Upload 4 files

app.py

@@ -0,0 +1,26 @@
+import streamlit as st
+import torch
+import numpy as np
+import matplotlib.pyplot as plt
+from model import Generator
+
+# Load model
+device = torch.device("cpu")
+generator = Generator()
+generator.load_state_dict(torch.load("generator_digit.pth", map_location=device))
+generator.eval()
+
+st.title("🧠 MNIST Digit Generator")
+digit = st.selectbox("Select a digit (0-9):", list(range(10)))
+
+if st.button("Generate 5 Images"):
+    noise = torch.randn(5, 100)
+    labels = torch.tensor([digit] * 5)
+    with torch.no_grad():
+        generated = generator(noise, labels)
+
+    fig, axs = plt.subplots(1, 5, figsize=(10, 2))
+    for i in range(5):
+        axs[i].imshow(generated[i].squeeze(), cmap='gray')
+        axs[i].axis('off')
+    st.pyplot(fig)

generator_digit.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72679173523a721a7ecf0499d0b3eda442158eddcb50c4989b4570ed5976ca0b
+size 5959844

model.py

@@ -0,0 +1,23 @@
+import torch
+import torch.nn as nn
+
+class Generator(nn.Module):
+    def __init__(self, noise_dim=100, num_classes=10, img_dim=28*28):
+        super(Generator, self).__init__()
+        self.label_emb = nn.Embedding(num_classes, num_classes)
+        self.model = nn.Sequential(
+            nn.Linear(noise_dim + num_classes, 256),
+            nn.ReLU(),
+            nn.Linear(256, 512),
+            nn.ReLU(),
+            nn.Linear(512, 1024),
+            nn.ReLU(),
+            nn.Linear(1024, img_dim),
+            nn.Tanh()
+        )
+
+    def forward(self, z, labels):
+        label_input = self.label_emb(labels)
+        x = torch.cat([z, label_input], dim=1)
+        x = self.model(x)
+        return x.view(-1, 1, 28, 28)

requirements.txt

@@ -1,3 +1,3 @@
-altair
-pandas
-streamlit
+torch
+streamlit
+matplotlib