I am running keros_ocr with tensorflow-macos 2.11.0 and tensorflow-metal 0.7.1 on Monterey 12.2 on an MacBook Pro 2021 with Apple M1 Pro. The environment I built according to this website https://developer.apple.com/metal/tensorflow-plugin/ Then I got this issue
Looking for /Users/macbookpro/.keras-ocr/craft_mlt_25k.h5
Metal device set to: Apple M1 Pro
systemMemory: 16.00 GB
maxCacheSize: 5.33 GB
2023-02-13 09:42:08.663860: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:306] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.
2023-02-13 09:42:08.664267: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:272] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)
Looking for /Users/macbookpro/.keras-ocr/crnn_kurapan.h5
2023-02-13 09:42:11.437347: W tensorflow/tsl/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz
2023-02-13 09:42:11.595474: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.
1/1 [==============================] - 2s 2s/step
2023-02-13 09:42:15.158322: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.
2023-02-13 09:42:15.438 python[2073:25218] -[MPSGraph coordinateAlongAxisTensor:withShapeTensor:name:]: unrecognized selector sent to instance 0x2bf029030
2023-02-13 09:42:15.487 python[2073:25218] *** Terminating app due to uncaught exception 'NSInvalidArgumentException', reason: '-[MPSGraph coordinateAlongAxisTensor:withShapeTensor:name:]: unrecognized selector sent to instance 0x2bf029030'
*** First throw call stack:
(
0 CoreFoundation 0x00000001ac40c1cc __exceptionPreprocess + 240
1 libobjc.A.dylib 0x00000001ac15d7b8 objc_exception_throw + 60
2 CoreFoundation 0x00000001ac49f1a0 -[NSObject(NSObject) __retain_OA] + 0
3 CoreFoundation 0x00000001ac36c360 ___forwarding___ + 1728
4 CoreFoundation 0x00000001ac36bbe0 _CF_forwarding_prep_0 + 96
5 libmetal_plugin.dylib 0x000000017795e25c ___ZN12metal_plugin17MPSSequenceOpBaseIiE15ProduceSequenceEPNS_15OpKernelContextEPNS_6TensorEiiix_block_invoke + 200
6 libmetal_plugin.dylib 0x000000017795c81c ___ZN12metal_plugin13MPSGraphCache17CreateCachedGraphERKNSt3__112basic_stringIcNS1_11char_traitsIcEENS1_9allocatorIcEEEEU13block_pointerFPNS_14MPSCachedGraphEvE_block_invoke + 492
7 libdispatch.dylib 0x00000001ac102bac _dispatch_client_callout + 20
8 libdispatch.dylib 0x00000001ac111e00 _dispatch_lane_barrier_sync_invoke_and_complete + 56
9 libmetal_plugin.dylib 0x000000017795decc _ZN12metal_plugin17MPSSequenceOpBaseIiE15ProduceSequenceEPNS_15OpKernelContextEPNS_6TensorEiiix + 828
10 libmetal_plugin.dylib 0x000000017795d468 _ZN12metal_plugin10MPSRangeOpIiE7ComputeEPNS_15OpKernelContextE + 2076
11 libmetal_plugin.dylib 0x000000017795cb78 _ZN12metal_pluginL15ComputeOpKernelINS_10MPSRangeOpIiEEEEvPvP18TF_OpKernelContext + 44
12 libtensorflow_framework.2.dylib 0x0000000156240028 _ZN10tensorflow15PluggableDevice7ComputeEPNS_8OpKernelEPNS_15OpKernelContextE + 148
13 libtensorflow_framework.2.dylib 0x00000001561bd510 _ZN10tensorflow12_GLOBAL__N_113ExecutorStateINS_15PropagatorStateEE7ProcessENS2_10TaggedNodeEx + 3784
14 libtensorflow_framework.2.dylib 0x00000001561bf170 _ZNSt3__110__function6__funcIZN10tensorflow12_GLOBAL__N_113ExecutorStateINS2_15PropagatorStateEE7RunTaskIZNS6_13ScheduleReadyEPN4absl12lts_2022062313InlinedVectorINS5_10TaggedNodeELm8ENS_9allocatorISB_EEEEPNS5_20TaggedNodeReadyQueueEEUlvE1_EEvOT_iEUlvE_NSC_ISL_EEFvvEEclEv + 56
15 _pywrap_tensorflow_internal.so 0x00000002a7faab78 _ZN5Eigen15ThreadPoolTemplIN3tsl6thread16EigenEnvironmentEE10WorkerLoopEi + 1496
16 _pywrap_tensorflow_internal.so 0x00000002a7faa474 _ZZN3tsl6thread16EigenEnvironment12CreateThreadENSt3__18functionIFvvEEEENKUlvE_clEv + 80
17 libtensorflow_framework.2.dylib 0x00000001560de878 _ZN3tsl12_GLOBAL__N_17PThread8ThreadFnEPv + 120
18 libsystem_pthread.dylib 0x00000001ac2c1240 _pthread_start + 148
19 libsystem_pthread.dylib 0x00000001ac2bc024 thread_start + 8
)
libc++abi: terminating with uncaught exception of type NSException
Her is my test code
import matplotlib.pyplot as plt
import keras_ocr
# keras-ocr will automatically download pretrained
# weights for the detector and recognizer.
pipeline = keras_ocr.pipeline.Pipeline()
# Get a set of three example images
images = [
keras_ocr.tools.read(url) for url in [
'https://pic4.zhimg.com/v2-97af8540d2b25dd0f01f45af74b65f2b_r.jpg',
'https://pic4.zhimg.com/v2-283a715d0c4628fb1e2020a2228e252b_r.jpg'
]
]
# Each list of predictions in prediction_groups is a list of
# (word, box) tuples.
prediction_groups = pipeline.recognize(images)
# Plot the predictions
fig, axs = plt.subplots(nrows=len(images), figsize=(20, 20))
for ax, image, predictions in zip(axs, images, prediction_groups):
keras_ocr.tools.drawAnnotations(image=image, predictions=predictions, ax=ax)
And libs version
keras 2.11.0 pypi_0 pypi
keras-applications 1.0.8 pypi_0 pypi
keras-ocr 0.0.0 pypi_0 pypi
numpy 1.24.2 pypi_0 pypi
opencv-python 4.7.0.68 pypi_0 pypi
openssl 1.1.1s h1a28f6b_0
pip 23.0 pypi_0 pypi
python 3.9.16 hc0d8a6c_0
python-dateutil 2.8.2 pypi_0 pypi
tensorboard 2.11.2 pypi_0 pypi
tensorboard-data-server 0.6.1 pypi_0 pypi
tensorboard-plugin-wit 1.8.1 pypi_0 pypi
tensorflow-deps 2.9.0 0 apple
tensorflow-estimator 2.11.0 pypi_0 pypi
tensorflow-macos 2.11.0 pypi_0 pypi
tensorflow-metal 0.7.1 pypi_0 pypi
I can running this code successfully
import tensorflow as tf
from tensorflow.keras.optimizers.legacy import Adam
cifar = tf.keras.datasets.cifar100
(x_train, y_train), (x_test, y_test) = cifar.load_data()
model = tf.keras.applications.ResNet50(
include_top=True,
weights=None,
input_shape=(32, 32, 3),
classes=100,)
loss_fn = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True)
# Instantiate legacy Adam optimizer
model.compile(optimizer=Adam(), loss=loss_fn, metrics=["accuracy"])
model.fit(x_train, y_train, epochs=5, batch_size=64)
