Merge branch 'main' of https://github.com/via-cs/birdmig

Author: mihikakrishna

.gitignore

@@ -47,4 +47,5 @@ model/outputs/models/Anser_albifrons_et_classifier_model.pkl
 # Ignore all TeX files except .tex and .pdf
 /techdoc/*
 !techdoc/tech-doc.pdf
-!techdoc/tech-doc.tex
+!techdoc/tech-doc.tex
+backend/app/base.py

README.md

@@ -1,8 +1,12 @@
 # Usage
 
 ## To start the Flask Backend:
-- cd into /birdmig/backend/app
-- enter ```uvicorn base:app --reload``` in your terminal
+- ```cd``` into ```/birdmig/backend/app```
+- Run the backend
+  - To run the backend for showcase and documentation purposes, enter ```fastapi dev base.py``` in your terminal.
+    - Access docs by navigating to ```localhost:8000/docs``` on your web browser
+  - To run the backend for debugging purposes, run ```uvicorn base:app --reload``` in your terminal.
+    - Note that this only runs the backend. It can not be navigated to on the web browser, nor will documentation be available.
 
 ## To start the React Frontend:
 

backend/.flaskenv

@@ -27,12 +27,8 @@ COPY app/data/ /code/data/
 
 COPY app/models/ /code/models/
 
-# Set environment variables
-ENV FLASK_APP=app/base.py
-ENV FLASK_RUN_HOST=0.0.0.0
-
 # Expose the port the app runs on
 EXPOSE 8000
 
 # Command to run the application
-CMD ["flask", "run"]
+CMD ["fastapi", "dev", "base.py"]

backend/Dockerfile

@@ -1,11 +1,3 @@
-# Data format:
-- ```info```: string
-- ```sdmData```: x-y coordinate pair.
-- ```timeSeriesData```: array of time-series values for climate variables
-  - ```precipiation```
-  - ```climate```
-  - ```temperature```
-
 # Running Tests:
 Write all tests for the backend in the ```tests``` directory.
 

backend/README.md

@@ -1,4 +1,3 @@
-import uvicorn
 from fastapi import FastAPI, HTTPException
 from fastapi.middleware.cors import CORSMiddleware
 from fastapi.responses import FileResponse, JSONResponse
@@ -9,19 +8,12 @@
 
 import os
 
-#from .config import AppConfig
 import pandas as pd
 import numpy as np
-from shapely.geometry import LineString
-import rasterio
-from rasterio.warp import calculate_default_transform, reproject, Resampling
 
 from PIL import Image
 from io import BytesIO
 import base64
-# DEBUGGING
-import sys
-from time import sleep
 
 app = FastAPI()
 frontend_origin = "http://localhost:3000"
@@ -184,33 +176,26 @@ def get_precipitation_data(selectedYear: int):
     return monthly_avg.to_dict(orient='records')
 
 
-'''app.get('/bird-data/{bird_name}')
-def get_bird_data(bird_name):
-  if bird_name in bird_data:
-    return bird_data'''
-      
-
 class PredictionInputs(BaseModel):
   bird: str
   year: int
   emissions: str
 
 @app.put('/prediction')
-async def predict(prediction_input: PredictionInputs):
+async def get_predictions(prediction_input: PredictionInputs):
   selected_bird = prediction_input.bird
   selected_year = str(prediction_input.year)
   emission_Type = prediction_input.emissions
-
   # For image data:
   output_path = f"../../model/outputs/png-images/{birdsModelDirs[selected_bird]}/{emission_Type}/{selected_year}.png"
   dataImg = Image.open(output_path)
   buffer = BytesIO()
   dataImg.save(buffer, format="png")
 
-  # Artificial delay to simulate model prediction time
-  #sleep(2.5)
+  # FastAPI requests can handle asynchronous predictions.
+  # In practice, this would be waiting for a machine learning model to generate
+  # predictions. In this version, waiting wasn't necessary, so it returns immediately.
 
-  #If we'll need to encapsulate a file, use this:
   return {
     "prediction": base64.b64encode(buffer.getvalue()).decode(),
     "resFormat": dataImg.format
@@ -230,8 +215,8 @@ def get_bird_info(bird_name):
     raise HTTPException(
       status_code= 404,
       detail= f"data for bird {bird_name} does not exist.")
-      
-
+  
+  
 @app.get('/json/{filename}')
 def send_json(filename):
   climate_file_loc = os.path.join('climate_data/json_data', filename)
@@ -244,6 +229,7 @@ def send_json(filename):
 
 @app.get('/get_trajectory_data')
 def get_trajectory_data(bird: str, birdID: str):
+    
   filename = f'./data/{bird}.csv'
   try:
     df = pd.read_csv(filename)
@@ -253,77 +239,29 @@ def get_trajectory_data(bird: str, birdID: str):
     if bird_data.empty:
       raise HTTPException(
         status_code= 404,
-        details= 'No trajectory data found for given bird ID')
+        detail= 'No trajectory data found for given bird ID')
 
     # Convert data to dictionary format
     trajectory_data = bird_data[['LATITUDE', 'LONGITUDE', 'TIMESTAMP']].to_dict(orient='records')
     return trajectory_data
   except FileNotFoundError:
     raise HTTPException(
       status_code= 404,
-      details= f'CSV file for {bird} not found')
+      detail= f'CSV file for {filename} not found')
 
 
 @app.get('/get_bird_ids')
 def get_bird_ids(bird: str):
   filename = f'./data/{bird}.csv'
+  
   try:
     df = pd.read_csv(filename)
     bird_ids = df['ID'].unique().tolist()
     return bird_ids
   except FileNotFoundError:
     raise HTTPException(
       status_code=404,
-      detail= f'CSV file for {bird} not found')
-
-
-@app.get('/get_general_migration')
-def get_general_migration(selected_bird: str):
-  filename = f'./data/{selected_bird}.csv'
-  
-  try:
-    df = pd.read_csv(filename, low_memory=False)
-    
-    # Convert TIMESTAMP column to datetime
-    df['TIMESTAMP'] = pd.to_datetime(df['TIMESTAMP'])
-    
-    # Group by ID and get the start and end months for each group
-    grouped = df.groupby('ID')
-    print(grouped)
-    simplified_polylines = []
-    for _, group in grouped:
-      # Sort by timestamp
-      group = group.sort_values(by='TIMESTAMP')
-      
-      # Get coordinates
-      coordinates = list(zip(group['LATITUDE'], group['LONGITUDE']))
-      
-      # Apply Douglas-Peucker algorithm for simplification
-      simplified_line = simplify_line(coordinates)
-      
-      # Calculate direction for simplified line
-      start_point = simplified_line[0]
-      end_point = simplified_line[-1]
-      delta_lat = end_point[0] - start_point[0]
-      delta_lon = end_point[1] - start_point[1]
-      direction = np.arctan2(delta_lat, delta_lon) * (180 / np.pi)
-      
-      # Append simplified line with direction to simplified_polylines
-      simplified_polylines.append({
-          'coordinates': simplified_line,
-          'direction': direction
-      })
-    
-    return {'segmented_polylines': simplified_polylines}
-  
-  except Exception as e:
-    raise HTTPException(status_code= 400, detail=str(e))
-
-        
-def simplify_line(coordinates, tolerance=0.1):
-  line = LineString(coordinates)
-  simplified_line = line.simplify(tolerance, preserve_topology=False)
-  return list(zip(*simplified_line.xy))
+      detail= f'CSV file for {filename} not found')
 
 
 @app.get('/get_heatmap_data')
@@ -334,69 +272,4 @@ def get_heatmap_data(bird: str):
         heatmap_data = df[['LATITUDE', 'LONGITUDE']].values.tolist()
         return heatmap_data
     except Exception as e:
-      raise HTTPException(status_code=400, detail= str(e))
-
-source_epsg = 'epsg:4326'  # Input coordinates in EPSG 4326 (WGS84)
-target_epsg = 'epsg:3587' 
-    
-@app.get('/get_SDM_data')
-def get_SDM_data(prediction_input: PredictionInputs):
-  selected_bird = prediction_input.bird
-  selected_year = str(prediction_input.year)
-  emission_Type = prediction_input.emissions
-
-  tiff_file_path = f"../model/outputs/tiff-images/{birdsModelDirs[selected_bird]}/{emission_Type}/{selected_year}/probability_1.0.tif"
-
-  with rasterio.open(tiff_file_path) as src:
-      # Read the data from the TIFF file
-      data = src.read(1)  # Assuming a single band for simplicity
-
-        # Get the transformation parameters for the reprojected image
-      transform, width, height = calculate_default_transform(src.crs, target_epsg, src.width, src.height, *src.bounds)
-
-        # Reproject the image
-      data_reprojected = np.empty((height, width), dtype=data.dtype)
-      reproject(
-          source=data,
-          destination=data_reprojected,
-          src_transform=src.transform,
-          src_crs=src.crs,
-          dst_transform=transform,
-          dst_crs=target_epsg,
-          resampling=Resampling.nearest
-      )
-
-    # Convert numpy array to list for JSON serialization
-  data_list = data_reprojected.tolist()
-
-  # Prepare data to send to frontend
-  converted_data = {'data': data_list}
-
-  # Convert data to JSON
-  json_data = json.dumps(converted_data)
-
-  # Send converted data to frontend
-  return JSONResponse(content=converted_data)
-
-'''@api.route('/get_SDM_data')
-@cross_origin(supports_credentials=True)
-def get_SDM_data():
-    tiff_file_path = f"../model/outputs/tiff-images/{birdsModelDirs[session['bird']]}/{session['emissions']}/{session['year']}/probability_1.0.tif"
-    try:
-        # Open the image using Pillow
-        img = Image.open(tiff_file_path)
-        dataset = rasterio.open(tiff_file_path)
-        print(dataset.bounds)
-        # Extract data from the image
-        img_array = np.array(img)
-        response_data = {'tiff_data': img_array.tolist()}
-        # Send the TIFF data as JSON response
-        return jsonify(response_data)
-    
-    except Exception as e:
-        return f'An error occurred: {e}'
-'''
-
-if __name__ == '__main__':
-    api.run(debug=True)
-    socket_io.run(api, debug=True, port=5000)
+      raise HTTPException(status_code=400, detail= str(e))

backend/app/base.py

@@ -1,11 +1,9 @@
 fastapi
-
-python-dotenv
-jinja2
-pandas
-Shapely
-
-pytest
+pydantic
 requests
-pytest-cov
-flask_testing
+pandas
+numpy
+shapely
+rasterio
+pillow
+pytest