Support for Tensorflow 2.4

[lucas-langlois]

Hi,
I've just acquired a new machine with the lastest GPU RTX3090. This GPU is only supported for CUDA 11 meaning that I can only use the latest version of tensorflow (2.4).
I know that for now segmentation-models only works for tensorflow 2.1. I would really like to use your library to do some semantic segmentation but it seems that I am stuck.
Would you be able to estimate when tensorflow 2.4 support might be available ? Or any other ideas for me in the midtime ?

Thank you very much

[pluniak]

Are you sure it's not working? I'm using the repo with TF2.3.0.

[lucas-langlois]

Yeah I had it running fine previously. But now with Tensorflow 2.4.3 (the only one that would recognize the RTX3090) here are the error message I'm getting :

1. module 'keras.utils' has no attribute 'generic_utils' = found the way around that with %env SM_FRAMEWORK=tf.keras before loading segmentation-models
2. when training the model

ValueError Traceback (most recent call last)
in
1 # train model
----> 2 history = model.fit_generator(
3 train_dataloader,
4 steps_per_epoch=len(train_dataloader),
5 epochs=EPOCHS,

\AppData\Local\R-MINI1\envs\tensorflow24\lib\site-packages\tensorflow\python\keras\engine\training.py in fit_generator(self, generator, steps_per_epoch, epochs, verbose, callbacks, validation_data, validation_steps, validation_freq, class_weight, max_queue_size, workers, use_multiprocessing, shuffle, initial_epoch)
1845 'will be removed in a future version. '
1846 'Please use Model.fit, which supports generators.')
-> 1847 return self.fit(
1848 generator,
1849 steps_per_epoch=steps_per_epoch,

\AppData\Local\R-MINI1\envs\tensorflow24\lib\site-packages\tensorflow\python\keras\engine\training.py in fit(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_batch_size, validation_freq, max_queue_size, workers, use_multiprocessing)
1098 _r=1):
1099 callbacks.on_train_batch_begin(step)
-> 1100 tmp_logs = self.train_function(iterator)
1101 if data_handler.should_sync:
1102 context.async_wait()

\AppData\Local\R-MINI1\envs\tensorflow24\lib\site-packages\tensorflow\python\eager\def_function.py in call(self, *args, **kwds)
826 tracing_count = self.experimental_get_tracing_count()
827 with trace.Trace(self._name) as tm:
--> 828 result = self._call(*args, **kwds)
829 compiler = "xla" if self._experimental_compile else "nonXla"
830 new_tracing_count = self.experimental_get_tracing_count()

\AppData\Local\R-MINI1\envs\tensorflow24\lib\site-packages\tensorflow\python\eager\def_function.py in _call(self, *args, **kwds)
869 # This is the first call of call, so we have to initialize.
870 initializers = []
--> 871 self._initialize(args, kwds, add_initializers_to=initializers)
872 finally:
873 # At this point we know that the initialization is complete (or less

\AppData\Local\R-MINI1\envs\tensorflow24\lib\site-packages\tensorflow\python\eager\def_function.py in _initialize(self, args, kwds, add_initializers_to)
723 self._graph_deleter = FunctionDeleter(self._lifted_initializer_graph)
724 self._concrete_stateful_fn = (
--> 725 self._stateful_fn._get_concrete_function_internal_garbage_collected( # pylint: disable=protected-access
726 *args, **kwds))
727

\AppData\Local\R-MINI1\envs\tensorflow24\lib\site-packages\tensorflow\python\eager\function.py in _get_concrete_function_internal_garbage_collected(self, *args, **kwargs)
2967 args, kwargs = None, None
2968 with self._lock:
-> 2969 graph_function, _ = self._maybe_define_function(args, kwargs)
2970 return graph_function
2971

\AppData\Local\R-MINI1\envs\tensorflow24\lib\site-packages\tensorflow\python\eager\function.py in _maybe_define_function(self, args, kwargs)
3359
3360 self._function_cache.missed.add(call_context_key)
-> 3361 graph_function = self._create_graph_function(args, kwargs)
3362 self._function_cache.primary[cache_key] = graph_function
3363

\AppData\Local\R-MINI1\envs\tensorflow24\lib\site-packages\tensorflow\python\eager\function.py in _create_graph_function(self, args, kwargs, override_flat_arg_shapes)
3194 arg_names = base_arg_names + missing_arg_names
3195 graph_function = ConcreteFunction(
-> 3196 func_graph_module.func_graph_from_py_func(
3197 self._name,
3198 self._python_function,

\AppData\Local\R-MINI1\envs\tensorflow24\lib\site-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, override_flat_arg_shapes)
988 _, original_func = tf_decorator.unwrap(python_func)
989
--> 990 func_outputs = python_func(*func_args, **func_kwargs)
991
992 # invariant: func_outputs contains only Tensors, CompositeTensors,

\AppData\Local\R-MINI1\envs\tensorflow24\lib\site-packages\tensorflow\python\eager\def_function.py in wrapped_fn(*args, **kwds)
632 xla_context.Exit()
633 else:
--> 634 out = weak_wrapped_fn().wrapped(*args, **kwds)
635 return out
636

\AppData\Local\R-MINI1\envs\tensorflow24\lib\site-packages\tensorflow\python\framework\func_graph.py in wrapper(*args, **kwargs)
975 except Exception as e: # pylint:disable=broad-except
976 if hasattr(e, "ag_error_metadata"):
--> 977 raise e.ag_error_metadata.to_exception(e)
978 else:
979 raise

ValueError: in user code:

C:\Users\jc238248\AppData\Local\R-MINI~1\envs\tensorflow24\lib\site-packages\tensorflow\python\keras\engine\training.py:805 train_function  *
    return step_function(self, iterator)
C:\Users\jc238248\AppData\Local\R-MINI~1\envs\tensorflow24\lib\site-packages\tensorflow\python\keras\engine\training.py:795 step_function  **
    outputs = model.distribute_strategy.run(run_step, args=(data,))
C:\Users\jc238248\AppData\Local\R-MINI~1\envs\tensorflow24\lib\site-packages\tensorflow\python\distribute\distribute_lib.py:1259 run
    return self._extended.call_for_each_replica(fn, args=args, kwargs=kwargs)
C:\Users\jc238248\AppData\Local\R-MINI~1\envs\tensorflow24\lib\site-packages\tensorflow\python\distribute\distribute_lib.py:2730 call_for_each_replica
    return self._call_for_each_replica(fn, args, kwargs)
C:\Users\jc238248\AppData\Local\R-MINI~1\envs\tensorflow24\lib\site-packages\tensorflow\python\distribute\distribute_lib.py:3417 _call_for_each_replica
    return fn(*args, **kwargs)
C:\Users\jc238248\AppData\Local\R-MINI~1\envs\tensorflow24\lib\site-packages\tensorflow\python\keras\engine\training.py:788 run_step  **
    outputs = model.train_step(data)
C:\Users\jc238248\AppData\Local\R-MINI~1\envs\tensorflow24\lib\site-packages\tensorflow\python\keras\engine\training.py:754 train_step
    y_pred = self(x, training=True)
C:\Users\jc238248\AppData\Local\R-MINI~1\envs\tensorflow24\lib\site-packages\tensorflow\python\keras\engine\base_layer.py:998 __call__
    input_spec.assert_input_compatibility(self.input_spec, inputs, self.name)
C:\Users\jc238248\AppData\Local\R-MINI~1\envs\tensorflow24\lib\site-packages\tensorflow\python\keras\engine\input_spec.py:204 assert_input_compatibility
    raise ValueError('Layer ' + layer_name + ' expects ' +

ValueError: Layer model_3 expects 1 input(s), but it received 2 input tensors. Inputs received: [<tf.Tensor 'IteratorGetNext:0' shape=(None, None, None, None) dtype=uint8>, <tf.Tensor 'IteratorGetNext:1' shape=(None, None, None, None) dtype=float32>]

[pluniak]

Sorry, can't help you with that.

[WeiChihChern]

Can confirm this repo works with my 3090: a docker container in linux.

[JordanMakesMaps]

See #412 if you haven't already, it looks like it might be related?

[lucas-langlois]

Yes that was it! Works like a charm now.
Thank you very much for the help. No chance I would have been able to debug that by myself.

[Kritz23]

Yes that was it! Works like a charm now.
Thank you very much for the help. No chance I would have been able to debug that by myself.

Hey, I tried the same approach its working fine, but I'm not able to use my GPU. Can you assist?

Thanks in advance!