0.1.2-py3-none-any.whl

Download URL fernsicht-0.1.2-py3-none-any.whl
Size 25.6 kB
Tags Python 3
SHA-256 checksum
How to use checksums
78b611c7b7084f6344edb0abc1fcb280907ec0b006c389c2c723f95fa7b12371
BLAKE2b-256 checksum
How to use checksums
55fd7c023a8c32ea954986163cf9cd70156144aa38d1eb8e04df443c8591a844
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.12.9

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Apr 23, 2026.

Transparency log

Release history Release notifications | RSS feed

This release

0.1.2 This release

2 release files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page