keras create/load model fails on Apple M2 cpu. Same code works on Ryzen #46

drmatthewclark · 2023-07-06T17:08:55Z

You can obtain the TensorFlow version with:
python -c "import tensorflow as tf; print(tf.version.GIT_VERSION, tf.version.VERSION)"
result:
v2.13.0-rc2-7-g1cb1a030a62 2.13.0

Can't load a keras model that I just created. The issue is not specific to Adam; all of the optimizers have the same issue.

Describe the problem.

code:

  from tensorflow import keras
  
  
  optimizer = keras.optimizers.Adam()
  
  vh  = keras.Input(shape=(2,3), name = 'vh')
  v1  = keras.layers.Dense(512)(vh)
  
  output  = keras.layers.Dense(1, activation='softmax', name='prediction')(v1)
  
  model = keras.Model(inputs=vh, outputs=[output], name="antibody_model")
  
  model.compile(optimizer=optimizer )
  
  model.save('nn_model.keras')
  test = keras.models.load_model('nn_model.keras')

output:

the legacy Adam is missing the method "build". This same code works on non-mac platforms.

WARNING:absl:At this time, the v2.11+ optimizer tf.keras.optimizers.Adam runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at tf.keras.optimizers.legacy.Adam.
WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., tf.keras.optimizers.legacy.Adam.
WARNING:absl:At this time, the v2.11+ optimizer tf.keras.optimizers.Adam runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at tf.keras.optimizers.legacy.Adam.
WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., tf.keras.optimizers.legacy.Adam.
Traceback (most recent call last):
File "/Users/matthewclark/eclipse-workspace/antibody_classification/test/test.py", line 17, in
test = keras.models.load_model('nn_model.keras')
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/Shared/anaconda3/envs/ability/lib/python3.11/site-packages/keras/src/saving/saving_api.py", line 230, in load_model
return saving_lib.load_model(
^^^^^^^^^^^^^^^^^^^^^^
File "/Users/Shared/anaconda3/envs/ability/lib/python3.11/site-packages/keras/src/saving/saving_lib.py", line 275, in load_model
raise e
File "/Users/Shared/anaconda3/envs/ability/lib/python3.11/site-packages/keras/src/saving/saving_lib.py", line 240, in load_model
model = deserialize_keras_object(
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/Shared/anaconda3/envs/ability/lib/python3.11/site-packages/keras/src/saving/serialization_lib.py", line 710, in deserialize_keras_object
instance.compile_from_config(compile_config)
File "/Users/Shared/anaconda3/envs/ability/lib/python3.11/site-packages/keras/src/engine/training.py", line 3582, in compile_from_config
self.optimizer.build(self.trainable_variables)
^^^^^^^^^^^^^^^^^^^^
File "/Users/Shared/anaconda3/envs/ability/lib/python3.11/site-packages/keras/src/optimizers/legacy/optimizer_v2.py", line 997, in getattribute
raise e
File "/Users/Shared/anaconda3/envs/ability/lib/python3.11/site-packages/keras/src/optimizers/legacy/optimizer_v2.py", line 987, in getattribute
return super().getattribute(name)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: 'Adam' object has no attribute 'build'

The text was updated successfully, but these errors were encountered:

tilakrayal · 2023-07-07T08:40:22Z

@drmatthewclark,
I tried to execute the code on both tensorflow v2.12 and v2.13 and it was executed without any error/issue. Kindly find the gist of it here and try to test your code in the virtual environment. Thank you!

jtpotato · 2023-07-18T12:48:32Z

Can confirm having the same issue (M1). Haven't had the chance to test on non-Apple Silicon hardware (that I own) yet, but it appears to be fine on Colab...

bossebandowski · 2023-07-19T11:54:16Z

Same here, on M1, python3.10, and tensorflow 2.13.0. Training, inference, and saving works fine. Loading fails.

Code extract:

train(model, train_ds, val_ds)
acc = model.evaluate(test_ds)[1]
print(f"test accuracy: {acc}")
model.save(MODEL_PATH)
test = tf.keras.models.load_model(MODEL_PATH)
print("successfully loaded model")

Output:

[...]
test accuracy: 0.9336022138595581
Traceback (most recent call last):
  File "[my_file]", line 252, in <module>
    test = tf.keras.models.load_model(MODEL_PATH)
  File "[my_venv]/site-packages/keras/src/saving/saving_api.py", line 230, in load_model
    return saving_lib.load_model(
  File "[my_venv]/site-packages/keras/src/saving/saving_lib.py", line 275, in load_model
    raise e
  File "[my_venv]/site-packages/keras/src/saving/saving_lib.py", line 240, in load_model
    model = deserialize_keras_object(
  File "[my_venv]/site-packages/keras/src/saving/serialization_lib.py", line 710, in deserialize_keras_object
    instance.compile_from_config(compile_config)
  File "[my_venv]/site-packages/keras/src/engine/training.py", line 3582, in compile_from_config
    self.optimizer.build(self.trainable_variables)
  File "[my_venv]/site-packages/keras/src/optimizers/legacy/optimizer_v2.py", line 997, in __getattribute__
    raise e
  File "[my_venv]/site-packages/keras/src/optimizers/legacy/optimizer_v2.py", line 987, in __getattribute__
    return super().__getattribute__(name)
AttributeError: 'Adam' object has no attribute 'build'

tilakrayal · 2023-07-20T08:05:06Z

@drmatthewclark,
I tried to execute the mentioned above code on both tensorflow v2.12 and tf-nighty(2.14.0-dev20230622), and it was executed without any issue/error. Kindly find the gist of it here
As it is failing only on tensorflow-macos, we request to raise the concern on the macos-apple forum for the quick resolution. Thank you!

bossebandowski · 2023-07-26T08:23:09Z

Hey @tilakrayal,

thank you for your input. I did a bit of digging because I found the issue too annoying and it is clearly a tensorflow problem and not a macos problem. So, here's what I found:

As you can easily see from the backtrace in my previous comment, loading a model on the M1/M2 macs causes the legacy optimizers to be loaded instead of the latest code. Sure enough, the legacy code does not have a build method while the latest code does. So, why is the legacy code executed in the first place?
To get an optimizer, the optimizers.get() method is called which is defined here on the latest tag v2.13.1. Whenever we have a mac that runs on the new chips, keras falls back to the legacy optimizers which do not work.
As of commit 5ddf84a, this issue seems to be fixed on master but there are no releases that contain the update.

[EDIT] @drmatthewclark to fix your problem install the nightly version of tensorflow and keras while we wait for a new release.

pip install tf-nightly keras-nightly

tilakrayal · 2023-08-03T08:10:51Z

There is an alternative workaround to make AdamW work on Apple Silicon with the latest version of tensorflow, tensorflow-addons.
All you need to do is to import AdamW from tensorflow_addons.optimizers. Could you please try and let us know if this works in your case.

github-actions · 2023-08-18T01:43:11Z

This issue is stale because it has been open for 14 days with no activity. It will be closed if no further activity occurs. Thank you.

mbergwall2222 · 2023-08-18T21:12:53Z

Can confirm same issue on m2 max mac. nightly tensorflow builds don't work and neither does tensorflow-addons.

tcivie · 2023-08-20T11:32:01Z

import AdamW from tensorflow_addons.optimizers

Also tried this and got the same result (M2 Mac):
AttributeError: 'AdamW' object has no attribute 'build'

Tried also the suggested solutions from this thread without any success

norwend · 2023-08-27T21:04:18Z

Same error with the SGD omptimizer. M1 Macbook Air.

Traceback (most recent call last):
File "***", line 17, in
model = keras.models.load_model(
^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/homebrew/lib/python3.11/site-packages/keras/src/saving/saving_api.py", line 230, in load_model
return saving_lib.load_model(
^^^^^^^^^^^^^^^^^^^^^^
File "/opt/homebrew/lib/python3.11/site-packages/keras/src/saving/saving_lib.py", line 275, in load_model
raise e
File "/opt/homebrew/lib/python3.11/site-packages/keras/src/saving/saving_lib.py", line 240, in load_model
model = deserialize_keras_object(
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/homebrew/lib/python3.11/site-packages/keras/src/saving/serialization_lib.py", line 710, in deserialize_keras_object
instance.compile_from_config(compile_config)
File "/opt/homebrew/lib/python3.11/site-packages/keras/src/engine/training.py", line 3582, in compile_from_config
self.optimizer.build(self.trainable_variables)
^^^^^^^^^^^^^^^^^^^^
File "/opt/homebrew/lib/python3.11/site-packages/keras/src/optimizers/legacy/optimizer_v2.py", line 997, in getattribute
raise e
File "/opt/homebrew/lib/python3.11/site-packages/keras/src/optimizers/legacy/optimizer_v2.py", line 987, in getattribute
return super().getattribute(name)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: 'SGD' object has no attribute 'build'

RobP-git · 2023-09-04T09:26:18Z

same issue here also. M1 mac.
current workaround is to save and load using the legacy .h5 file extension.

Go back to use .h5 files instead of .keras due to a problem in Keras loading .keras files on MacOS ARM (see https://github.com/keras-team/keras/issues/18278 )

samhva · 2023-09-08T18:46:32Z

Same issue here, fix plz.
M2 Pro

ptn24 · 2023-09-09T07:51:50Z

Ditto. I am using TensorFlow 2.13 on macOS with M2.

Go back to use .h5 files instead of .keras due to a problem in Keras loading .keras files on MacOS ARM (see https://github.com/keras-team/keras/issues/18278 )

LiWangSH · 2023-09-13T23:04:11Z

Same issue, using Tensorflow 2.13.0 and Python 3.10.13 on Mac M2.

LiWangSH · 2023-09-14T03:25:39Z

Saving the model to a different format (h5) solved the issue^ for me.

Source: https://stackoverflow.com/questions/76311807/attributeerror-adam-object-has-no-attribute-build-during-unpickling/76311808#76311808

Go back to use .h5 files instead of .keras due to a problem in Keras loading .keras files on MacOS ARM (see https://github.com/keras-team/keras/issues/18278 )

palminha · 2023-09-19T10:20:58Z

any news on this issue?
we are having the same problem here with tensorflow 2.13 in MacOS M1

palminha · 2023-09-19T13:14:21Z

Saving the model to a different format (h5) solved the issue^ for me.

Source: https://stackoverflow.com/questions/76311807/attributeerror-adam-object-has-no-attribute-build-during-unpickling/76311808#76311808

This workaround is not suitable for us because the h5 format doesn't support some attributes/parameters of our models

palminha · 2023-09-19T13:21:01Z

tf_env.txt

You can obtain the TensorFlow version with: python -c "import tensorflow as tf; print(tf.version.GIT_VERSION, tf.version.VERSION)" result: v2.13.0-rc2-7-g1cb1a030a62 2.13.0

Can't load a keras model that I just created. The issue is not specific to Adam; all of the optimizers have the same issue.

Describe the problem.

code:
  from tensorflow import keras
  
  
  optimizer = keras.optimizers.Adam()
  
  vh  = keras.Input(shape=(2,3), name = 'vh')
  v1  = keras.layers.Dense(512)(vh)
  
  output  = keras.layers.Dense(1, activation='softmax', name='prediction')(v1)
  
  model = keras.Model(inputs=vh, outputs=[output], name="antibody_model")
  
  model.compile(optimizer=optimizer )
  
  model.save('nn_model.keras')
  test = keras.models.load_model('nn_model.keras')
output:

the legacy Adam is missing the method "build". This same code works on non-mac platforms.

WARNING:absl:At this time, the v2.11+ optimizer tf.keras.optimizers.Adam runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at tf.keras.optimizers.legacy.Adam. WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., tf.keras.optimizers.legacy.Adam. WARNING:absl:At this time, the v2.11+ optimizer tf.keras.optimizers.Adam runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at tf.keras.optimizers.legacy.Adam. WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., tf.keras.optimizers.legacy.Adam. Traceback (most recent call last): File "/Users/matthewclark/eclipse-workspace/antibody_classification/test/test.py", line 17, in test = keras.models.load_model('nn_model.keras') ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Shared/anaconda3/envs/ability/lib/python3.11/site-packages/keras/src/saving/saving_api.py", line 230, in load_model return saving_lib.load_model( ^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Shared/anaconda3/envs/ability/lib/python3.11/site-packages/keras/src/saving/saving_lib.py", line 275, in load_model raise e File "/Users/Shared/anaconda3/envs/ability/lib/python3.11/site-packages/keras/src/saving/saving_lib.py", line 240, in load_model model = deserialize_keras_object( ^^^^^^^^^^^^^^^^^^^^^^^^^ File "/Users/Shared/anaconda3/envs/ability/lib/python3.11/site-packages/keras/src/saving/serialization_lib.py", line 710, in deserialize_keras_object instance.compile_from_config(compile_config) File "/Users/Shared/anaconda3/envs/ability/lib/python3.11/site-packages/keras/src/engine/training.py", line 3582, in compile_from_config self.optimizer.build(self.trainable_variables) ^^^^^^^^^^^^^^^^^^^^ File "/Users/Shared/anaconda3/envs/ability/lib/python3.11/site-packages/keras/src/optimizers/legacy/optimizer_v2.py", line 997, in getattribute raise e File "/Users/Shared/anaconda3/envs/ability/lib/python3.11/site-packages/keras/src/optimizers/legacy/optimizer_v2.py", line 987, in getattribute return super().getattribute(name) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ AttributeError: 'Adam' object has no attribute 'build'

I just tried to run this code in my MacOS M1 (using tensorflow==2.13.0 and keras [required: >=2.13.1,<2.14, installed: 2.13.1]) and got the following error:

WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.
WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.
WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.
WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.
Traceback (most recent call last):
  File "/opt/homebrew/Cellar/[email protected]/3.10.13/Frameworks/Python.framework/Versions/3.10/lib/python3.10/runpy.py", line 196, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "/opt/homebrew/Cellar/[email protected]/3.10.13/Frameworks/Python.framework/Versions/3.10/lib/python3.10/runpy.py", line 86, in _run_code
    exec(code, run_globals)
  File "/Users/palmito/Development/federated-demo/backend/app/test.py", line 15, in <module>
    test = keras.models.load_model('nn_model.keras')
  File "/Users/palmito/.local/share/virtualenvs/backend-tPS_SUas/lib/python3.10/site-packages/keras/src/saving/saving_api.py", line 230, in load_model
    return saving_lib.load_model(
  File "/Users/palmito/.local/share/virtualenvs/backend-tPS_SUas/lib/python3.10/site-packages/keras/src/saving/saving_lib.py", line 275, in load_model
    raise e
  File "/Users/palmito/.local/share/virtualenvs/backend-tPS_SUas/lib/python3.10/site-packages/keras/src/saving/saving_lib.py", line 240, in load_model
    model = deserialize_keras_object(
  File "/Users/palmito/.local/share/virtualenvs/backend-tPS_SUas/lib/python3.10/site-packages/keras/src/saving/serialization_lib.py", line 710, in deserialize_keras_object
    instance.compile_from_config(compile_config)
  File "/Users/palmito/.local/share/virtualenvs/backend-tPS_SUas/lib/python3.10/site-packages/keras/src/engine/training.py", line 3582, in compile_from_config
    self.optimizer.build(self.trainable_variables)
  File "/Users/palmito/.local/share/virtualenvs/backend-tPS_SUas/lib/python3.10/site-packages/keras/src/optimizers/legacy/optimizer_v2.py", line 997, in __getattribute__
    raise e
  File "/Users/palmito/.local/share/virtualenvs/backend-tPS_SUas/lib/python3.10/site-packages/keras/src/optimizers/legacy/optimizer_v2.py", line 987, in __getattribute__
    return super().__getattribute__(name)
AttributeError: 'Adam' object has no attribute 'build'

qlzh727 · 2023-09-21T03:00:05Z

Adding Neel for the saving issue.

tilakrayal · 2023-09-28T08:44:19Z

@drmatthewclark,
I tried to execute the mentioned code on Apple M1 with the tf-nightly(2.15.0-dev20230927) and it was executed without any fail or error. Kindly find the screenshot below for the reference. And also please take a look at the PR which is related to the issue.
keras-team/keras#18492

Thank you!

palminha · 2023-09-28T09:40:31Z

@tilakrayal i confirm that with tf-nightly the error it seems to be fixed

% pipenv graph
...
tf-nightly==2.15.0.dev20230928
└── tf-nightly-macos [required: ==2.15.0-dev20230928, installed: 2.15.0.dev20230928]

it seems it solves the error

% pipenv run adam
WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.
WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.
WARNING:absl:Skipping variable loading for optimizer 'Adam', because it has 9 variables whereas the saved optimizer has 1 variables.

I also reported the same finding here: tensorflow/tensorflow#61915 (comment)

eswar-janjanam · 2023-09-29T10:56:42Z

I'm using mac m1 pro and tensorflow version is 2.14.0. I was facing the same error "AttributeError: 'Adam' object has no attribute 'build'" while loading the saved model, if I save the model as below

model = Sequential()
model.add(Embedding(MAX_FEATURES+1, 32))
model.add(Bidirectional(LSTM(32, activation='tanh')))
model.add(Dense(128, activation='relu'))
model.add(Dense(256, activation='relu'))
model.add(Dense(128, activation='relu'))
model.add(Dense(1, activation='sigmoid'))
model.compile(loss='BinaryCrossentropy', optimizer='Adam')
model.fit(train, epochs=1, validation_data=val)
model.save('data/model_data.keras')

I updated the optimizer as below and it is working fine now

import tensorflow as tf
model.compile(loss='BinaryCrossentropy', optimizer=tf.keras.optimizers.Adam())

nkovela1 · 2023-09-29T15:58:51Z

@EswarJanjanam The fix is on tf-nightly. Please do pip install tf-nightly and see if your code works, thanks!

phpmac · 2023-10-02T00:56:34Z

$pip install tf-nightly
Looking in indexes: https://pypi.tuna.tsinghua.edu.cn/simple
ERROR: Could not find a version that satisfies the requirement tf-nightly (from versions: none)
ERROR: No matching distribution found for tf-nightly

phpmac · 2023-10-02T01:01:19Z

$pip install tf-nightly
Looking in indexes: https://pypi.tuna.tsinghua.edu.cn/simple
ERROR: Could not find a version that satisfies the requirement tf-nightly (from versions: none)
ERROR: No matching distribution found for tf-nightly

Crap, it's my mirroring problem, sorry!

But it didn't solve the problem.


AttributeError: 'Adam' object has no attribute 'build'

hanamurayuki · 2023-10-16T13:34:03Z

$pip install tf-nightly
Looking in indexes: https://pypi.tuna.tsinghua.edu.cn/simple
ERROR: Could not find a version that satisfies the requirement tf-nightly (from versions: none)
ERROR: No matching distribution found for tf-nightly

Same error exists on M2 mac.

AttributeError: 'Adam' object has no attribute 'build'

.h5 format is one of solution instead of .keras format for now.
I hope I can save model as .keras format.

phpmac · 2023-10-16T21:43:14Z

$pip install tf-nightly
Looking in indexes: https://pypi.tuna.tsinghua.edu.cn/simple
ERROR: Could not find a version that satisfies the requirement tf-nightly (from versions: none)
ERROR: No matching distribution found for tf-nightly

Same error exists on M2 mac.
AttributeError: 'Adam' object has no attribute 'build'
.h5 format is one of solution instead of .keras format for now. I hope I can save model as .keras format.

Yeah, I mean can't he just automatically adapt to the solution?

MurtadhaInit · 2023-10-27T22:36:06Z

Having the exact same issue, and can confirm that tf-nightly is NOT working for me.

richardokonicha · 2024-03-03T00:40:11Z

same issue

saarangbond · 2024-06-17T20:24:54Z

Having the same issue as well on Mac M1. The same code works on cloud workstation on linux.

eyash24 · 2024-08-18T23:44:42Z

After much trial and error, installing keras-nightly solved the issue and my model started working.
pip install tf-nightly keras-nightly

google-ml-butler bot assigned tilakrayal Jul 6, 2023

tilakrayal added the stat:awaiting response from contributor label Jul 7, 2023

tilakrayal added stat:awaiting response from contributor and removed stat:awaiting response from contributor labels Jul 20, 2023

lmoneta referenced this issue in lmoneta/root Sep 4, 2023

[tmva][pymva] Use .h5 file instead of .keras

9946206

Go back to use .h5 files instead of .keras due to a problem in Keras loading .keras files on MacOS ARM (see https://github.com/keras-team/keras/issues/18278 )

lmoneta mentioned this issue Sep 4, 2023

[tmva][pymva] Use .h5 file instead of .keras root-project/root#13592

Merged

lmoneta referenced this issue in root-project/root Sep 6, 2023

[tmva][pymva] Use .h5 file instead of .keras

5e38958

Go back to use .h5 files instead of .keras due to a problem in Keras loading .keras files on MacOS ARM (see https://github.com/keras-team/keras/issues/18278 )

tilakrayal added type:bug and removed stat:awaiting response from contributor labels Sep 9, 2023

lmoneta referenced this issue in lmoneta/root Sep 11, 2023

[tmva][pymva] Use .h5 file instead of .keras

a3088ca

Go back to use .h5 files instead of .keras due to a problem in Keras loading .keras files on MacOS ARM (see https://github.com/keras-team/keras/issues/18278 )

lmoneta referenced this issue in lmoneta/root Sep 14, 2023

[tmva][pymva] Use .h5 file instead of .keras

8794972

Go back to use .h5 files instead of .keras due to a problem in Keras loading .keras files on MacOS ARM (see https://github.com/keras-team/keras/issues/18278 )

lmoneta referenced this issue in root-project/root Sep 14, 2023

[tmva][pymva] Use .h5 file instead of .keras

7793c20

Go back to use .h5 files instead of .keras due to a problem in Keras loading .keras files on MacOS ARM (see https://github.com/keras-team/keras/issues/18278 )

tilakrayal added the keras-team-review-pending label Sep 19, 2023

palminha mentioned this issue Sep 19, 2023

'Adam' object has no attribute 'build' (saving and loading keras.optimizers.Adam) tensorflow/tensorflow#61915

Closed

qlzh727 assigned nkovela1 Sep 21, 2023

qlzh727 removed the keras-team-review-pending label Sep 21, 2023

sachinprasadhs transferred this issue from keras-team/keras Sep 22, 2023

nkovela1 closed this as completed Sep 29, 2023

mikivee mentioned this issue Nov 7, 2024

Update optimizer path rewiringamerica/surrogate_modeling#26

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

keras create/load model fails on Apple M2 cpu. Same code works on Ryzen #46

keras create/load model fails on Apple M2 cpu. Same code works on Ryzen #46

drmatthewclark commented Jul 6, 2023

tilakrayal commented Jul 7, 2023

jtpotato commented Jul 18, 2023

bossebandowski commented Jul 19, 2023

tilakrayal commented Jul 20, 2023

bossebandowski commented Jul 26, 2023 •

edited

Loading

tilakrayal commented Aug 3, 2023

github-actions bot commented Aug 18, 2023

mbergwall2222 commented Aug 18, 2023

tcivie commented Aug 20, 2023

norwend commented Aug 27, 2023 •

edited

Loading

RobP-git commented Sep 4, 2023

samhva commented Sep 8, 2023

ptn24 commented Sep 9, 2023

LiWangSH commented Sep 13, 2023 •

edited

Loading

LiWangSH commented Sep 14, 2023

palminha commented Sep 19, 2023

palminha commented Sep 19, 2023 •

edited

Loading

palminha commented Sep 19, 2023 •

edited

Loading

qlzh727 commented Sep 21, 2023

tilakrayal commented Sep 28, 2023 •

edited

Loading

palminha commented Sep 28, 2023

eswar-janjanam commented Sep 29, 2023 •

edited

Loading

nkovela1 commented Sep 29, 2023

phpmac commented Oct 2, 2023

phpmac commented Oct 2, 2023

hanamurayuki commented Oct 16, 2023

phpmac commented Oct 16, 2023

MurtadhaInit commented Oct 27, 2023

richardokonicha commented Mar 3, 2024

saarangbond commented Jun 17, 2024

eyash24 commented Aug 18, 2024 •

edited

Loading

keras create/load model fails on Apple M2 cpu. Same code works on Ryzen #46

keras create/load model fails on Apple M2 cpu. Same code works on Ryzen #46

Comments

drmatthewclark commented Jul 6, 2023

tilakrayal commented Jul 7, 2023

jtpotato commented Jul 18, 2023

bossebandowski commented Jul 19, 2023

tilakrayal commented Jul 20, 2023

bossebandowski commented Jul 26, 2023 • edited Loading

tilakrayal commented Aug 3, 2023

github-actions bot commented Aug 18, 2023

mbergwall2222 commented Aug 18, 2023

tcivie commented Aug 20, 2023

norwend commented Aug 27, 2023 • edited Loading

RobP-git commented Sep 4, 2023

samhva commented Sep 8, 2023

ptn24 commented Sep 9, 2023

LiWangSH commented Sep 13, 2023 • edited Loading

LiWangSH commented Sep 14, 2023

palminha commented Sep 19, 2023

palminha commented Sep 19, 2023 • edited Loading

palminha commented Sep 19, 2023 • edited Loading

qlzh727 commented Sep 21, 2023

tilakrayal commented Sep 28, 2023 • edited Loading

palminha commented Sep 28, 2023

eswar-janjanam commented Sep 29, 2023 • edited Loading

nkovela1 commented Sep 29, 2023

phpmac commented Oct 2, 2023

phpmac commented Oct 2, 2023

hanamurayuki commented Oct 16, 2023

phpmac commented Oct 16, 2023

MurtadhaInit commented Oct 27, 2023

richardokonicha commented Mar 3, 2024

saarangbond commented Jun 17, 2024

eyash24 commented Aug 18, 2024 • edited Loading

bossebandowski commented Jul 26, 2023 •

edited

Loading

norwend commented Aug 27, 2023 •

edited

Loading

LiWangSH commented Sep 13, 2023 •

edited

Loading

palminha commented Sep 19, 2023 •

edited

Loading

palminha commented Sep 19, 2023 •

edited

Loading

tilakrayal commented Sep 28, 2023 •

edited

Loading

eswar-janjanam commented Sep 29, 2023 •

edited

Loading

eyash24 commented Aug 18, 2024 •

edited

Loading