Use attention tests from master

Author: Ervin Teng

ml-agents/mlagents/trainers/tests/torch/test_attention.py

@@ -1,12 +1,14 @@
+import pytest
 from mlagents.torch_utils import torch
 import numpy as np
 
 from mlagents.trainers.torch.utils import ModelUtils
 from mlagents.trainers.torch.layers import linear_layer, LinearEncoder
 from mlagents.trainers.torch.attention import (
     MultiHeadAttention,
-    EntityEmbeddings,
+    EntityEmbedding,
     ResidualSelfAttention,
+    get_zero_entities_mask,
 )
 
 
@@ -71,7 +73,7 @@ def generate_input_helper(pattern):
     input_1 = generate_input_helper(masking_pattern_1)
     input_2 = generate_input_helper(masking_pattern_2)
 
-    masks = EntityEmbeddings.get_masks([input_1, input_2])
+    masks = get_zero_entities_mask([input_1, input_2])
     assert len(masks) == 2
     masks_1 = masks[0]
     masks_2 = masks[1]
@@ -83,13 +85,60 @@ def generate_input_helper(pattern):
         assert masks_2[0, 1] == 0 if i % 2 == 0 else 1
 
 
+@pytest.mark.parametrize("mask_value", [0, 1])
+def test_all_masking(mask_value):
+    # We make sure that a mask of all zeros or all ones will not trigger an error
+    np.random.seed(1336)
+    torch.manual_seed(1336)
+    size, n_k, = 3, 5
+    embedding_size = 64
+    entity_embeddings = EntityEmbedding(size, n_k, embedding_size)
+    entity_embeddings.add_self_embedding(size)
+    transformer = ResidualSelfAttention(embedding_size, n_k)
+    l_layer = linear_layer(embedding_size, size)
+    optimizer = torch.optim.Adam(
+        list(entity_embeddings.parameters())
+        + list(transformer.parameters())
+        + list(l_layer.parameters()),
+        lr=0.001,
+        weight_decay=1e-6,
+    )
+    batch_size = 20
+    for _ in range(5):
+        center = torch.rand((batch_size, size))
+        key = torch.rand((batch_size, n_k, size))
+        with torch.no_grad():
+            # create the target : The key closest to the query in euclidean distance
+            distance = torch.sum(
+                (center.reshape((batch_size, 1, size)) - key) ** 2, dim=2
+            )
+            argmin = torch.argmin(distance, dim=1)
+            target = []
+            for i in range(batch_size):
+                target += [key[i, argmin[i], :]]
+            target = torch.stack(target, dim=0)
+            target = target.detach()
+
+        embeddings = entity_embeddings(center, key)
+        masks = [torch.ones_like(key[:, :, 0]) * mask_value]
+        prediction = transformer.forward(embeddings, masks)
+        prediction = l_layer(prediction)
+        prediction = prediction.reshape((batch_size, size))
+        error = torch.mean((prediction - target) ** 2, dim=1)
+        error = torch.mean(error) / 2
+        optimizer.zero_grad()
+        error.backward()
+        optimizer.step()
+
+
 def test_predict_closest_training():
     np.random.seed(1336)
     torch.manual_seed(1336)
     size, n_k, = 3, 5
     embedding_size = 64
-    entity_embeddings = EntityEmbeddings(size, [size], embedding_size, [n_k])
-    transformer = ResidualSelfAttention(embedding_size, [n_k])
+    entity_embeddings = EntityEmbedding(size, n_k, embedding_size)
+    entity_embeddings.add_self_embedding(size)
+    transformer = ResidualSelfAttention(embedding_size, n_k)
     l_layer = linear_layer(embedding_size, size)
     optimizer = torch.optim.Adam(
         list(entity_embeddings.parameters())
@@ -114,8 +163,8 @@ def test_predict_closest_training():
             target = torch.stack(target, dim=0)
             target = target.detach()
 
-        embeddings = entity_embeddings(center, [key])
-        masks = EntityEmbeddings.get_masks([key])
+        embeddings = entity_embeddings(center, key)
+        masks = get_zero_entities_mask([key])
         prediction = transformer.forward(embeddings, masks)
         prediction = l_layer(prediction)
         prediction = prediction.reshape((batch_size, size))
@@ -135,14 +184,12 @@ def test_predict_minimum_training():
     n_k = 5
     size = n_k + 1
     embedding_size = 64
-    entity_embeddings = EntityEmbeddings(
-        size, [size], embedding_size, [n_k], concat_self=False
-    )
+    entity_embedding = EntityEmbedding(size, n_k, embedding_size)  # no self
     transformer = ResidualSelfAttention(embedding_size)
     l_layer = LinearEncoder(embedding_size, 2, n_k)
     loss = torch.nn.CrossEntropyLoss()
     optimizer = torch.optim.Adam(
-        list(entity_embeddings.parameters())
+        list(entity_embedding.parameters())
         + list(transformer.parameters())
         + list(l_layer.parameters()),
         lr=0.001,
@@ -166,8 +213,8 @@ def test_predict_minimum_training():
         sliced_oh = onehots[:, : num + 1]
         inp = torch.cat([inp, sliced_oh], dim=2)
 
-        embeddings = entity_embeddings(inp, [inp])
-        masks = EntityEmbeddings.get_masks([inp])
+        embeddings = entity_embedding(inp, inp)
+        masks = get_zero_entities_mask([inp])
         prediction = transformer(embeddings, masks)
         prediction = l_layer(prediction)
         ce = loss(prediction, argmin)

ml-agents/mlagents/trainers/torch/attention.py

@@ -10,22 +10,41 @@
 from mlagents.trainers.exception import UnityTrainerException
 
 
-class MultiHeadAttention(torch.nn.Module):
+def get_zero_entities_mask(observations: List[torch.Tensor]) -> List[torch.Tensor]:
     """
-    Multi Head Attention module. We do not use the regular Torch implementation since
-    Barracuda does not support some operators it uses.
-    Takes as input to the forward method 3 tensors:
-     - query: of dimensions (batch_size, number_of_queries, embedding_size)
-     - key: of dimensions (batch_size, number_of_keys, embedding_size)
-     - value: of dimensions (batch_size, number_of_keys, embedding_size)
-    The forward method will return 2 tensors:
-     - The output: (batch_size, number_of_queries, embedding_size)
-     - The attention matrix: (batch_size, num_heads, number_of_queries, number_of_keys)
+    Takes a List of Tensors and returns a List of mask Tensor with 1 if the input was
+    all zeros (on dimension 2) and 0 otherwise. This is used in the Attention
+    layer to mask the padding observations.
     """
+    with torch.no_grad():
+        # Generate the masking tensors for each entities tensor (mask only if all zeros)
+        key_masks: List[torch.Tensor] = [
+            (torch.sum(ent ** 2, axis=2) < 0.01).float() for ent in observations
+        ]
+    return key_masks
+
+
+class MultiHeadAttention(torch.nn.Module):
 
     NEG_INF = -1e6
 
     def __init__(self, embedding_size: int, num_heads: int):
+        """
+        Multi Head Attention module. We do not use the regular Torch implementation since
+        Barracuda does not support some operators it uses.
+        Takes as input to the forward method 3 tensors:
+        - query: of dimensions (batch_size, number_of_queries, embedding_size)
+        - key: of dimensions (batch_size, number_of_keys, embedding_size)
+        - value: of dimensions (batch_size, number_of_keys, embedding_size)
+        The forward method will return 2 tensors:
+        - The output: (batch_size, number_of_queries, embedding_size)
+        - The attention matrix: (batch_size, num_heads, number_of_queries, number_of_keys)
+        :param embedding_size: The size of the embeddings that will be generated (should be
+        dividable by the num_heads)
+        :param total_max_elements: The maximum total number of entities that can be passed to
+        the module
+        :param num_heads: The number of heads of the attention module
+        """
         super().__init__()
         self.n_heads = num_heads
         self.head_size: int = embedding_size // self.n_heads
@@ -82,7 +101,7 @@ def forward(
         return value_attention, att
 
 
-class EntityEmbeddings(torch.nn.Module):
+class EntityEmbedding(torch.nn.Module):
     """
     A module used to embed entities before passing them to a self-attention block.
     Used in conjunction with ResidualSelfAttention to encode information about a self
@@ -92,95 +111,69 @@ class EntityEmbeddings(torch.nn.Module):
 
     def __init__(
         self,
-        x_self_size: int,
-        entity_sizes: List[int],
+        entity_size: int,
+        entity_num_max_elements: Optional[int],
         embedding_size: int,
-        entity_num_max_elements: Optional[List[int]] = None,
-        concat_self: bool = True,
     ):
         """
-        Constructs an EntityEmbeddings module.
+        Constructs an EntityEmbedding module.
         :param x_self_size: Size of "self" entity.
-        :param entity_sizes: List of sizes for other entities. Should be of length
-            equivalent to the number of entities.
-        :param embedding_size: Embedding size for entity encoders.
-        :param entity_num_max_elements: Maximum elements in an entity, None for unrestricted.
+        :param entity_size: Size of other entities.
+        :param entity_num_max_elements: Maximum elements for a given entity, None for unrestricted.
             Needs to be assigned in order for model to be exportable to ONNX and Barracuda.
-        :param concat_self: Whether to concatenate x_self to entites. Set True for ego-centric
+        :param embedding_size: Embedding size for the entity encoder.
+        :param concat_self: Whether to concatenate x_self to entities. Set True for ego-centric
             self-attention.
         """
         super().__init__()
-        self.self_size: int = x_self_size
-        self.entity_sizes: List[int] = entity_sizes
-        self.entity_num_max_elements: List[int] = [-1] * len(entity_sizes)
+        self.self_size: int = 0
+        self.entity_size: int = entity_size
+        self.entity_num_max_elements: int = -1
         if entity_num_max_elements is not None:
             self.entity_num_max_elements = entity_num_max_elements
-
-        self.concat_self: bool = concat_self
-        # If not concatenating self, input to encoder is just entity size
-        if not concat_self:
-            self.self_size = 0
+        self.embedding_size = embedding_size
         # Initialization scheme from http://www.cs.toronto.edu/~mvolkovs/ICML2020_tfixup.pdf
-        self.ent_encoders = torch.nn.ModuleList(
-            [
-                LinearEncoder(
-                    self.self_size + ent_size,
-                    1,
-                    embedding_size,
-                    kernel_init=Initialization.Normal,
-                    kernel_gain=(0.125 / embedding_size) ** 0.5,
-                )
-                for ent_size in self.entity_sizes
-            ]
+        self.self_ent_encoder = LinearEncoder(
+            self.entity_size,
+            1,
+            self.embedding_size,
+            kernel_init=Initialization.Normal,
+            kernel_gain=(0.125 / self.embedding_size) ** 0.5,
         )
-        self.embedding_norm = LayerNorm()
 
-    def forward(
-        self, x_self: torch.Tensor, entities: List[torch.Tensor]
-    ) -> Tuple[torch.Tensor, int]:
-        if self.concat_self:
-            # Concatenate all observations with self
-            self_and_ent: List[torch.Tensor] = []
-            for num_entities, ent in zip(self.entity_num_max_elements, entities):
-                if num_entities < 0:
-                    if exporting_to_onnx.is_exporting():
-                        raise UnityTrainerException(
-                            "Trying to export an attention mechanism that doesn't have a set max \
-                            number of elements."
-                        )
-                    num_entities = ent.shape[1]
-                expanded_self = x_self.reshape(-1, 1, self.self_size)
-                expanded_self = torch.cat([expanded_self] * num_entities, dim=1)
-                self_and_ent.append(torch.cat([expanded_self, ent], dim=2))
-        else:
-            self_and_ent = entities
-            # Encode and concatenate entites
-        encoded_entities = torch.cat(
-            [ent_encoder(x) for ent_encoder, x in zip(self.ent_encoders, self_and_ent)],
-            dim=1,
+    def add_self_embedding(self, size: int) -> None:
+        self.self_size = size
+        self.self_ent_encoder = LinearEncoder(
+            self.self_size + self.entity_size,
+            1,
+            self.embedding_size,
+            kernel_init=Initialization.Normal,
+            kernel_gain=(0.125 / self.embedding_size) ** 0.5,
         )
-        encoded_entities = self.embedding_norm(encoded_entities)
-        return encoded_entities
 
-    @staticmethod
-    def get_masks(observations: List[torch.Tensor]) -> List[torch.Tensor]:
-        """
-        Takes a List of Tensors and returns a List of mask Tensor with 1 if the input was
-        all zeros (on dimension 2) and 0 otherwise. This is used in the Attention
-        layer to mask the padding observations.
-        """
-        with torch.no_grad():
-            # Generate the masking tensors for each entities tensor (mask only if all zeros)
-            key_masks: List[torch.Tensor] = [
-                (torch.sum(ent ** 2, axis=2) < 0.01).float() for ent in observations
-            ]
-        return key_masks
+    def forward(self, x_self: torch.Tensor, entities: torch.Tensor) -> torch.Tensor:
+        if self.self_size > 0:
+            num_entities = self.entity_num_max_elements
+            if num_entities < 0:
+                if exporting_to_onnx.is_exporting():
+                    raise UnityTrainerException(
+                        "Trying to export an attention mechanism that doesn't have a set max \
+                        number of elements."
+                    )
+                num_entities = entities.shape[1]
+            expanded_self = x_self.reshape(-1, 1, self.self_size)
+            expanded_self = torch.cat([expanded_self] * num_entities, dim=1)
+            # Concatenate all observations with self
+            entities = torch.cat([expanded_self, entities], dim=2)
+        # Encode entities
+        encoded_entities = self.self_ent_encoder(entities)
+        return encoded_entities
 
 
 class ResidualSelfAttention(torch.nn.Module):
     """
     Residual self attentioninspired from https://arxiv.org/pdf/1909.07528.pdf. Can be used
-    with an EntityEmbeddings module, to apply multi head self attention to encode information
+    with an EntityEmbedding module, to apply multi head self attention to encode information
     about a "Self" and a list of relevant "Entities".
     """
 
@@ -189,7 +182,7 @@ class ResidualSelfAttention(torch.nn.Module):
     def __init__(
         self,
         embedding_size: int,
-        entity_num_max_elements: Optional[List[int]] = None,
+        entity_num_max_elements: Optional[int] = None,
         num_heads: int = 4,
     ):
         """
@@ -205,8 +198,7 @@ def __init__(
         super().__init__()
         self.max_num_ent: Optional[int] = None
         if entity_num_max_elements is not None:
-            _entity_num_max_elements = entity_num_max_elements
-            self.max_num_ent = sum(_entity_num_max_elements)
+            self.max_num_ent = entity_num_max_elements
 
         self.attention = MultiHeadAttention(
             num_heads=num_heads, embedding_size=embedding_size
@@ -237,11 +229,14 @@ def __init__(
             kernel_init=Initialization.Normal,
             kernel_gain=(0.125 / embedding_size) ** 0.5,
         )
+        self.embedding_norm = LayerNorm()
         self.residual_norm = LayerNorm()
 
     def forward(self, inp: torch.Tensor, key_masks: List[torch.Tensor]) -> torch.Tensor:
         # Gather the maximum number of entities information
         mask = torch.cat(key_masks, dim=1)
+
+        inp = self.embedding_norm(inp)
         # Feed to self attention
         query = self.fc_q(inp)  # (b, n_q, emb)
         key = self.fc_k(inp)  # (b, n_k, emb)