Skip to main content

Client for

Project description


Collective microprediction client leveraging

pip install microprediction 

Read client

It is possible to retrieve most quantities at with direct web calls such as For example:

import requests
lagged_values = requests.get('').json()
lagged        = requests.get('').json()

However the reader client adds a little convenience.

from microprediction import MicroReader
mr = MicroReader()

summary       = mr.get_summary('cop.json')
current_value = mr.get('cop.json')
lagged_values = mr.get_lagged_values('cop.json') 
lagged_times  = mr.get_lagged_times('cop.json')
delayed       = mr.get_delayed('cop.json',delay=70)

Your best reference is the code

Write client

The write client is used to submit predictions or to create a data stream.

Submitting predictions

To predict a data stream at is to supply a collection of scenarios. These scenarios are quarantined for different horizons (see delays parameter at ). When the data is updated by the stream owner, rewards are calculated. People and machines making accurate probabilistic forecasts will see their balances (at rise.

Step 1: Obtaining a write_key (

Click on to create a write_key. Hash memorable keys are explained at

Step 2: Instantiate a writer

from microprediction import MicroWriter
mw = MicroWriter(write_key=)    # Sub in your own write_key 

Step 3: Submitting scenarios

scenarios = [ i*0.001 for i in range(mw.num_predictions) ] 

There is no difference when predicting regular streams and derived streams. For example to predict the implied z-score:

my_scenarios = sorted(list(np.random.randn(mw.num_predictions))
mw.submit(name="z1~airp-06820.json", write_key="ce169feeb3565b282d50a850dc62e0db", values = my_scenarios, delay=15)

Step 4: Examine performance

Visit leaderboards such as or look across all streams with:


Submitting data to be predicted

You can also use the writer to create a stream of live data that clever algorithms and humans can predict.

mw = MicroWriter(write_key=write_key)

However there is a higher barrier to entry...

Step 1: Obtaining a rare write_key

To create a new stream you need:

muid.difficulty(write_key)  >  official minimum difficulty     # 12 at time of writing

See for the current values of min_len, which is the official minimum difficulty to create a stream. To mine for write_keys with this property you can cut and paste this bash command into terminal:

/bin/bash -c "$(curl -fsSL"

or use the MUID library ( ...

$pip install muid
>>> import muid
>>> muid.mine(skip_intro=True)

See or for more on MUIDs. Use a URL like to reveal the hidden "spirit animal" in a MUID. The difficulty is the length of the animal, not including the space.

Step 2: Updating the current value

To create a new live data source or update its value:

prctl = mw.put(name='mystream.json',value=3.14157) 

By default this returns a percentile so you know how surprising the data point is, relative to the CDF of predictions made by others at some time in the past.

Step 3: Retrieve the distribution of future values

You can see what others think about the future of your data as follows:

 cdf = mw.get_cdf('cop.json',delay=mr.delays[0],values=[0,0.5])

where the delay parameter, in seconds, acts as a prediction horizon. This call will reveal the probability that your future value is less than 0.0, and the probability that it is less than 0.5.

Step 4: Hope that your write_key does not go broke

When you create a stream you automatically participate in the prediction of the stream. A benchmark empirical sampling algorithm with some recency adjustment is used for this purpose. If nobody can do a better job that this, your write_key balance will generally neither rise nor fall.

However once smart people and algorithms enter the fray, you can expect this default model to be beaten and the balance on your write_key to trend downwards. On an ongoing basis you also need the write_key balance not to fall below a threshold bankruptcy level. The minimum balance for a key of difficulty 8 is also found at At time of writing, and assuming this parameter is -10, we have:

write_key difficulty bankruptcy write_key difficulty bankruptcy
8 -10 11 -40,960
9 -160 12 -655,360
10 -2,560 13 -10,485,760

Higher dimensional prediction (copulas, Z-curves)

Advanced functionality is available to those with write_keys of difficulty 1 more than the stream minimum. More details to follow on that.

Stream name rules

  • Must end in .json
  • Must contain only alphanumeric, hyphens, underscores, colons (discouraged) and at most one period.
  • Must not contain double colon.




which are also available directly. For example:

error_log = requests.get('').json()
error_log = requests.get('').json()

Project details

Release history Release notifications | RSS feed

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for microprediction, version 0.2.2
Filename, size File type Python version Upload date Hashes
Filename, size microprediction-0.2.2-py3-none-any.whl (25.1 kB) File type Wheel Python version py3 Upload date Hashes View
Filename, size microprediction-0.2.2.tar.gz (19.4 kB) File type Source Python version None Upload date Hashes View

Supported by

Pingdom Pingdom Monitoring Google Google Object Storage and Download Analytics Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page