tensor deduplication working now

_bighash.py

@@ -45,7 +45,7 @@ def tensor_python_tuple_hash(items, out_or_in_place):
 
 def hash(buffer, *incoming_unhashed_ints):
     if len(buffer) < 16:
-        return buffer.__hash__()
+        return bytes(buffer).__hash__()
 
     # first pass
     # - allocate storage

_safetensors.py

@@ -15,24 +15,25 @@ class WritingSafeTensors:
         self.weight_map = {}
         self.hash_map = {}
     def add(self, name, tensor):
-        print(name, '...')
         tensor_hash = self.file.add(name, tensor)
-        image_of = self.hash_map.get(tensor_hash)
-        if image_of is not None:
-            self.file.undo(name, tensor)
+        image_of = self.hash_map.setdefault(tensor_hash, name)
+        if image_of is not name:
+            self.file.undo(name, tensor, True)
             imaged_hash = self.weight_map[image_of].add(name, tensor, image_of)
             assert imaged_hash == tensor_hash
-        elif self.file.size >= self.file_size:
-            self.file.undo(name, tensor)
-            ct = len(self.files)
-            if len(self.files) == 1:
-                self.file.rename(f'{self.name}-{ct:05}.safetensors')
-                self.file.set_metadata(index = str(ct))
-                self.files = {self.file.filename:self.file}
-            ct += 1
-            self.file = self.File(f'{self.name}-{ct:05}.safetensors', index = ct)
-            self.files[self.file.filename] = self.file
-            self.file.add(name, tensor)
+        else:
+            print(name, '...')
+            if self.file.size >= self.file_size:
+                self.file.undo(name, tensor, False)
+                ct = len(self.files)
+                if len(self.files) == 1:
+                    self.file.rename(f'{self.name}-{ct:05}.safetensors')
+                    self.file.set_metadata(index = str(ct))
+                    self.files = {self.file.filename:self.file}
+                ct += 1
+                self.file = self.File(f'{self.name}-{ct:05}.safetensors', index = ct)
+                self.files[self.file.filename] = self.file
+                self.file.add(name, tensor)
             self.weight_map[name] = self.file
     def finalize(self):
         if len(self.files) == 1:
@@ -114,6 +115,8 @@ class WritingSafeTensors:
                         self.mmapview[start : end],
                         dtype=tensor.dtype, count=tensor.numel(),
                 ).view(tensor.shape or [1])[:] = tensor
+                assert end >= self.size
+                self.size = end
             else:
                 image = self.header[image_of]
                 start, end = image['data_offsets']
@@ -121,7 +124,7 @@ class WritingSafeTensors:
                 assert (tensor == torch.frombuffer(
                         self.mmapview[start : end],
                         dtype=tensor.dtype, count=tensor.numel(),
-                ).view(tensor.shape or [1])[:]).all()
+                ).view(tensor.shape)).all()
 
             tensor.flatten()
             tensor_hash = hash(self.mmapview[start : end])
@@ -140,12 +143,13 @@ class WritingSafeTensors:
                 'data_offsets':
                     [start, end],
             }
-            self.size = end
             return tensor_hash
-        def undo(self, name, tensor):
-            length = tensor.numel() * tensor.dtype.itemsize
+        def undo(self, name, tensor, is_image):
+            if not is_image:
+                length = tensor.numel() * tensor.dtype.itemsize
+                assert [self.size - length, self.size] == self.header[name]['data_offsets']
+                self.size -= length
             del self.header[name]
-            self.size -= length
         def set_metadata(self, **metadata):
             m = self.header['__metadata__']
             for k, v in metadata.items():