Frames represent a product’s state at a moment or over an interval in time. Once you have a frame, you can retrieve any of its datapoints or tiles.

Frame Requests

Basic Usage

Here’s how to request the latest frames for a product.

from skywiseplatform import Product

product = Product.find('skywise-24hr-precip-analysis')
frames  = product.frames()

If you want to retrieve frames over a date range, use the start/end keyword arguments.

frames = product.frames(start='2015-05-01', end='2015-05-31')

date/datetime keyword arguments in `skywise-platform` can be anything parseable by the arrow library.

Forecast Frames

Products with forecasts link all frames to a particular forecast. The frames() method on a forecast product will only return frames for the latest forecast. If you’d like to retrieve frames for a different forecast, you’ll first need to retrieve your forecast of interest using the product’s forecasts() method. This will return all active forecasts. Here’s an example of retrieving frames for the latest forecast as well as the oldest forecast:

forecast_product = Product.find('weatherops-24hr-precip-forecast')

# Retrieving Frames using Product.frames()
frames = forecast_product.frames()

# Is synonymous with retrieving frames for the latest forecast
latest_forecast = product.forecasts().pop()
frames = latest_forecast.frames()

# Retrieve Forecasts (sorted from oldest to latest)
oldest_forecast = product.forecasts().pop(0)

# Retrieve Frames from a Specific Forecast
frames = oldest_forecast.frames()


A datapoint represents the value of a product’s frame at a particular point on the globe. This value is derived from the highest resolution tile that contains the point. Using datapoints saves you from the hassle of pulling down the tile containing your point and inspecting the value for yourself.

>>> import arrow
>>> from skywiseplatform import Product
>>> product = Product.find('skywise-24hr-high-temperature-analysis')
>>> frames = product.frames()
>>> datapoints = [frame.datapoint(35.46, -97.52) for frame in frames]
>>> for dp in datapoints:
...     print "Frame %s - %s - %.3f %s" % (, dp.validTime, dp.value, dp.unit['label'])
Frame 1f421458-3498-4193-b9c9-88874102ca25 - 2016-09-14 00:00:00+00:00 - 32.487 °C
Frame a90e6a01-2935-4f37-a7c2-825a0d4b5bd6 - 2016-09-15 00:00:00+00:00 - 27.084 °C
Frame 907714bd-f33a-4707-aa1a-8aad5ed0cd7a - 2016-09-16 00:00:00+00:00 - 28.496 °C
Frame 188bf605-f07d-4322-8e9c-ac7b356a52ea - 2016-09-17 00:00:00+00:00 - 28.560 °C
Frame 35ec9859-1e43-4b25-85b9-9ff0d2419ffa - 2016-09-18 00:00:00+00:00 - 29.350 °C
Frame c0e34b8d-0367-4b7d-83e8-c6e6fdea11ee - 2016-09-19 00:00:00+00:00 - 31.525 °C
Frame 86d5a13d-119a-4a49-a48d-cf8df5f7ba56 - 2016-09-20 00:00:00+00:00 - 35.735 °C
Frame 2fe7432d-b33b-494d-91f0-9d1015114db9 - 2016-09-21 00:00:00+00:00 - 34.587 °C
Frame 5c2b8012-262f-43e1-984c-deb8b613b511 - 2016-09-22 00:00:00+00:00 - 31.879 °C
Frame 003e4208-fd80-4561-a0e0-83af52a6273b - 2016-09-23 00:00:00+00:00 - 32.031 °C


You can request tiles for a frame using either a Google Maps XYZ-coordinate or with a Bing Maps quadkey. Here’s how to create a stack of tiles at a particular coordinate for your latest frames:

from skywiseplatform import Product

# Get tiles for the latest frames
product = Product.find('skywise-24hr-high-temperature-analysis')

# Google Maps
tiles = [frame.tile(x=0, y=0, z=1) for frame in product.frames()]

# Bing Maps
tiles = [frame.tile(quadkey="0") for frame in product.frames()]

Once you’ve received the tile, you can retrieve its contents using the content() method.

tile = tiles.pop()
with open('my_tile.tiff', 'w') as f:


You can specify a style for your tile requests using either a style id or Style object:

# Grab the latest frame
frame = product.frames().pop()

# Use a Style object
style = product.styles().pop()
stylized_tile = frame.tile(x=0, y=0, z=1, style=style)

# Use a style id
stylized_tile = frame.tile(x=0, y=0, z=1, style="my-style-id")

Media Types

You can inspect which media types are available for your frame using the mediaTypes attribute:

>>> frame = product.frames().pop()
>>> frame.mediaTypes
["image/jpeg", "image/png", "image/tiff"]

By default, tile requests will use “image/tiff” if it is available. You can also specify any of the media types supported by the frame:

>>> png = frame.tile(x=0, y=0, z=1, media_type='image/png')
>>> jpg = frame.tile(x=0, y=0, z=1, media_type='image/jpeg')


If you’re needing to make a large number of tile or datapoint calls, requesting them one at a time will most likely be too slow. You can fire off batches of tile calls using the async methods provided for both datapoints and tiles in conjunction with map_async():

from skywiseplatform import Product, map_async
product = Product.find('skywise-24hr-high-temperature-analysis')
frames = product.frames()

# Request Multiple Tiles at Once
tile_batch = [frame.tile_async(x=0, y=0, z=1) for frame in product.frames()]
tiles = map_async(tile_batch)

# Request Multiple Datapoints at Once
dp_batch = [frame.datapoint_async(35.46, -97.52) for frame in frames]
datapoints = map_async(dp_batch)