Fixed the llama model (#769)

* fixed input_pos is None

Signed-off-by: yiliu30 <[email protected]>

* add test

Signed-off-by: yiliu30 <[email protected]>

* update the test

Signed-off-by: yiliu30 <[email protected]>

* update the docstring

Signed-off-by: yiliu30 <[email protected]>

* update the docstring

Signed-off-by: yiliu30 <[email protected]>

---------

Signed-off-by: yiliu30 <[email protected]>

Author: yiliu30

test/test_ao_models.py

@@ -0,0 +1,26 @@
+import pytest
+import torch
+from torchao._models.llama.model import Transformer
+
+_AVAILABLE_DEVICES = ["cpu"] + (["cuda"] if torch.cuda.is_available() else [])
+
+
+def init_model(name="stories15M", device="cpu", precision=torch.bfloat16):
+    model = Transformer.from_name(name)
+    model.to(device=device, dtype=precision)
+    return model.eval()
+
+
+@pytest.mark.parametrize("device", _AVAILABLE_DEVICES)
+@pytest.mark.parametrize("batch_size", [1, 4])
+@pytest.mark.parametrize("is_training", [True, False])
+def test_ao_llama_model_inference_mode(device, batch_size, is_training):
+    random_model = init_model(device=device)
+    seq_len = 16
+    input_ids = torch.randint(0, 1024, (batch_size, seq_len)).to(device)
+    input_pos = None if is_training else torch.arange(seq_len).to(device)
+    with torch.device(device):
+        random_model.setup_caches(max_batch_size=batch_size, max_seq_length=seq_len, training=is_training)
+    for i in range(3):
+        out = random_model(input_ids, input_pos)
+        assert out is not None, "model failed to run"

torchao/_models/llama/model.py

@@ -193,19 +193,33 @@ def setup_caches(self, max_batch_size, max_seq_length, training: bool=False, kv_
 
 
     def forward(self, idx: Tensor, input_pos: Optional[Tensor] = None) -> Tensor:
+        """Forward pass of the model.
+
+        Args:
+            idx  (`torch.LongTensor` of shape `(batch_size, seq_length)`): 
+                Indices of input sequence tokens in the vocabulary.
+            input_pos (`torch.LongTensor` of shape `(batch_size, seq_length)`, *optional*):
+                Indices of positions of each input sequence tokens in the position embeddings.
+                This argument is optional for training mode but required for
+                inference mode(when model.setup_caches(training=False) is used).
+
+        Returns:
+            Tensor: The output logits tensor.
+        """
         assert self.freqs_cis is not None, "Caches must be initialized first"
 
         if input_pos is None: 
             mask = None
             freqs_cis = self.freqs_cis[:idx.shape[1]]
-        elif not self.linear_causal_mask:
-            mask = self.causal_mask[None, None, input_pos]
-        elif len(input_pos)>1 and self.linear_causal_mask: # prefill for linear causal mask
-            mask = torch.tril(torch.ones(len(input_pos), self.max_seq_length, dtype=torch.bool, device=input_pos.device)).unsqueeze(0).unsqueeze(0)
-        else: # decode_one_token for linear causal mask
-            self.causal_mask[0,0,0,input_pos] = 1
-            mask = self.causal_mask
-        freqs_cis = self.freqs_cis[input_pos]
+        else:
+            if not self.linear_causal_mask:
+                mask = self.causal_mask[None, None, input_pos]
+            elif len(input_pos)>1 and self.linear_causal_mask: # prefill for linear causal mask
+                mask = torch.tril(torch.ones(len(input_pos), self.max_seq_length, dtype=torch.bool, device=input_pos.device)).unsqueeze(0).unsqueeze(0)
+            else: # decode_one_token for linear causal mask
+                self.causal_mask[0,0,0,input_pos] = 1
+                mask = self.causal_mask
+            freqs_cis = self.freqs_cis[input_pos]
 
         x = self.tok_embeddings(idx)