Trapper workflow
a tutorial by wsyxbcl
based on the official tutorial of trapper-project.
Overview
Terminology
We'll stick with these concepts throughout the whole workflow. They are designed to match Trapper but also tweaked to fit historical data and standards in Shanshui, so it's more like a pragmatic adaptation.
- Project
- Deployment
- Location
- Resource
- Collection
- Classificator
Tools
There are a couple tools we use throughout the workflow which are worth introducing here.
- trap_info
- Maze
- Trapper
- Serval
- Emberiza
- Ptilopsis
- trapper-client
- digikam
Miscs
Due to historical reasons and the strong insistence of my colleagues, the species list will contaminated by a few specific tags, check Species tagging convention for details.
Data wrangling
The data wrangling includes checking trap_info, renaming files, generating deployment table and tag analysis (for historical data), check Data wrangling for details.
The ensuing steps rely on the directory layout below:
.
├─ project_name
| ├─ collection_name_1
| | ├─ deployment_id_1
| | | ├─ filename_1
| | | ├─ filename_2
| | | └─ filename_3
| | ├─ deployment_id_2
| | | ├─ filename_4
| | | ├─ filename_5
| | | └─ filename_6
| | └─ ...
| ├─ collection_name_2
| | ├─ deployment_id_3
| | | ├─ ...
| | └─ ...
| └─ ...
└─ deployments.csv
Configure trapper
The trapper web interface
Register and login your trapper account at trapper, and you are ready to go.
Note
Maze is not open to to public, make you're on the same network (specifically the WLAN named 'Maze' or 'shanshui' if you're in the Beijing office) to connect to the Trapper server.
Note
There will be a security warning from your browser as we use a self-signed SSL signature, you can safely ignore it at this moment.
The trapper-client
Skip this section if you are using Serval for data upload.
Follow this section in the official trapper tutorial to download and setup trapper-client.
In the advanced settings, disable Verify SSL,
and because we use FTP for large package update, don't forget to configure the FTP account mentioned above.
The FTP host is the same with the trapper server, i.e. 10.134.1.13,
and you can use the username and the password of your trapper account to login.
Note
Just a reminder, make sure that you select the correct Project in the Settings tab before making next move,
or trapper will not find the corresponding deployments.
FYI., trapper-client save user's password in an unencrypted way,
so I would strongly suggest not including your password when "Save Settings"
and manually entering the password every time you use trapper-client instead.
Create a research project
Projects in trapper are already created by admins, skip this part if you're working on existing projects. If this is not the case, feel free to add your own project by following the official guide.
Upload resource to trapper server
Use trapper-client
Get deployment tables
Create data packages
Upload
Use Serval
Upload classifications to trapper
First, create classificator and classification project, by following this section of the official tutorial.