Subject: Marine Geospatial Ecology Tools (MGET) help
Text archives
| From: | Jason Roberts <> |
|---|---|
| To: | Raquel Romero Chaves <> |
| Cc: | "" <> |
| Subject: | RE: [mget-help] Trouble with Predict GAM from Rasters tool |
| Date: | Fri, 26 Jun 2015 02:20:41 +0000 |
| Accept-language: | en-US |
| Authentication-results: | gmail.com; dkim=none (message not signed) header.d=none; |
|
Hi Raquel, Thanks for your interest in MGET and your detailed description of the problem. I can help you with it, but I’m travelling and look at this in detail until next
week. In the meantime, could you please: 1.
Enable verbose logging (instructions
here). 2.
Run the Fit GAM tool again and send me the complete output. Copy it into an email; there will be a lot of it. 3.
Then run Predict GAM From Rasters with the parameters that produce the UnboundLocalError and send me that output. 4.
If you like, you could run Predict GAM From Rasters again using the switched ONI and Area and send that. I’m more concerned about the UnboundLocalError,
but also interested in why the second way of doing it did not work for you. I’ll look at the output and can then offer some advice. (You can disable verbose logging by just deleting the Logging.ini file.) Best regards, Jason From: Raquel Romero Chaves [mailto:]
Good Afternoon, This is my data: we have 20 stations distributed along the pacific cost of costa rica, they were sampled once a year by trawl fishing and got the fish data for each trawl made. This was done for only 4 years.
The problem comes here, when using Predict GAM from raster for the year 2008. At the "Constant predictors" box I have to include the Area and ONI with their values. If i include area first and write down "1" at the value section (like
in the webinar by Jaso R, although i dont really know if that is correct) and then add the ONI category with "Normal" (that year had no niño or niña event) i get this response: "UnboundLocalError: local variable 'value' referenced before assignment". And
if i include ONI "Normal" first and Area "1" second, the tool does not recognize the ONI data and it says that ONI was not taken into account for that prediction, and then creates a raster BUT i can not see it and have a single wrong value. Raquel |