chore: working on the inference with trees

play_with_endpoint.py

@@ -4,9 +4,6 @@ import os, sys
 
 from pathlib import Path
 
-from sklearn.datasets import make_classification
-from sklearn.model_selection import train_test_split
-
 from concrete.ml.deployment import FHEModelClient
 
 import requests
@@ -24,7 +21,7 @@ def from_json(python_object):
 
 
 # TODO: put the right link `API_URL` for your entryp point
-API_URL = "https://puqif7goarh132kl.us-east-1.aws.endpoints.huggingface.cloud"
+API_URL = "https://yw1dgyuig6ff5pft.us-east-1.aws.endpoints.huggingface.cloud"
 headers = {
     "Authorization": "Bearer " + os.environ.get("HF_TOKEN"),
     "Content-Type": "application/json",
@@ -38,9 +35,24 @@ def query(payload):
 
 path_to_model = Path("compiled_model")
 
-# BEGIN: replace this part with your privacy-preserving application
-x, y = make_classification(n_samples=1000, class_sep=2, n_features=30, random_state=42)
-_, X_test, _, y_test = train_test_split(x, y, test_size=0.2, random_state=42)
+# Decision-tree in FHE
+from sklearn.datasets import fetch_openml
+from sklearn.model_selection import train_test_split
+import numpy
+
+features, classes = fetch_openml(data_id=44, as_frame=False, cache=True, return_X_y=True)
+classes = classes.astype(numpy.int64)
+
+_, X_test, _, Y_test = train_test_split(
+    features,
+    classes,
+    test_size=0.15,
+    random_state=42,
+)
+
+NB_SAMPLES = 2
+X_test = X_test[:NB_SAMPLES]
+Y_test = Y_test[:NB_SAMPLES]
 
 # Recover parameters for client side
 fhemodel_client = FHEModelClient(path_to_model)
@@ -60,7 +72,7 @@ is_first = True
 for i in range(nb_samples):
 
     # Quantize the input and encrypt it
-    encrypted_inputs = fhemodel_client.quantize_encrypt_serialize([X_test[i]])
+    encrypted_inputs = fhemodel_client.quantize_encrypt_serialize(X_test[i].reshape(1, -1))
 
     # Prepare the payload, including the evaluation keys which are needed server side
     payload = {
@@ -94,4 +106,3 @@ print(f"Accuracy on {nb_samples} samples is {nb_good * 1. / nb_samples}")
 print(f"Total time: {time.time() - time_start} seconds")
 print(f"Duration in inferences: {duration} seconds")
 print(f"Duration per inference: {duration / nb_samples} seconds")
-# END: replace this part with your privacy-preserving application