costiash
commented
1 year ago It appears that the problem is related to KerasCV's implementation because they are using their own implementation of the GroupNormalization layer, which is causing this error, whereas this repo's implementation uses the GroupNormalization layer from tensorflow-addons. I was able to save the diffusion model without errors by using this repo.
!pip install git+https://github.com/divamgupta/stable-diffusion-tensorflow --upgrade --quiet
!pip install tensorflow tensorflow_addons ftfy --upgrade --quiet
from stable_diffusion_tf.stable_diffusion import StableDiffusion
model = StableDiffusion(
img_height=512,
img_width=512,
)
model.diffusion_model.save('/content/saved_model')**EDIT: While saving is possible, converting to TFLite appears to be impossible. It appears that the protobuf has a size limit of 2GB. Because the saved model exceeds 3GB, the conversion returns the following error:
[/usr/local/lib/python3.7/dist-packages/tensorflow/lite/python/lite.py] in wrapper(self, *args, **kwargs)
931 def wrapper(self, *args, **kwargs):
932 # pylint: disable=protected-access
--> 933 return self._convert_and_export_metrics(convert_func, *args, **kwargs)
934 # pylint: enable=protected-access
935
[/usr/local/lib/python3.7/dist-packages/tensorflow/lite/python/lite.py] in _convert_and_export_metrics(self, convert_func, *args, **kwargs)
909 self._save_conversion_params_metric()
910 start_time = time.process_time()
--> 911 result = convert_func(self, *args, **kwargs)
912 elapsed_time_ms = (time.process_time() - start_time) * 1000
913 if result:
[/usr/local/lib/python3.7/dist-packages/tensorflow/lite/python/lite.py] in convert(self)
1345
1346 graph_def, input_tensors, output_tensors, frozen_func = (
-> 1347 self._freeze_keras_model())
1348
1349 graph_def = self._optimize_tf_model(graph_def, input_tensors,
[/usr/local/lib/python3.7/dist-packages/tensorflow/lite/python/convert_phase.py] in wrapper(*args, **kwargs)
213 except Exception as error:
214 report_error_message(str(error))
--> 215 raise error from None # Re-throws the exception.
216
217 return wrapper
[/usr/local/lib/python3.7/dist-packages/tensorflow/lite/python/convert_phase.py] in wrapper(*args, **kwargs)
203 def wrapper(*args, **kwargs):
204 try:
--> 205 return func(*args, **kwargs)
206 except ConverterError as converter_error:
207 if converter_error.errors:
[/usr/local/lib/python3.7/dist-packages/tensorflow/lite/python/lite.py] in _freeze_keras_model(self)
1302 frozen_func, graph_def = (
1303 _convert_to_constants.convert_variables_to_constants_v2_as_graph(
-> 1304 self._funcs[0], lower_control_flow=False))
1305
1306 input_tensors = [
[/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/convert_to_constants.py] in convert_variables_to_constants_v2_as_graph(func, lower_control_flow, aggressive_inlining)
1249
1250 frozen_func = _construct_concrete_function(func, output_graph_def,
-> 1251 converted_input_indices)
1252 return frozen_func, output_graph_def
1253
[/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/convert_to_constants.py] in _construct_concrete_function(func, output_graph_def, converted_input_indices)
1094 new_func = wrap_function.function_from_graph_def(output_graph_def,
1095 new_input_names,
-> 1096 new_output_names)
1097
1098 # Manually propagate shape for input tensors where the shape is not correctly
[/usr/local/lib/python3.7/dist-packages/tensorflow/python/eager/wrap_function.py] in function_from_graph_def(graph_def, inputs, outputs, captures)
653 graph.add_capture(captures[c], graph.get_tensor_by_name(str(c) + ":0"))
654
--> 655 wrapped_import = wrap_function(_imports_graph_def, [])
656 import_graph = wrapped_import.graph
657 return wrapped_import.prune(
[/usr/local/lib/python3.7/dist-packages/tensorflow/python/eager/wrap_function.py] in wrap_function(fn, signature, name)
624 signature=signature,
625 add_control_dependencies=False,
--> 626 collections={}),
627 variable_holder=holder,
628 signature=signature)
[/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/func_graph.py] in func_graph_from_py_func(name, python_func, args, kwargs, signature, func_graph, autograph, autograph_options, add_control_dependencies, arg_names, op_return_value, collections, capture_by_value, acd_record_initial_resource_uses)
1281 _, original_func = tf_decorator.unwrap(python_func)
1282
-> 1283 func_outputs = python_func(*func_args, **func_kwargs)
1284
1285 # invariant: `func_outputs` contains only Tensors, CompositeTensors,
[/usr/local/lib/python3.7/dist-packages/tensorflow/python/eager/wrap_function.py] in __call__(self, *args, **kwargs)
81
82 def __call__(self, *args, **kwargs):
---> 83 return self.call_with_variable_creator_scope(self._fn)(*args, **kwargs)
84
85 def call_with_variable_creator_scope(self, fn):
[/usr/local/lib/python3.7/dist-packages/tensorflow/python/eager/wrap_function.py] in wrapped(*args, **kwargs)
87 def wrapped(*args, **kwargs):
88 with variable_scope.variable_creator_scope(self.variable_creator_scope):
---> 89 return fn(*args, **kwargs)
90
91 return wrapped
[/usr/local/lib/python3.7/dist-packages/tensorflow/python/eager/wrap_function.py] in _imports_graph_def()
647
648 def _imports_graph_def():
--> 649 importer.import_graph_def(graph_def, name="")
650 graph = ops.get_default_graph()
651 if captures is not None:
[/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/deprecation.py] in new_func(*args, **kwargs)
559 'in a future version' if date is None else ('after %s' % date),
560 instructions)
--> 561 return func(*args, **kwargs)
562
563 doc = _add_deprecated_arg_notice_to_docstring(
[/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/importer.py] in import_graph_def(***failed resolving arguments***)
406 return_elements=return_elements,
407 name=name,
--> 408 producer_op_list=producer_op_list)
409
410
[/usr/local/lib/python3.7/dist-packages/tensorflow/python/framework/importer.py] in _import_graph_def_internal(graph_def, input_map, return_elements, validate_colocation_constraints, name, producer_op_list)
495 # _ProcessNewOps.
496 with graph._mutation_lock(): # pylint: disable=protected-access
--> 497 with c_api_util.tf_buffer(graph_def.SerializeToString()) as serialized:
498 try:
499 with graph._c_graph.get() as c_graph: # pylint: disable=protected-access
ValueError: Message tensorflow.GraphDef exceeds maximum protobuf size of 2GB: 3440790027
charbull
joe-albert
soten355
rvalitov
SkyyySi
Christianb233
Hi, Leaving this here in case someone is also trying to convert to a TF lite model.
From the keras_cv documentation:
Load the model:
It seems a similar error occurs when trying to save the model.
The error in conversion: