Spaces:
Sleeping
Sleeping
ManjinderUNCC
commited on
Commit
•
d2e6a80
1
Parent(s):
c2eb30b
Update python_Code/evaluate_model.py
Browse files- python_Code/evaluate_model.py +46 -55
python_Code/evaluate_model.py
CHANGED
@@ -1,55 +1,46 @@
|
|
1 |
-
import
|
2 |
-
|
3 |
-
|
4 |
-
|
5 |
-
|
6 |
-
|
7 |
-
|
8 |
-
|
9 |
-
|
10 |
-
|
11 |
-
|
12 |
-
|
13 |
-
|
14 |
-
|
15 |
-
|
16 |
-
|
17 |
-
|
18 |
-
|
19 |
-
predicted_labels
|
20 |
-
|
21 |
-
|
22 |
-
|
23 |
-
|
24 |
-
|
25 |
-
|
26 |
-
|
27 |
-
|
28 |
-
|
29 |
-
|
30 |
-
|
31 |
-
|
32 |
-
|
33 |
-
|
34 |
-
|
35 |
-
|
36 |
-
|
37 |
-
|
38 |
-
|
39 |
-
|
40 |
-
|
41 |
-
|
42 |
-
|
43 |
-
|
44 |
-
|
45 |
-
|
46 |
-
|
47 |
-
"detailed_classification_report": report
|
48 |
-
}
|
49 |
-
|
50 |
-
return result
|
51 |
-
|
52 |
-
# Gradio Interface
|
53 |
-
output = gr.outputs.Label(type="json", label="Evaluation Metrics")
|
54 |
-
iface = gr.Interface(fn=evaluate_model_script, outputs=output, title="Evaluate Model Script")
|
55 |
-
iface.launch()
|
|
|
1 |
+
import spacy
|
2 |
+
import jsonlines
|
3 |
+
from sklearn.metrics import classification_report, accuracy_score, f1_score, precision_score, recall_score
|
4 |
+
|
5 |
+
# Load the trained spaCy model
|
6 |
+
nlp = spacy.load("./my_trained_model")
|
7 |
+
|
8 |
+
# Load the golden evaluation data
|
9 |
+
golden_eval_data = []
|
10 |
+
with jsonlines.open("data/goldenEval.jsonl") as reader:
|
11 |
+
for record in reader:
|
12 |
+
golden_eval_data.append(record)
|
13 |
+
|
14 |
+
# Predict labels for each record using your model
|
15 |
+
predicted_labels = []
|
16 |
+
for record in golden_eval_data:
|
17 |
+
text = record["text"]
|
18 |
+
doc = nlp(text)
|
19 |
+
predicted_labels.append(doc.cats)
|
20 |
+
|
21 |
+
# Extract ground truth labels from the golden evaluation data
|
22 |
+
true_labels = [record["accept"] for record in golden_eval_data]
|
23 |
+
|
24 |
+
# Convert label format to match sklearn's classification report format
|
25 |
+
true_labels_flat = [label[0] if label else "reject" for label in true_labels]
|
26 |
+
predicted_labels_flat = [max(pred, key=pred.get) for pred in predicted_labels]
|
27 |
+
|
28 |
+
# Calculate evaluation metrics
|
29 |
+
accuracy = accuracy_score(true_labels_flat, predicted_labels_flat)
|
30 |
+
precision = precision_score(true_labels_flat, predicted_labels_flat, average='weighted')
|
31 |
+
recall = recall_score(true_labels_flat, predicted_labels_flat, average='weighted')
|
32 |
+
f1 = f1_score(true_labels_flat, predicted_labels_flat, average='weighted')
|
33 |
+
|
34 |
+
# Additional classification report
|
35 |
+
report = classification_report(true_labels_flat, predicted_labels_flat)
|
36 |
+
|
37 |
+
# Print or save the evaluation metrics
|
38 |
+
print("Evaluation Metrics:")
|
39 |
+
print(f"Accuracy: {accuracy}")
|
40 |
+
print(f"Precision: {precision}")
|
41 |
+
print(f"Recall: {recall}")
|
42 |
+
print(f"F1-Score: {f1}")
|
43 |
+
|
44 |
+
# Print or save the detailed classification report
|
45 |
+
print("Detailed Classification Report:")
|
46 |
+
print(report)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|