Upload app.py

app.py

@@ -1,33 +1,101 @@
-
 import gradio as gr
 import torch
 import torch.nn as nn
 
-# Define your custom model class
+# Define your custom model class with detailed layer structures
+class Head(nn.Module):
+    def __init__(self, head_size):
+        super().__init__()
+        self.key = nn.Linear(64, head_size, bias=False)
+        self.query = nn.Linear(64, head_size, bias=False)
+        self.value = nn.Linear(64, head_size, bias=False)
+        self.register_buffer('tril', torch.tril(torch.ones(32, 32)))
+        self.dropout = nn.Dropout(0.1)
+
+    def forward(self, x):
+        B, T, C = x.shape
+        k = self.key(x)
+        q = self.query(x)
+        wei = q @ k.transpose(-2, -1) * C**-0.5
+        wei = wei.masked_fill(self.tril[:T, :T] == 0, float('-inf'))
+        wei = nn.functional.softmax(wei, dim=-1)
+        wei = self.dropout(wei)
+        v = self.value(x)
+        return wei @ v
+
+class MultiHeadAttention(nn.Module):
+    def __init__(self, num_heads, head_size):
+        super().__init__()
+        self.heads = nn.ModuleList([Head(head_size) for _ in range(num_heads)])
+        self.proj = nn.Linear(64, 64)
+        self.dropout = nn.Dropout(0.1)
+
+    def forward(self, x):
+        out = torch.cat([h(x) for h in self.heads], dim=-1)
+        return self.dropout(self.proj(out))
+
+class FeedForward(nn.Module):
+    def __init__(self, n_embd):
+        super().__init__()
+        self.net = nn.Sequential(
+            nn.Linear(n_embd, 4 * n_embd),
+            nn.ReLU(),
+            nn.Linear(4 * n_embd, n_embd),
+            nn.Dropout(0.1),
+        )
+
+    def forward(self, x):
+        return self.net(x)
+
+class Block(nn.Module):
+    def __init__(self, n_embd, n_head):
+        super().__init__()
+        head_size = n_embd // n_head
+        self.sa = MultiHeadAttention(n_head, head_size)
+        self.ffwd = FeedForward(n_embd)
+        self.ln1 = nn.LayerNorm(n_embd)
+        self.ln2 = nn.LayerNorm(n_embd)
+
+    def forward(self, x):
+        x = x + self.sa(self.ln1(x))
+        x = x + self.ffwd(self.ln2(x))
+        return x
+
 class BigramLanguageModel(nn.Module):
     def __init__(self):
         super().__init__()
-        # Example layers (adjust as needed for your model)
         self.token_embedding_table = nn.Embedding(61, 64)
         self.position_embedding_table = nn.Embedding(32, 64)
-        self.blocks = nn.Sequential(*[nn.Linear(64, 64) for _ in range(4)])
+        self.blocks = nn.Sequential(*[Block(64, n_head=4) for _ in range(4)])
         self.ln_f = nn.LayerNorm(64)
         self.lm_head = nn.Linear(64, 61)
 
-    def forward(self, idx):
-        # Implement the forward pass
-        pass
+    def forward(self, idx, targets=None):
+        B, T = idx.shape
+        tok_emb = self.token_embedding_table(idx)
+        pos_emb = self.position_embedding_table(torch.arange(T, device=idx.device))
+        x = tok_emb + pos_emb
+        x = self.blocks(x)
+        x = self.ln_f(x)
+        logits = self.lm_head(x)
+        return logits, None
 
-    def generate(self, idx, max_new_tokens=250):
-        # Implement the generate method
-        pass
+    def generate(self, idx, max_new_tokens):
+        for _ in range(max_new_tokens):
+            idx_cond = idx[:, -32:]
+            logits, _ = self(idx_cond)
+            logits = logits[:, -1, :]
+            probs = nn.functional.softmax(logits, dim=-1)
+            idx_next = torch.multinomial(probs, num_samples=1)
+            idx = torch.cat((idx, idx_next), dim=1)
+        return idx
 
-# Load your model
+# Load the model with strict=False to handle missing or unexpected keys
 def load_model():
     model = BigramLanguageModel()
     model_url = "https://huggingface.co/yoonusajwardapiit/triptuner/resolve/main/pytorch_model.bin"
     model_weights = torch.hub.load_state_dict_from_url(model_url, map_location=torch.device('cpu'), weights_only=True)
-    model.load_state_dict(model_weights)
+    model.load_state_dict(model_weights, strict=False)
     model.eval()
     return model