serhii.net

Related: 230529-1413 Plants datasets taxonomy

Key info

PlantCLEF

• PlantCLEF 2021 and 2022 summary papers, no doi :(
  • 2021¹:
  • 2022:²
  • other years
• Latest datasets not available, previous ones use eol and therefore are a mix of stuff
• Tasks and datasets differ by year (=can’t reliably do baseline), and main ideas differ too:
  • 2021: use Herbaria to supplement lacking real-life photos
    • Best methods were the ones that used domain-specific adaptations as opposed to simple CNNs
  • 2022: multi-image(/metadata) class. problem with A LOT of classes (80k)
    • Classes mean a lot of gimmicks to handle this memory-wise
• Why this doesn’t work for us:
  • datasets not available!
  • the ones that are are a mix of stuff
  • A lot of methods that work there well are specific to the task, as opposed to the general thing
  • People can use their own datasets for training
  • Metrics: MRR (=not comparable to some other literature, even if there were results on the same dataset)

PlantNet300k³ paper

• Dataset is a representative subsample of the big PlantNet dataaset that “covers over 35K species illustrated by nearly 12 million validated images”
  • Subset has “306,146 plant images covering 1,081 species.
• Long-tailed distribution of classes:
  • 80% of the species account for only 11% of the total number of images”
  • Top1 accuracy is OK, but not meaningful
  • Macro-average top-1 accuracy differs by A LOT
• The paper does a baselines using a lot of networks

Useful stuff

Citizen science

• Citizen science - Wikipedia

  Citizen science (similar to [..] participatory/volunteer monitoring) is scientific research conducted with participation from the general public

  most citizen science research publications being in the fields of biology and conservation

• can mean multiple things, usually using citizens acting volunteers to help monitor/classify/.. stuff (but also citizens initiating stuff; also: educating the public about scientific methods, e.g. schools)

  • allowed users to upload photos of a plant species and its components, enter its characteristics (such as color and size), compare it against a catalog photo and classify it. The classification results are juried by crowdsourced ratings.⁴

Papers

• Paper about using Pl@ntNet⁵ for CS:

  • “Here we present two Pl@ntNet citizen science initiatives used by conservation practitioners in Europe (France) and Africa (Kenya).”

  • paper citing it are interesting: Bonnet: How citizen scientists contribute to monitor… - Google Scholar
  • Pl@ntNet can be
    • limited for subsets of plants
    • limiting plants based on GPS coordinates
    • made to train better certain species by manually adding good examples as done in the Lewa Conservatory in Kenya
• Assessing accuracy in citizen science-based plant phenology monitoring | SpringerLink <@fuccilloAssessingAccuracyCitizen2015 (2015) z>

  • Volunteers demonstrated greatest overall accuracy identifying unfolded leaves, ripe fruits, and open flowers.

  • Maybe we’ll want to compare the areas where people are better at than ML in our paper?
• Similar to the above, but detecting weeds:Assessing citizen science data quality: an invasive species case study <@crallAssessingCitizenScience2011 Assessing citizen science data quality (2011) z>
  • Compare to the paper about detecting weeds with DL: <@chenPerformanceEvaluationDeep2021 (2021) z>

Centralized repositories of stuff

• GBIF (ofc)
• https://bien.nceas.ucsb.edu/bien/ more than 200k observations, and
  • This:

    • Georeferenced plant observations from herbarium, plot, and trait records;
    • Plot inventories and surveys;
    • Species geographic distribution maps;
    • Plant traits;
    • A species-level phylogeny for all plants in the New World;
    • Cross-continent, continent, and country-level species lists.

• No names known to me in their Data contributors

Biodiversity

• Really nice paper: <@ortizReviewInteractionsBiodiversity2021 A review of the interactions between biodiversity, agriculture, climate change, and international trade (2021) z/d>
• TL;DR climate change is not the worst wrt biodiversity

Positioning / strategy

Main bits

• Plant classification as a method to monitor biodiversity in the context of citizen science

Why plant classification is hard

• A lot of cleanly labeled herbaria, few labeled pictures (esp. tropical), but trasferring learned stuff from herbarium sheets to field photos is challenging:

  • (e.g. strong colour variation and the transformation of 3D objects after pressing like fruits and flowers) <@waldchenMachineLearningImage2018 (2018) z>

  • PlantCLEF2021 was entirely dedicated to using herbaria+photos, and there domain adaptations (joint representation space between herb+field) dramatically outperform best classical CNN, esp. the most difficult plants.<@goeau2021overview (2021) z><@goeauAIbasedIdentificationPlant2021 (2021) z>
• Connected to the above: lab-based VS field-based investigations
  • lab-based has strict protocols for aquisition, people with mobile phones don’t

    • “Lab-based setting is often used by biologist that brings the specimen (e.g. insects or plants) to the lab for inspecting them, to identify them and mostly to archive them. In this setting, the image acquisition can be controlled and standardised. In contrast to field-based investigations, where images of the specimen are taken in-situ without a controllable capturing procedure and system. For fieldbased investigations, typically a mobile device or camera is used for image acquisition and the specimen is alive when taking the picture (Martineau et al., 2017). ”<@waldchenMachineLearningImage2018 (2018) z>

• Phenology (growth stages / seasons -> flowers) make life harder
  • Plants sometimes have strong phenology (like bright red flowers) that make it more different and easier to find (esp. here in detecting them in satellite pictures: <@pearseDeepLearningPhenology2021 (2021) z>, but there DL failed less without flowers than non-DL), but sometimes don’t
  • And ofc. a plant with and without flowers looks like a totally different plant
  • Related:
    • Plant growth form has been the most helpful species metadata in PlantCLEF2021, but some plants at different stages of growth look like different plant stages.
• Intra-species variability
  • Plants of the same species can have flowers of different colors: (<@NEURIPS_DATASETS_AND_BENCHMARKS2021_7e7757b1 (2021) z/d>)
    • Esp. if only a part of the plant is photographed
• The Pl@ntNet-300k paper mentions
  • epistemic (model) uncertainty (flowers etc.)
  • aleatoric (data) uncertainty (small information given to make a decision)
    • Plants belonging to the same genus can be visually very similar to each other: (same paper)
• long-tailed distribution, which: <@walkerHarnessingLargeScaleHerbarium2022 (2022) z/d>
  • is representative of RL
  • is a problem because DL is “data-hungry”
• some say there are a lot of mislabeled specimens <@goodwinWidespreadMistakenIdentity2015 (2015) z/d>

Datasets

EDIT separate post about this: 230529-1413 Plants datasets taxonomy

• We can classify existing datasets in two types:

  • Pl@ntNet / iNaturalist? / …: people with phones
  • Clean standardized things like the Plant seedling classification dataset (<@giselssonPublicImageDatabase2017 (2017) z>), common weeds in Denmark dataset <@leminenmadsenOpenPlantPhenotype2020 (2020) z/d> etc.
    • I’d put leaf datasets in this category too
    • FloraIncognita is an interesting case:

      • FloraCapture requests contributors to photograph plants from at least five precisely defined perspectives

• There are some special datasets, satellite and whatever, but especially:

  • Leaf datasets exist and are used surprisingly often (if not exclusively) in overviews like the one we want to do:
    • (pic from <@kayaAnalysisTransferLearning2019 (2019) z/d>)
    • ds-peruvian, as used in <@bermejoComparativeStudyPlant2022 (2022) z/d>
  • Flower datasets / “Natural flower classification”

    • is one of challenging aspects of plant classification because of non-rigid deformation, illumination changes and inter-class similarity. (<@al-qurranPlantClassificationWild2022 Plant classification in the wild (2022) z/d>)

  • Seedlings etc. seem to be useful in industry (and they go hand-in-hand with weed-control)
  • Fruit classification/segmentation and other very specific stuff we don’t really care about (<@mamatAdvancedTechnologyAgriculture2022 Advanced Technology in Agriculture Industry by Implementing Image Annotation Technique and Deep Learning Approach (2022) z/d> has an excellent overview of these)

• Additional info present in datasets or useful:

  • PlantCLEF2021 had additional metadata at the species level: growth form, habitat (forest, wetland, ..), and three others
  • • PlantCLEF2022: 3/5 taxonomic levels where used in various ways. Taxonomic loss is a thing (TODO - was this useful?)⁶
  • Pl@ntNet and FloraIncognita apps (can) use GPS coordinates during the decision
  • TODO Phenology / phenological stage: is this true to begin with?

Research questions similar to ours

Plant classification (a.k.a. species identification) on pictures

• Things like ecology and habitat loss, citizen science etc.
• Industry:
  • Weed detection

Crop identification (sattellites)

Crop stage identification / phenology (sattellites)

Paper outline sketch

Introduction

• Tasks about plants are important
• Ecology: global warming etc., means different distribution of plant species, phenology stages changed, broken balances and stuff and one needs to track it; herbaria and digitization / labeling of herbaria
• Industry: crops stages identification, crops/weeds identification, fruit ripeness identification, etc. long list
• automatic methods have been used, starting from SVM/manual-feature-xxx, later - DL
• DL has been especially nice and improved stuff in all of these different sub-areas, show the examples that compare DL-vs-non-DL in the narrow fields
• The closest relevant thing is PlantCLEF competition that’s really really nice but \textbf{TODO what are we doing that PlantCLEF isn’t?}
• Goal of this paper is:
  1. Do a short overview of the tasks-connected-to-plants that exist and are usually tackled using AI magic
  2. Along the way: WHICH AI magic is usually used for the tasks that are formalized as image classification (TODO and object segmentation/detection?)
  3. Show that while

Goal: Interact with Zotero from within Obsidian
Solution: “Citations”¹ plugin for Obsidian, “Better Bibtex”² plugin for Zotero!

• Creating a local self-updating bibtex export:
  • In Zotero, File->Export, format is “Better Bibtex”
  • this shows an additional checkmark for keeping it autoupdated, check it
  • file is now at the resulting path
• Setting up Obsidian with Citations (in Citations plugin settings):
  • set the path to the one above
  • and the format to BibLaTeX (or it’ll will fail with a generic error)
  • Through the Palette run “Refresh citation database” (and do it every time something changes)
• Operation
  • Search in palette for “Citations”
  • Pandoc format citations³ are the default, but can be changed (almost anything can be changed!)

Neat bits:

• There’s a configurable “Citations: Insert Markdown Citation” thing!

  • My current template:

    <_`@{{citekey}}` {{titleShort}} ({{year}}) [z]({{zoteroSelectURI}})/[d](https://doi.org/{{DOI}})_>
    

  • Legal fields:

    - {{citekey}}
    - {{abstract}}
    - {{authorString}}
    - {{containerTitle}}
    - {{DOI}}
    - {{eprint}}
    - {{eprinttype}}
    - {{eventPlace}}
    - {{page}}
    - {{publisher}}
    - {{publisherPlace}}
    - {{title}}
    - {{titleShort}}
    - {{URL}}
    - {{year}}
    - {{zoteroSelectURI}}
    

• ishnid/zotero-obsidian

• After playing with it for an hour or so: Zotero is freaking awesome
• Sync
  • If you add an account (in Preferences), all groups you’re part of will be synced after you click the sync green thing top-left¹
    • even if you enabled “automatic sync” through Preferences
• Adding/importing
  • Adding by DOI (“Add items by identifier”) is wonderful
    • Works for DOI, ISBN, arxiv ID, and other IDs I never heard of
  • Bibtex “Import from clipboard” is magic for the items that don’t have a DOI/ISBN/… (incl. direct arxiv links!)
    • You can even import multiple papers that way! Neat for copying other .bib files
  • You can add links to files online, and attach local copies of files
    • There’s also “Find available PDF”! Wow!
• notes and highlights; PDF viewer
  • If the item has a (file, not link) PDF, you can highlight inside it!
    • text and rectangle, multiple colors
    • they are synced
    • can be seen and edited in the web version too!
  • Item/file notes
    • can contain citations (= links to other files)!
      • that are clickable!
    • can be created based on existing annotations (=highlights!)
    • support templates²
  • (AWESOME) tutorial: New PDF Reader Available in Zotero 6 | Lane Library Blog
  • PDF reader shortcuts: kb:keyboard shortcuts [Zotero Documentation] (incomplete) and PDF Reader shortcuts (default and missing Zotero-specific ones mentioned in the first link)
• Taxonomy
  • Items have
    • tags (auto-completed; really nice search interface)
    • related items/files
      • symmetrical relationship, being related works both ways
  • Nested collections work as expected (~ nested categories, parent shows all items belonging to all it’s descendants)
  • Advanced search exists and is about as expected
    • good support for ALL THE METADATA FIELDS
    • no support for regex :(
    • you can save useful searches
• Addons/Plugins
  • Info and link to full list: Plug-ins and Integrations
  • Zutilo³ is an addon for macros/shortcuts, found it when looking for a shortcut for “add items by identifier”
    • “Lookup item by identifier”, now <C-N> for me.
    • It’s awesome:
  • “Better Bibtex”⁴ allows exporting as bibtex
    • including automatic export (used in 230507-1620 Zotero and Obsidian)
    • and is EXTREMELY configurable⁵
  • Zotero Robust Links for archiving links with Web Archive and friends
  • Integrations
    • Overleaf Zotero integration is a Premium Overleaf feature
• Other features
  • “Timeline” is neat but not too useful right now:
• Misc
  • (Almost) all keyboard shortcuts (for PDF see above): kb:keyboard shortcuts [Zotero Documentation]
• TODO
  • zotero:// links don’t work for me, and the default .desktop file they provide seems broken - TODO later

Gitstats is the best I know: tomgi/git_stats: GitStats is a git repository statistics generator.

gitstats /path/to/repo /path/to/output/dir

Generates comprehensive static html reports with graphs. Authors, files, times of the day/week/month, ….

4. More Control Flow Tools — Python 3.10.11 documentation:

def http_error(status):
    match status:
        case 400:
            return "Bad request"
        case 404:
            return "Not found"
        case 418:
            return "I'm a teapot"
        case _:
            return "Something's wrong with the internet"

Also

case 401 | 403 | 404:
    return "Not allowed"

and

match points:
    case []:
        print("No points")
    case [Point(0, 0)]:
        print("The origin")
    case [Point(x, y)]:
        print(f"Single point {x}, {y}")
    case [Point(0, y1), Point(0, y2)]:
        print(f"Two on the Y axis at {y1}, {y2}")
    case _:
        print("Something else")

Lastly, you can capture subpatterns:

case (Point(x1, y1), Point(x2, y2) as p2): ...

Generally - #todo - I should systematically read up on new things in the not-latest-anymore Python versions, e.g.:

• Python 3.9: Cool New Features for You to Try – Real Python
• Python 3.10: Cool New Features for You to Try – Real Python
• Python 3.11: Cool New Features for You to Try – Real Python'

TIL Pycharm can automatically reformat files, incl. things like json. The action is “Reformat file”, on my install <C-S-a-L>

If not all files are seen in pycharm project view:

• In the project view, you can pick which files to show: usually it’s all files form the project, but there’s a settings for “only open files” etc. It being on might make it look like not all files are present
• ALSO, there’s the setting for synchronizing file system at each tab change or action in settings/appearance / system changes / synchronize …, enabled by default for me
• There’s an option for “refresh file system”

A typo in a keybinding randomly led me to the graph view in Obsidian, never thought about it - but now apparently I have a lot of notes and it’s quite pretty!

I wanted to remove the #zc tag from graph view to make it clearer (since ALL notes have it basically.)

(177) How to hide tags, but keep notes with them in graph : ObsidianMD mentioned a way to do just that, though I’m not sure I understand it:

-(-path:folder (#tag1 OR #tag2 OR #tag3))

For me that’s:

-(-path:garden/it (#zc OR #zc/it))

# Display all columns and rows:
pd.set_option('display.max_columns', None)
pd.set_option('display.max_rows', None)
# Don't truncate values
pd.set_option('display.max_colwidth', None)

This of course works:

with pd.option_context('display.max_colwidth', None):
  display(df)

Make cells 100% wide in Jupyter:

from IPython.core.display import display, HTML 
display(HTML("<style>.container { width:100% !important; }</style>"))

And https://stackoverflow.com/a/51593236 has this function remarkably similar to the old one I’ve had, except that I changed print->display:

def print_full(x):
    pd.set_option('display.max_rows', None)
    pd.set_option('display.max_columns', None)
    pd.set_option('display.width', 2000)
    pd.set_option('display.float_format', '{:20,.2f}'.format)
    pd.set_option('display.max_colwidth', None)
    #print(x)
	display(x)
    pd.reset_option('display.max_rows')
    pd.reset_option('display.max_columns')
    pd.reset_option('display.width')
    pd.reset_option('display.float_format')
    pd.reset_option('display.max_colwidth')

Pandas convert column to categorial

pd.row_name.astype('category')

Pandas select numeric columns¹:

ds.select_dtypes(include=[np.number])

Pandas divide columns by other column²:

(ds.T / ds.col2.T).T

python - Divide multiple columns by another column in pandas - Stack Overflow

3D plotting in matplotlib: Three-Dimensional Plotting in Matplotlib | Python Data Science Handbook & the official docu: 3D plotting — Matplotlib 3.7.1 documentation

Color palettes

seaborn.color_palette — seaborn 0.12.2 documentation:

sns.color_palette("flare")

# to reverse
sns.color_palette("flare_r")

# to uniform (=not categorical)

sns.color_palette("flare_r",as_cmap=True)

Markers

Properties of Mark objects — seaborn 0.12.2 documentation & matplotlib.markers — Matplotlib 3.7.1 documentation

Really nice relevant tutorial: How to handle time series data with ease — pandas 2.1.0.dev0+658.gc9de03effb documentation

Make pd.Timedelta plottable

sns.boxplot(data=s_dsm_conv, y='Dauer', x='Parameter')

> TypeError: Neither the `x` nor `y` variable appears to be numeric.

pd.TimeDelta is indeed not numeric, but can be made one through

s_dsm_conv['Dauer'] = s_dsm_conv['Dauer'].astype('timedelta64[h]')

Pandas filter by deltas

# Gaps longer than one day
real_gaps=gaps[gaps>pd.Timedelta(1,"d")]