Update gradio_interface.py

gradio_interface.py

@@ -1,6 +1,5 @@
 import gradio as gr
 import spacy
-import jsonlines
 from sklearn.metrics import classification_report, accuracy_score, f1_score, precision_score, recall_score
 
 # Load the trained spaCy model
@@ -13,46 +12,48 @@ def classify_text(text):
     predicted_labels = doc.cats
     return predicted_labels
 
-# Function to evaluate the model based on user input
-def evaluate_model(input_text):
-    # Load the golden evaluation data
-    golden_eval_data = []
-    with jsonlines.open("data/goldenEval.jsonl") as reader:
-        for record in reader:
-            golden_eval_data.append(record)
-
-    # Predict labels for the input text using the model
+# Function to evaluate the predicted labels for the input text
+def evaluate_text(input_text):
+    # Get the predicted labels for the input text
     predicted_labels = classify_text(input_text)
+    
+    # Assuming you have ground truth labels for the input text, you would compare the predicted labels with the ground truth labels here.
+    # For demonstration purposes, let's assume the ground truth labels are provided here.
+    ground_truth_labels = {
+        "CapitalRequirements": 0,
+        "ConsumerProtection": 1,
+        "RiskManagement": 0,
+        "ReportingAndCompliance": 1,
+        "CorporateGovernance": 0
+    }
 
-    # Extract ground truth labels from the golden evaluation data
-    true_labels = [record["accept"] for record in golden_eval_data]
-
-    # Convert label format to match sklearn's classification report format
-    true_labels_flat = [label[0] if label else "reject" for label in true_labels]
-    predicted_labels_flat = [max(predicted_labels, key=predicted_labels.get) for _ in range(len(true_labels))]
+    # Convert predicted and ground truth labels to lists
+    predicted_labels_list = [1 if predicted_labels[label] > 0.5 else 0 for label in predicted_labels]
+    ground_truth_labels_list = [ground_truth_labels[label] for label in predicted_labels]
 
     # Calculate evaluation metrics
-    accuracy = accuracy_score(true_labels_flat, predicted_labels_flat)
-    precision = precision_score(true_labels_flat, predicted_labels_flat, average='weighted')
-    recall = recall_score(true_labels_flat, predicted_labels_flat, average='weighted')
-    f1 = f1_score(true_labels_flat, predicted_labels_flat, average='weighted')
+    accuracy = accuracy_score(ground_truth_labels_list, predicted_labels_list)
+    precision = precision_score(ground_truth_labels_list, predicted_labels_list, average='weighted')
+    recall = recall_score(ground_truth_labels_list, predicted_labels_list, average='weighted')
+    f1 = f1_score(ground_truth_labels_list, predicted_labels_list, average='weighted')
 
     # Additional classification report
-    report = classification_report(true_labels_flat, predicted_labels_flat)
-
-    # Return evaluation metrics and report
-    return {
-        "accuracy": accuracy,
-        "precision": precision,
-        "recall": recall,
-        "f1": f1,
-        "classification_report": report
+    report = classification_report(ground_truth_labels_list, predicted_labels_list)
+
+    # Construct output
+    evaluation_results = {
+        "Accuracy": accuracy,
+        "Precision": precision,
+        "Recall": recall,
+        "F1-Score": f1,
+        "Classification Report": report
     }
+    
+    return evaluation_results
 
-# Gradio Interface for text classification
-classification_interface = gr.Interface(fn=classify_text, inputs="textbox", outputs="json", title="Text Classifier", description="Enter your text")
-classification_interface.launch(share=True)
+# Gradio Interface
+inputs = gr.inputs.Textbox(lines=7, label="Enter your text")
+outputs = gr.outputs.JSON(label="Evaluation Results")
 
-# Gradio Interface for model evaluation
-evaluation_interface = gr.Interface(fn=evaluate_model, inputs="textbox", outputs="json", title="Model Evaluation", description="Enter text to evaluate the model")
-evaluation_interface.launch(share=True)
+iface = gr.Interface(fn=evaluate_text, inputs=inputs, outputs=outputs, title="Text Evaluation")
+iface.launch(share=True)