Enhance GlobalWorkspace to support dtype and device in zero padding; …

…add setup fixture for default tensor settings
Neuro-Flex · Dec 27, 2024 · d65db2b · d65db2b
1 parent e9e9d99
commit d65db2b
Show file tree

Hide file tree

Showing 52 changed files with 17 additions and 1 deletion.
diff --git a/__pycache__/test_init.cpython-310-pytest-8.3.4.pyc b/__pycache__/test_init.cpython-310-pytest-8.3.4.pyc
diff --git a/core/__pycache__/config.cpython-310.pyc b/core/__pycache__/config.cpython-310.pyc
diff --git a/models/__pycache__/__init__.cpython-310.pyc b/models/__pycache__/__init__.cpython-310.pyc
diff --git a/models/__pycache__/abstract_reasoning.cpython-310.pyc b/models/__pycache__/abstract_reasoning.cpython-310.pyc
diff --git a/models/__pycache__/attention.cpython-310.pyc b/models/__pycache__/attention.cpython-310.pyc
diff --git a/models/__pycache__/base_model.cpython-310.pyc b/models/__pycache__/base_model.cpython-310.pyc
diff --git a/models/__pycache__/consciousness.cpython-310.pyc b/models/__pycache__/consciousness.cpython-310.pyc
diff --git a/models/__pycache__/consciousness_model.cpython-310.pyc b/models/__pycache__/consciousness_model.cpython-310.pyc
diff --git a/models/__pycache__/consciousness_state.cpython-310.pyc b/models/__pycache__/consciousness_state.cpython-310.pyc
diff --git a/models/__pycache__/dynamic_attention.cpython-310.pyc b/models/__pycache__/dynamic_attention.cpython-310.pyc
diff --git a/models/__pycache__/error_handling.cpython-310.pyc b/models/__pycache__/error_handling.cpython-310.pyc
diff --git a/models/__pycache__/ethical_safety.cpython-310.pyc b/models/__pycache__/ethical_safety.cpython-310.pyc
diff --git a/models/__pycache__/global_workspace.cpython-310.pyc b/models/__pycache__/global_workspace.cpython-310.pyc
diff --git a/models/__pycache__/information_integration.cpython-310.pyc b/models/__pycache__/information_integration.cpython-310.pyc
diff --git a/models/__pycache__/intentionality.cpython-310.pyc b/models/__pycache__/intentionality.cpython-310.pyc
diff --git a/models/__pycache__/long_term_memory.cpython-310.pyc b/models/__pycache__/long_term_memory.cpython-310.pyc
diff --git a/models/__pycache__/memory.cpython-310.pyc b/models/__pycache__/memory.cpython-310.pyc
diff --git a/models/__pycache__/reasoning.cpython-310.pyc b/models/__pycache__/reasoning.cpython-310.pyc
diff --git a/models/__pycache__/self_awareness.cpython-310.pyc b/models/__pycache__/self_awareness.cpython-310.pyc
diff --git a/models/__pycache__/simulated_emotions.cpython-310.pyc b/models/__pycache__/simulated_emotions.cpython-310.pyc
diff --git a/models/__pycache__/working_memory.cpython-310.pyc b/models/__pycache__/working_memory.cpython-310.pyc
diff --git a/models/global_workspace.py b/models/global_workspace.py
@@ -75,7 +75,11 @@ def forward(self, modalities: Union[Dict[str, torch.Tensor], None] = None, senso
 
         # Pad remaining slots with zeros if needed
         while len(integrated_features) < self.num_modalities:
-            zero_features = torch.zeros_like(integrated_features[0])
+            zero_features = torch.zeros_like(
+                integrated_features[0], 
+                dtype=integrated_features[0].dtype,
+                device=integrated_features[0].device
+            )
             integrated_features.append(zero_features)
 
         # Stack and reshape for attention
@@ -121,6 +125,7 @@ def forward(self, modalities: Union[Dict[str, torch.Tensor], None] = None, senso
                 self.num_modalities - attention_weights.size(1),
                 seq_len,
                 attention_weights.size(3),
+                dtype=attention_weights.dtype,
                 device=attention_weights.device
             )
             attention_weights = torch.cat([attention_weights, padding], dim=1)
@@ -132,6 +137,7 @@ def forward(self, modalities: Union[Dict[str, torch.Tensor], None] = None, senso
                 attention_weights.size(1),
                 3 - attention_weights.size(2),
                 attention_weights.size(3),
+                dtype=attention_weights.dtype,
                 device=attention_weights.device
             )
             attention_weights = torch.cat([attention_weights, padding], dim=2)

diff --git a/tests/__pycache__/conftest.cpython-310-pytest-8.3.4.pyc b/tests/__pycache__/conftest.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/__pycache__/test_consciousness.cpython-310-pytest-8.3.4.pyc b/tests/__pycache__/test_consciousness.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/__pycache__/test_consciousness_integration.cpython-310-pytest-8.3.4.pyc b/tests/__pycache__/test_consciousness_integration.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/__pycache__/test_dynamic_attention.cpython-310-pytest-8.3.4.pyc b/tests/__pycache__/test_dynamic_attention.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/__pycache__/test_environment.cpython-310-pytest-8.3.4.pyc b/tests/__pycache__/test_environment.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/__pycache__/test_error_correction.cpython-310-pytest-8.3.4.pyc b/tests/__pycache__/test_error_correction.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/__pycache__/test_error_handling.cpython-310-pytest-8.3.4.pyc b/tests/__pycache__/test_error_handling.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/__pycache__/test_ethical_safety.cpython-310-pytest-8.3.4.pyc b/tests/__pycache__/test_ethical_safety.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/__pycache__/test_intentionality.cpython-310-pytest-8.3.4.pyc b/tests/__pycache__/test_intentionality.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/__pycache__/test_lint.cpython-310-pytest-8.3.4.pyc b/tests/__pycache__/test_lint.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/__pycache__/test_long_term_memory.cpython-310-pytest-8.3.4.pyc b/tests/__pycache__/test_long_term_memory.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/__pycache__/test_reasoning.cpython-310-pytest-8.3.4.pyc b/tests/__pycache__/test_reasoning.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/__pycache__/test_self_awareness.cpython-310-pytest-8.3.4.pyc b/tests/__pycache__/test_self_awareness.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/benchmarks/__pycache__/test_arc_reasoning.cpython-310-pytest-8.3.4.pyc b/tests/benchmarks/__pycache__/test_arc_reasoning.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/benchmarks/__pycache__/test_bigbench_reasoning.cpython-310-pytest-8.3.4.pyc b/tests/benchmarks/__pycache__/test_bigbench_reasoning.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/conftest.py b/tests/conftest.py
@@ -44,3 +44,13 @@ def jit_compile():
     def decorator(func):
         return torch.jit.script(func)
     return decorator
+
+@pytest.fixture(autouse=True)
+def setup_torch_defaults():
+    # Use recommended alternatives to set_default_tensor_type
+    torch.set_default_dtype(torch.float32)
+    # Only set default device if CUDA is available
+    if torch.cuda.is_available():
+        torch.set_default_device('cuda')
+    else:
+        torch.set_default_device('cpu')
diff --git a/tests/unit/__pycache__/__init__.cpython-310.pyc b/tests/unit/__pycache__/__init__.cpython-310.pyc
diff --git a/tests/unit/__pycache__/test_base.cpython-310-pytest-8.3.4.pyc b/tests/unit/__pycache__/test_base.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/unit/__pycache__/test_cognition_progress.cpython-310-pytest-8.3.4.pyc b/tests/unit/__pycache__/test_cognition_progress.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/unit/__pycache__/test_consciousness.cpython-310-pytest-8.3.4.pyc b/tests/unit/__pycache__/test_consciousness.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/unit/__pycache__/test_global_workspace.cpython-310-pytest-8.3.4.pyc b/tests/unit/__pycache__/test_global_workspace.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/unit/__pycache__/test_simulated_emotions.cpython-310-pytest-8.3.4.pyc b/tests/unit/__pycache__/test_simulated_emotions.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/unit/attention/__pycache__/test_attention.cpython-310-pytest-8.3.4.pyc b/tests/unit/attention/__pycache__/test_attention.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/unit/attention/__pycache__/test_attention_mechanisms.cpython-310-pytest-8.3.4.pyc b/tests/unit/attention/__pycache__/test_attention_mechanisms.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/unit/integration/__pycache__/test_cognitive_integration.cpython-310-pytest-8.3.4.pyc b/tests/unit/integration/__pycache__/test_cognitive_integration.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/unit/integration/__pycache__/test_state_management.cpython-310-pytest-8.3.4.pyc b/tests/unit/integration/__pycache__/test_state_management.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/unit/memory/__pycache__/test_integration.cpython-310-pytest-8.3.4.pyc b/tests/unit/memory/__pycache__/test_integration.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/unit/memory/__pycache__/test_memory.cpython-310-pytest-8.3.4.pyc b/tests/unit/memory/__pycache__/test_memory.cpython-310-pytest-8.3.4.pyc
diff --git a/tests/unit/memory/__pycache__/test_memory_components.cpython-310-pytest-8.3.4.pyc b/tests/unit/memory/__pycache__/test_memory_components.cpython-310-pytest-8.3.4.pyc
diff --git a/...s/unit/state/__pycache__/test_consciousness_state_management.cpython-310-pytest-8.3.4.pyc b/...s/unit/state/__pycache__/test_consciousness_state_management.cpython-310-pytest-8.3.4.pyc