Update app.py

app.py

@@ -7,7 +7,9 @@ import numpy as np
 from sklearn.model_selection import train_test_split
 import gradio as gr
 import matplotlib.pyplot as plt
-import timm
+from skops import hub_utils
+import pickle
+
 
 
 #Data preparation
@@ -35,6 +37,37 @@ def visualize_input_data():
     # plt.show()
     return fig
 
+def download_model_skops():
+    repo_id="sklearn-docs/anomaly-detection"
+    download_repo = "downloaded-model"
+    hub_utils.download(repo_id=repo_id, dst=download_repo)
+    #     repo_copy = mkdtemp(prefix="skops")
+    # hub_utils.download(repo_id=repo_id, dst=repo_copy, token=token)
+    # print(os.listdir(download_repo))
+
+
+from sklearn.inspection import DecisionBoundaryDisplay
+
+def plot_decision_boundary(classifier):
+  disp = DecisionBoundaryDisplay.from_estimator(
+    classifier,
+    X,
+    response_method="predict",
+    alpha=0.5,
+)
+  fig = plt.figure(1, facecolor="w", figsize=(5, 5))
+  scatter = plt.scatter(X[:, 0], X[:, 1], c=y, s=20, edgecolor="k") 
+  disp.ax_.scatter(X[:, 0], X[:, 1], c=y, s=20, edgecolor="k")
+  handles, labels = scatter.legend_elements()
+  disp.ax_.set_title("Binary decision boundary \nof IsolationForest")
+  plt.axis("square")
+  
+  plt.legend(handles=handles, labels=["outliers", "inliers"], title="true class")
+  # plt.savefig('decision_boundary.png',dpi=300, bbox_inches = "tight")
+  return fig
+
+
+
 title = " An example using IsolationForest for anomaly detection."
 
 with gr.Blocks(title=title) as demo:
@@ -47,8 +80,23 @@ with gr.Blocks(title=title) as demo:
         
     btn = gr.Button(value="Visualize input dataset")
     btn.click(visualize_input_data, outputs= gr.Plot(label='Visualizing input dataset') )
+    # download
+    repo_id="sklearn-docs/anomaly-detection"
+    download_repo = "downloaded-model"
+    hub_utils.download(repo_id=repo_id, dst=download_repo)
+    if os.listdir(download_repo):
+      # hub_utils.download(repo_id=repo_id, dst=download_repo)
+    
 
-    model_reloaded = timm.create_model('hf_hub:sklearn-docs/anomaly-detection', pretrained=True)
+  
+    
+      # print("Empty directory")
+
+      # print(os.listdir(download_repo))
+      loaded_model = pickle.load(open('isolation_forest.pkl', 'rb'))
+      btn = gr.Button(value="Plot decision boundary")
+      btn.click(plot_decision_boundary, inputs=[loaded_model], outputs= gr.Plot(label='Visualizing input dataset') )
+    
     
-    gr.Markdown( f"## In progress")
+    gr.Markdown( f"## Success")
 demo.launch()