Add common name of birds

README.md

@@ -63,23 +63,19 @@ These are saved as parquet files and are checked into the repository.
 
 ```bash
 python -m birdcall_distribution.commands.model_assets intercept_car data/ee_v3_americas_5.parquet data/processed/models/intercept_car/americas/5 --n-species 10 --cores 4 --samples 5000
-
 python -m birdcall_distribution.commands.model_assets intercept_car data/ee_v3_western_us_2.parquet data/processed/models/intercept_car/western_us/2 --n-species 10 --cores 4 --samples 5000
-
 python -m birdcall_distribution.commands.model_assets intercept_car data/ee_v3_ca_1.parquet data/processed/models/intercept_car/ca/1 --n-species 10 --cores 4 --samples 5000
 
-
 python -m birdcall_distribution.commands.model_assets intercept_covariate_car data/ee_v3_americas_5.parquet data/processed/models/intercept_covariate_car/americas/5 --n-species 10 --cores 4 --samples 5000
-
 python -m birdcall_distribution.commands.model_assets intercept_covariate_car data/ee_v3_western_us_2.parquet data/processed/models/intercept_covariate_car/western_us/2 --n-species 10 --cores 4 --samples 5000
-
 python -m birdcall_distribution.commands.model_assets intercept_covariate_car data/ee_v3_ca_1.parquet data/processed/models/intercept_covariate_car/ca/1 --n-species 10 --cores 4 --samples 5000
 ```
 
 We also generate the manifest:
 
 ```bash
 python -m birdcall_distribution.commands.generate_manifest data/processed data/processed/manifest.json
+python -m birdcall_distribution.commands.bird_name_mapping data/processed data/processed
 ```
 
 ```bash

app/src/routes/+page.svelte

@@ -13,6 +13,7 @@
     "https://storage.googleapis.com/iyse6420-birdcall-distribution/processed";
 
   let manifest = [];
+  let species_mapper = {};
   let model = "intercept_covariate_car";
   let region = "americas";
   let specie = null;
@@ -22,8 +23,10 @@
   let should_show = true;
 
   onMount(async () => {
-    const res = await fetch(`${url}/manifest.json`);
+    let res = await fetch(`${url}/manifest.json`);
     manifest = await res.json();
+    res = await fetch(`${url}/species_mapping.json`);
+    species_mapper = await res.json();
   });
 
   $: models = uniq(manifest.map((item) => item.model));
@@ -43,6 +46,7 @@
     fetch(`${url}/${selected.path}/${selected.traces.trace}`)
       .then((res) => res.json())
       .then((data) => (trace = [...data]));
+  $: common_name = specie ? species_mapper[specie] : null;
   // NOTE: we don't really need the ppc data atm, but we could use it to show a
   // histogram or something
 
@@ -59,15 +63,30 @@
 
 <h3>Options</h3>
 
-<ModelOptions {models} {regions} {species} bind:model bind:region bind:specie />
+<ModelOptions
+  {models}
+  {regions}
+  {species}
+  {species_mapper}
+  bind:model
+  bind:region
+  bind:specie
+  bind:should_show
+/>
 
 {#if selected}
   <h3>
-    {specie},
+    {common_name} (<a href="https://ebird.org/species/{specie}">{specie}</a>),
     {region},
-    {selected.grid_size} degree resolution,
+    {selected.grid_size}&deg; resolution,
     {trace.filter((x) => x.index.includes("phi[")).length} cells
   </h3>
+
+  <p>
+    Find more information about the {common_name} on its
+    <a href="https://ebird.org/species/{specie}">eBird page.</a>
+  </p>
+
   <div class="primary">
     <h4>linear scale</h4>
     <div>
@@ -90,12 +109,21 @@
 
   <h3>options</h3>
 
-  <ModelOptions {models} {regions} {species} bind:model bind:region bind:specie bind:should_show />
+  <ModelOptions
+    {models}
+    {regions}
+    {species}
+    {species_mapper}
+    bind:model
+    bind:region
+    bind:specie
+    bind:should_show
+  />
 
   <h3>
-    {specie},
+    {common_name} (<a href="https://ebird.org/species/{specie}">{specie}</a>),
     {region},
-    {selected.grid_size} degree resolution,
+    {selected.grid_size}&deg; resolution,
     {trace.filter((x) => x.index.includes("phi[")).length} cells
   </h3>
 

app/src/routes/ModelOptions.svelte

@@ -2,6 +2,7 @@
   export let models = [];
   export let regions = [];
   export let species = [];
+  export let species_mapper = {};
   export let model;
   export let region;
   export let specie;
@@ -37,7 +38,7 @@
     {#each species as item}
       <label>
         <input type="radio" bind:group={specie} value={item} />
-        {item}
+        {species_mapper[item]}
       </label>
     {/each}
   </div>

birdcall_distribution/commands/bird_name_mapping.py

@@ -0,0 +1,46 @@
+import json
+from argparse import ArgumentParser
+from functools import partial
+from pathlib import Path
+
+import matplotlib.pyplot as plt
+import numpy as np
+import pandas as pd
+import tqdm
+
+from birdcall_distribution.geo import get_grid_meta, get_modis_land_cover_name
+from birdcall_distribution.plot import dataframe_color_getter, plot_grid
+
+
+def parse_args():
+    """Parse args for the input and output paths"""
+    parser = ArgumentParser()
+    parser.add_argument("input", type=str, help="Path to the input dataset")
+    parser.add_argument("output", type=str, help="Path to the output directory")
+    return parser.parse_args()
+
+
+def main():
+    args = parse_args()
+    input_path = Path(args.input)
+    output_path = Path(args.output)
+
+    taxonomy_df = pd.read_csv(
+        "https://storage.googleapis.com/birdclef-eda-f22/data/raw/birdclef-2022/eBird_Taxonomy_v2021.csv"
+    )
+    taxonomy_df.columns = taxonomy_df.columns.str.lower()
+    species_mapper = taxonomy_df.set_index("species_code")["primary_com_name"].to_dict()
+
+    # read manifest file
+    with open(input_path / "manifest.json") as f:
+        manifest = json.load(f)
+
+    # get all the set of all species
+    species = set([row["primary_label"] for row in manifest])
+    mapper_subset = {k: v for k, v in species_mapper.items() if k in species}
+    with open(output_path / "species_mapping.json", "w") as f:
+        json.dump(mapper_subset, f, indent=2)
+
+
+if __name__ == "__main__":
+    main()

notebooks/2022-12-04-quail.ipynb

@@ -1921,7 +1921,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.5"
+   "version": "3.10.5 (tags/v3.10.5:f377153, Jun  6 2022, 16:14:13) [MSC v.1929 64 bit (AMD64)]"
   },
   "orig_nbformat": 4,
   "vscode": {