Upload [skooldio]_thai_handwritten_recognition_app_by_gradio (2).py

[skooldio]_thai_handwritten_recognition_app_by_gradio (2).py

@@ -0,0 +1,102 @@
+# -*- coding: utf-8 -*-
+"""[Skooldio] Thai Handwritten Recognition App by Gradio
+
+Automatically generated by Colab.
+
+Original file is located at
+    https://colab.research.google.com/drive/1Feza00drEEejwPVzgdTmz60KArcms_w_
+"""
+
+!pip install gradio==3.35.0
+!pip install torchvision
+
+!wget https://github.com/biodatlab/deep-learning-skooldio/raw/master/saved_model/thai_digit_net.pth
+
+import numpy as np
+import torch
+from pathlib import Path
+import torch.nn as nn
+import torch.nn.functional as F
+from PIL import Image
+from torchvision import transforms
+import gradio as gr
+
+
+transform = transforms.Compose([
+    transforms.Resize((28, 28)),
+    transforms.Grayscale(),
+    transforms.ToTensor()
+])
+labels = ["๐ (ศูนย์)", "๑ (หนึ่ง)", "๒ (สอง)", "๓ (สาม)", "๔ (สี่)", "๕ (ห้า)", "๖ (หก)", "๗ (เจ็ด)", "๘ (แปด)", "๙ (เก้า)"]
+LABELS = {i:k for i, k in enumerate(labels)} # dictionary of index and label
+
+
+# Load model using DropoutThaiDigit instead
+class DropoutThaiDigit(nn.Module):
+    def __init__(self):
+        super(DropoutThaiDigit, self).__init__()
+        self.fc1 = nn.Linear(28 * 28, 392)
+        self.fc2 = nn.Linear(392, 196)
+        self.fc3 = nn.Linear(196, 98)
+        self.fc4 = nn.Linear(98, 10)
+        self.dropout = nn.Dropout(0.1)
+
+    def forward(self, x):
+        x = x.view(-1, 28 * 28)
+        x = self.fc1(x)
+        x = F.relu(x)
+        x = self.dropout(x)
+        x = self.fc2(x)
+        x = F.relu(x)
+        x = self.dropout(x)
+        x = self.fc3(x)
+        x = F.relu(x)
+        x = self.dropout(x)
+        x = self.fc4(x)
+        return x
+
+
+model = DropoutThaiDigit()
+model.load_state_dict(torch.load("thai_digit_net.pth"))
+model.eval()
+
+def predict(img):
+    if img.get("composite") is not None:
+        if img["composite"].sum() == 0:
+            return {"No input sketch": 0.0}
+
+    img_data = img['composite']
+    img_gray = Image.fromarray(img_data).convert('L').resize((28, 28))
+    img_tensor = transforms.ToTensor()(img_gray).unsqueeze(0)
+
+    with torch.no_grad():
+        probs = model(img_tensor).softmax(dim=1).squeeze()
+
+    probs, indices = torch.topk(probs, 5)
+    return {LABELS[i]: float(p) for i, p in zip(indices.tolist(), probs.tolist())}
+
+demo = gr.Interface(
+    fn=predict,
+    inputs=gr.Sketchpad(
+        label="Draw Here",
+        image_mode="L",
+        width=400,
+        height=350
+    ),
+    outputs=gr.Label(label="Guess"),
+    title="Thai Digit Handwritten Classification",
+    description="วาดเลขไทยตั้งแต่ ๐ ถึง ๙",
+    live=True
+)
+
+if __name__ == "__main__":
+    demo.launch()
+
+gr.Interface(
+    fn=predict,
+    inputs=gr.Sketchpad(label="Draw Here", brush_radius=5, type="pil", shape=(120, 120)),
+    outputs=gr.Label(label="Guess"),
+    title="Thai Digit Handwritten Classification",
+    live=True
+).launch(enable_queue=True)
+