Upload 5 files

Dockerfile

@@ -0,0 +1,13 @@
+FROM python:3.9-slim
+
+WORKDIR /app
+
+COPY ./requirements.txt /app/requirements.txt
+
+RUN  pip install --no-cache-dir -r requirements.txt
+
+COPY ./ /app
+
+EXPOSE 8000
+
+CMD ["uvicorn","main:app","--host","0.0.0.0","--port","8000"]

classification_gaussian_binary_model_0v.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a63771c2a046b3a514d302a93739f68011be2f9c40031ec5740df5530826d9f
+size 5640

main.py

@@ -0,0 +1,47 @@
+from typing import List,Dict
+from pydantic import BaseModel
+import numpy as np 
+from fastapi import FastAPI
+import torch
+from model import BinaryClassificationWithLogits
+import __main__
+
+model_path="classification_gaussian_binary_model_0v.pt"
+model=BinaryClassificationWithLogits(in_features=4,
+                                    out_features=1,
+                                    hidden_features=10)
+
+model = torch.jit.load(model_path,map_location="cpu")
+
+
+class ClassificationFeatures(BaseModel):
+  feature_1:float
+  feature_2:float
+  feature_3:float
+  feature_4:float
+
+
+# Creando una instacnia de FastAPI
+app=FastAPI()
+# Definiendo la ruta raiz
+@app.get("/")
+def home_page():
+  return "Welcome the API with pytorch"
+# Definiendo ruta para inferencias
+@app.post("/predict")
+def predict_sample(cls_features:ClassificationFeatures) -> Dict:
+  
+  input_data=np.array([[
+        cls_features.feature_1,
+        cls_features.feature_2,
+        cls_features.feature_3,
+        cls_features.feature_4,
+    ]])
+  X=torch.tensor(input_data,dtype=torch.float32)
+  model.eval()
+  with torch.inference_mode():
+    logit=model(X)
+    pred_prob=torch.sigmoid(logit)
+    pred_label=torch.round(pred_prob)
+  
+  return {"prediction":pred_label.item()}

model.py

@@ -0,0 +1,17 @@
+
+from torch import nn
+
+class BinaryClassificationWithLogits(nn.Module):
+  def __init__(self,in_features,out_features, hidden_features):
+    super().__init__()
+    self.linear_block1=nn.Linear(in_features=in_features,out_features=hidden_features)
+    self.linear_block2=nn.Linear(in_features=hidden_features,out_features=out_features)
+
+  def forward(self,X):
+  
+    x=self.linear_block1(X)
+   
+    x=self.linear_block2(x)
+   
+    # return LOGITS
+    return x

requirements.txt

@@ -0,0 +1,5 @@
+torch
+fastapi
+uvicorn
+scikit-learn
+