Upload 4 files

classifier.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96623293ecbd08c7b7dd73b1e5bb8985919aea941b024c478f64ffa15c6355df
+size 2590

main.py

@@ -0,0 +1,45 @@
+import imghdr
+from os import name
+from fastapi import FastAPI, Response, UploadFile
+from pydantic import BaseModel 
+import time
+import pickle 
+import numpy as np 
+import uvicorn
+from models import classe
+
+app = FastAPI()
+
+pickle_in = open("classifier.pkl","rb")
+classifier = pickle.load(pickle_in)
+
+@app.get("/")
+def index():
+    return {"hello": "FastAPI"}
+
+@app.get('/{name}')
+def get_name(name: str):
+    return {'message': f'hello, {name}'}
+
+@app.post('/predict')
+def predict_species(data: classe):
+    Sepal_Length = data.Sepal_Length
+    Sepal_Width = data.Sepal_Width
+    Petal_Length = data.Petal_Length
+    Petal_Width = data.Petal_Width
+
+    prediction = classifier.predict([[Sepal_Length, Sepal_Width, Petal_Length, Petal_Width]])
+
+    if prediction[0] == 0:
+        species = "setosa"
+    elif prediction[0] == 1:
+        species = "virginica"
+    elif prediction[0] == 2:
+        species = "versicolor"
+    else:
+        species = "unknown"
+
+    return {'prediction': species}
+
+if __name__ == "__main__":
+    uvicorn.run(app, host="127.0.0.1", port=8000)

models.py

@@ -0,0 +1,7 @@
+from pydantic import BaseModel
+
+class classe(BaseModel):
+    Sepal_Length: float
+    Sepal_Width: float
+    Petal_Length: float
+    Petal_Width: float

requirements.txt

@@ -0,0 +1,7 @@
+fastapi
+uvicorn
+pydantic
+numpy
+scikit-learn==1.2.2
+tensorflow
+h5py