Greatly speed up MSA processing code

openfold/data/data_transforms.py

@@ -45,6 +45,7 @@ def cast_to_64bit_ints(protein):
     for k, v in protein.items():
         if v.dtype == torch.int32:
             protein[k] = v.type(torch.int64)
+
     return protein
 
 
@@ -97,6 +98,7 @@ def fix_templates_aatype(protein):
         protein["template_aatype"] = torch.gather(
             new_order, 1, index=protein["template_aatype"]
         )
+
     return protein
 
 
@@ -120,6 +122,7 @@ def correct_msa_restypes(protein):
                 22,
             ], "num_dim for %s out of expected range: %s" % (k, num_dim)
             protein[k] = torch.dot(protein[k], perm_matrix[:num_dim, :num_dim])
+    
     return protein
 
 
@@ -147,6 +150,7 @@ def squeeze_features(protein):
     for k in ["seq_length", "num_alignments"]:
         if k in protein:
             protein[k] = protein[k][0]
+
     return protein
 
 
@@ -169,6 +173,7 @@ def randomly_replace_msa_with_unknown(protein, replace_proportion):
     )
     return protein
 
+
 @curry1
 def sample_msa(protein, max_seq, keep_extra, seed=None):
     """Sample MSA randomly, remaining sequences are stored are stored as `extra_*`."""
@@ -190,6 +195,7 @@ def sample_msa(protein, max_seq, keep_extra, seed=None):
                     protein[k], 0, not_sel_seq
                 )
             protein[k] = torch.index_select(protein[k], 0, sel_seq)
+
     return protein
 
 
@@ -210,6 +216,7 @@ def crop_extra_msa(protein, max_extra_msa):
             protein["extra_" + k] = torch.index_select(
                 protein["extra_" + k], 0, select_indices
             )
+    
     return protein
 
 
@@ -290,28 +297,24 @@ def nearest_neighbor_clusters(protein, gap_agreement_weight=0.0):
 
 def unsorted_segment_sum(data, segment_ids, num_segments):
     """
-    Computes the sum along segments of a tensor. Analogous to tf.unsorted_segment_sum.
+    Computes the sum along segments of a tensor. Similar to 
+    tf.unsorted_segment_sum, but only supports 1-D indices.
 
     :param data: A tensor whose segments are to be summed.
-    :param segment_ids: The segment indices tensor.
+    :param segment_ids: The 1-D segment indices tensor.
     :param num_segments: The number of segments.
     :return: A tensor of same data type as the data argument.
     """
-    # segment_ids.shape should be a prefix of data.shape
-    assert all([i in data.shape for i in segment_ids.shape])
-
-    # segment_ids is a 1-D tensor repeat it to have the same shape as data
-    if len(segment_ids.shape) == 1:
-        s = torch.prod(torch.tensor(data.shape[1:])).long()
-        segment_ids = segment_ids.repeat_interleave(s).view(
-            segment_ids.shape[0], *data.shape[1:]
+    assert (
+        len(segment_ids.shape) == 1 and
+        segment_ids.shape[0] == data.shape[0]
     )
-
-    # data.shape and segment_ids.shape should be equal
-    assert data.shape == segment_ids.shape
-
+    segment_ids = segment_ids.view(
+        segment_ids.shape[0], *((1,) * len(data.shape[1:]))
+    )
+    segment_ids = segment_ids.expand(data.shape)
     shape = [num_segments] + list(data.shape[1:])
-    tensor = torch.zeros(*shape).scatter_add(0, segment_ids, data.float())
+    tensor = torch.zeros(*shape).scatter_add_(0, segment_ids, data.float())
     tensor = tensor.type(data.dtype)
     return tensor
 
@@ -332,7 +335,6 @@ def summarize_clusters(protein):
     msa_sum = csum(mask[:, :, None] * make_one_hot(protein["extra_msa"], 23))
     msa_sum += make_one_hot(protein["msa"], 23)  # Original sequence
     protein["cluster_profile"] = msa_sum / mask_counts[:, :, None]
-
     del msa_sum
 
     del_sum = csum(mask * protein["extra_deletion_matrix"])
@@ -464,7 +466,6 @@ def make_fixed_size(
     num_templates=0,
 ):
     """Guess at the MSA and sequence dimension to make fixed size."""
-
     pad_size_map = {
         NUM_RES: num_res,
         NUM_MSA_SEQ: msa_cluster_size,
@@ -1169,4 +1170,5 @@ def random_crop_to_size(
         protein[k] = v[slices]
 
     protein["seq_length"] = protein["seq_length"].new_tensor(num_res_crop_size)
+    
     return protein

openfold/data/input_pipeline.py

@@ -175,6 +175,7 @@ def process_tensors_from_config(tensors, common_cfg, mode_cfg):
         common_cfg,
         mode_cfg,
     )
+
     tensors = compose(nonensembled)(tensors)
 
     if("no_recycling_iters" in tensors):