我一直致力于训练 tfdf.keras.GradientBoostedTreesModel，而且我经常遇到问题，我加载的模型给我一个 ValueError，说它“找不到匹配的函数来调用从 SavedModel 加载的函数。” TFDF的“known issues”页面指出，这可能是由于训练数据和我要预测的数据的数据类型不同造成的。
我一直在从同一个文件加载我的训练数据和我的预测数据（我正在为我的分类问题制作一个混淆矩阵），我使用 pandas 加载到 python 中，然后转换为张量流数据集：
labelStrings = pd.read_hdf('labelStrings.hdf5','labels').values.tolist()[0]
rawData = pd.read_hdf("identifiedData.hdf5","identified",stop=1000000)
boostedDecisionTree = tf.keras.models.load_model('bgGenBDT')
rawData['label']=rawData['label'].map(labelStrings.index) #Converting the string labels to integer ones
rawData = rawData.astype('float32')
evalData = tfdf.keras.pd_dataframe_to_tf_dataset(rawData,label='label')
results = boostedDecisionTree.predict(evalData,verbose=2)
这种转换为 float32 是我尝试解决任何不匹配数据类型的问题，并且它已经工作了几个月。然而，即使我在训练脚本中使用几乎完全相同的代码（唯一改变的是我从第 2 行的文件中读取的数据），我还是收到了 ValueError。输入功能列表一团糟，但我确实注意到错误是声称我的输入具有以下形式
{feature:<tf.Tensor 'inputs_126:0' shape=(None, 1) dtype=float32>}{feature: TensorSpec(shape=(None, 1), dtype=tf.float32, name='inputs/feature')}编辑：这是一个简化的示例和我得到的输出。
这是我写的第一个脚本，按预期工作：
import tensorflow_decision_forests as tfdf
import pandas as pd
import tensorflow as tf
import random
#Generating data to train on
rawData = [[random.uniform(0,2),random.uniform(0,2),random.randrange(0,3)] for _ in range(1000)]
data = pd.DataFrame.from_records(rawData,columns=['a','b','label']).astype('float32')
trainingData = tfdf.keras.pd_dataframe_to_tf_dataset(data,label='label')
#Training and saving the model
boostedDecisionTree = tfdf.keras.GradientBoostedTreesModel(max_depth= 6, num_trees = 10)
boostedDecisionTree.fit(trainingData)
boostedDecisionTree.save('BDT')
#Generating test data
rawData = [[random.uniform(0,2),random.uniform(0,2),random.randrange(0,3)] for _ in range(1000)]
data = pd.DataFrame.from_records(rawData,columns=['a','b','label'])
predictData = tfdf.keras.pd_dataframe_to_tf_dataset(data,label='label')
#Evaluating model on test data
results = boostedDecisionTree.predict(predictData)
resultsPD = pd.DataFrame.from_records(results)
print(resultsPD)
及其输出：
2023-05-16 14:23:40.621493: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /cvmfs/oasis.opensciencegrid.org/gluex/root-6.22.06/x86_64/lib:/cvmfs/oasis.opensciencegrid.org/gluex/xrootd/5.0.3/lib64:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/evtgen/evtgen-01.07.00/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/photos/Photos-3.61/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/hepmc/HepMC-2.06.10/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/gluex_root_analysis/gluex_root_analysis-1.18.0^hdr4270/Linux_CentOS7-x86_64-gcc4.8.5-cntr/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/diracxx/Diracxx-1.0.1/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/diracxx/Diracxx-1.0.1:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/evio/evio-4.4.6/Linux-x86_64/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/rcdb/rcdb_0.06.00/cpp/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/ccdb/ccdb_1.06.07/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/geant4/geant4.10.02.p02/lib64:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/root/root-6.08.06/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/xerces-c/xerces-c-3.1.4/lib:/.singularity.d/libs:/cvmfs/oasis.opensciencegrid.org/gluex/python3:/cvmfs/oasis.opensciencegrid.org/gluex/hdf5-1.12.0/lib
2023-05-16 14:23:40.621581: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
WARNING:root:TF Parameter Server distributed training not available.
2023-05-16 14:23:44.102919: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcuda.so.1'; dlerror: libcuda.so.1: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /cvmfs/oasis.opensciencegrid.org/gluex/root-6.22.06/x86_64/lib:/cvmfs/oasis.opensciencegrid.org/gluex/xrootd/5.0.3/lib64:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/evtgen/evtgen-01.07.00/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/photos/Photos-3.61/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/hepmc/HepMC-2.06.10/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/gluex_root_analysis/gluex_root_analysis-1.18.0^hdr4270/Linux_CentOS7-x86_64-gcc4.8.5-cntr/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/diracxx/Diracxx-1.0.1/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/diracxx/Diracxx-1.0.1:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/evio/evio-4.4.6/Linux-x86_64/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/rcdb/rcdb_0.06.00/cpp/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/ccdb/ccdb_1.06.07/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/geant4/geant4.10.02.p02/lib64:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/root/root-6.08.06/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/xerces-c/xerces-c-3.1.4/lib:/.singularity.d/libs:/cvmfs/oasis.opensciencegrid.org/gluex/python3:/cvmfs/oasis.opensciencegrid.org/gluex/hdf5-1.12.0/lib
2023-05-16 14:23:44.102996: W tensorflow/stream_executor/cuda/cuda_driver.cc:269] failed call to cuInit: UNKNOWN ERROR (303)
2023-05-16 14:23:44.103044: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:156] kernel driver does not appear to be running on this host (cn446): /proc/driver/nvidia/version does not exist
2023-05-16 14:23:44.103717: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-05-16 14:23:51.864524: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:185] None of the MLIR Optimization Passes are enabled (registered 2)
16/16 [==============================] - 8s 821us/step
[INFO kernel.cc:736] Start Yggdrasil model training
[INFO kernel.cc:737] Collect training examples
[INFO kernel.cc:392] Number of batches: 16
[INFO kernel.cc:393] Number of examples: 1000
[INFO kernel.cc:759] Dataset:
Number of records: 1000
Number of columns: 3
Number of columns by type:
NUMERICAL: 2 (66.6667%)
CATEGORICAL: 1 (33.3333%)
Columns:
NUMERICAL: 2 (66.6667%)
0: "a" NUMERICAL mean:0.980919 min:1.23745e-05 max:1.99877 sd:0.571513
1: "b" NUMERICAL mean:1.02036 min:0.00265199 max:1.99999 sd:0.57927
CATEGORICAL: 1 (33.3333%)
2: "__LABEL" CATEGORICAL integerized vocab-size:4 no-ood-item
Terminology:
nas: Number of non-available (i.e. missing) values.
ood: Out of dictionary.
manually-defined: Attribute which type is manually defined by the user i.e. the type was not automatically inferred.
tokenized: The attribute value is obtained through tokenization.
has-dict: The attribute is attached to a string dictionary e.g. a categorical attribute stored as a string.
vocab-size: Number of unique values.
[INFO kernel.cc:762] Configure learner
[WARNING gradient_boosted_trees.cc:1643] Subsample hyperparameter given but sampling method does not match.
[WARNING gradient_boosted_trees.cc:1656] GOSS alpha hyperparameter given but GOSS is disabled.
[WARNING gradient_boosted_trees.cc:1665] GOSS beta hyperparameter given but GOSS is disabled.
[WARNING gradient_boosted_trees.cc:1677] SelGB ratio hyperparameter given but SelGB is disabled.
[INFO kernel.cc:787] Training config:
learner: "GRADIENT_BOOSTED_TREES"
features: "a"
features: "b"
label: "__LABEL"
task: CLASSIFICATION
[yggdrasil_decision_forests.model.gradient_boosted_trees.proto.gradient_boosted_trees_config] {
num_trees: 10
decision_tree {
max_depth: 6
min_examples: 5
in_split_min_examples_check: true
missing_value_policy: GLOBAL_IMPUTATION
allow_na_conditions: false
categorical_set_greedy_forward {
sampling: 0.1
max_num_items: -1
min_item_frequency: 1
}
growing_strategy_local {
}
categorical {
cart {
}
}
num_candidate_attributes_ratio: -1
axis_aligned_split {
}
internal {
sorting_strategy: PRESORTED
}
}
shrinkage: 0.1
validation_set_ratio: 0.1
early_stopping: VALIDATION_LOSS_INCREASE
early_stopping_num_trees_look_ahead: 30
l2_regularization: 0
lambda_loss: 1
mart {
}
adapt_subsample_for_maximum_training_duration: false
l1_regularization: 0
use_hessian_gain: false
l2_regularization_categorical: 1
apply_link_function: true
compute_permutation_variable_importance: false
}
[INFO kernel.cc:790] Deployment config:
num_threads: 6
[INFO kernel.cc:817] Train model
[INFO gradient_boosted_trees.cc:404] Default loss set to MULTINOMIAL_LOG_LIKELIHOOD
[INFO gradient_boosted_trees.cc:1001] Training gradient boosted tree on 1000 example(s) and 2 feature(s).
[INFO gradient_boosted_trees.cc:1044] 923 examples used for training and 77 examples used for validation
[INFO gradient_boosted_trees.cc:1426] num-trees:1 train-loss:1.088802 train-accuracy:0.431203 valid-loss:1.101725 valid-accuracy:0.285714
[INFO gradient_boosted_trees.cc:1428] num-trees:2 train-loss:1.076015 train-accuracy:0.489707 valid-loss:1.101731 valid-accuracy:0.233766
[INFO gradient_boosted_trees.cc:1426] num-trees:10 train-loss:1.013323 train-accuracy:0.553629 valid-loss:1.097824 valid-accuracy:0.350649
[INFO gradient_boosted_trees.cc:229] Truncates the model to 27 tree(s) i.e. 9 iteration(s).
[INFO gradient_boosted_trees.cc:263] Final model num-trees:9 valid-loss:1.093618 valid-accuracy:0.363636
[INFO kernel.cc:828] Export model in log directory: /tmp/tmp_tu93ado
[INFO kernel.cc:836] Save model in resources
[INFO kernel.cc:988] Loading model from path
[INFO decision_forest.cc:590] Model loaded with 27 root(s), 793 node(s), and 2 input feature(s).
[INFO abstract_model.cc:993] Engine "GradientBoostedTreesGeneric" built
[INFO kernel.cc:848] Use fast generic engine
2023-05-16 14:23:55.500919: W tensorflow/python/util/util.cc:348] Sets are not currently considered sequences, but this may change in the future, so consider avoiding using them.
0 1 2
0 0.359499 0.328202 0.312299
1 0.331626 0.335902 0.332472
2 0.372847 0.307992 0.319160
3 0.371833 0.312442 0.315726
4 0.332559 0.198808 0.468633
.. ... ... ...
995 0.333704 0.358646 0.307650
996 0.347348 0.333341 0.319311
997 0.404316 0.313154 0.282530
998 0.344656 0.335357 0.319986
999 0.369685 0.349962 0.280352
[1000 rows x 3 columns]
这是我用来加载模型的脚本：
import pandas as pd
import tensorflow_decision_forests as tfdf
import tensorflow as tf
import random
#Generating test data
rawData = [[random.uniform(0,2),random.uniform(0,2),random.randrange(0,3)] for _ in range(1000)]
rawData = pd.DataFrame.from_records(rawData,columns=['a','b','label']).astype('float32')
rawData=rawData.drop('label',axis=1)
evalData = tfdf.keras.pd_dataframe_to_tf_dataset(rawData)
#Loading model and evaluating data
boostedDecisionTree= tf.keras.models.load_model('BDT')
results = boostedDecisionTree(evalData)
resultsPD = pd.DataFrame.from_records(results)
print(resultsPD)
我得到的错误：
2023-05-16 14:25:39.903926: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /cvmfs/oasis.opensciencegrid.org/gluex/root-6.22.06/x86_64/lib:/cvmfs/oasis.opensciencegrid.org/gluex/xrootd/5.0.3/lib64:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/evtgen/evtgen-01.07.00/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/photos/Photos-3.61/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/hepmc/HepMC-2.06.10/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/gluex_root_analysis/gluex_root_analysis-1.18.0^hdr4270/Linux_CentOS7-x86_64-gcc4.8.5-cntr/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/diracxx/Diracxx-1.0.1/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/diracxx/Diracxx-1.0.1:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/evio/evio-4.4.6/Linux-x86_64/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/rcdb/rcdb_0.06.00/cpp/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/ccdb/ccdb_1.06.07/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/geant4/geant4.10.02.p02/lib64:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/root/root-6.08.06/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/xerces-c/xerces-c-3.1.4/lib:/.singularity.d/libs:/cvmfs/oasis.opensciencegrid.org/gluex/python3:/cvmfs/oasis.opensciencegrid.org/gluex/hdf5-1.12.0/lib
2023-05-16 14:25:39.904012: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
WARNING:root:TF Parameter Server distributed training not available.
2023-05-16 14:25:43.117385: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcuda.so.1'; dlerror: libcuda.so.1: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /cvmfs/oasis.opensciencegrid.org/gluex/root-6.22.06/x86_64/lib:/cvmfs/oasis.opensciencegrid.org/gluex/xrootd/5.0.3/lib64:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/evtgen/evtgen-01.07.00/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/photos/Photos-3.61/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/hepmc/HepMC-2.06.10/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/gluex_root_analysis/gluex_root_analysis-1.18.0^hdr4270/Linux_CentOS7-x86_64-gcc4.8.5-cntr/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/diracxx/Diracxx-1.0.1/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/diracxx/Diracxx-1.0.1:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/evio/evio-4.4.6/Linux-x86_64/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/rcdb/rcdb_0.06.00/cpp/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/ccdb/ccdb_1.06.07/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/geant4/geant4.10.02.p02/lib64:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/root/root-6.08.06/lib:/group/halld/Software/builds/Linux_CentOS7-x86_64-gcc4.8.5-cntr/xerces-c/xerces-c-3.1.4/lib:/.singularity.d/libs:/cvmfs/oasis.opensciencegrid.org/gluex/python3:/cvmfs/oasis.opensciencegrid.org/gluex/hdf5-1.12.0/lib
2023-05-16 14:25:43.117498: W tensorflow/stream_executor/cuda/cuda_driver.cc:269] failed call to cuInit: UNKNOWN ERROR (303)
2023-05-16 14:25:43.117549: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:156] kernel driver does not appear to be running on this host (cn446): /proc/driver/nvidia/version does not exist
2023-05-16 14:25:43.118303: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-05-16 14:25:43.443369: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:185] None of the MLIR Optimization Passes are enabled (registered 2)
[INFO kernel.cc:988] Loading model from path
[INFO decision_forest.cc:590] Model loaded with 27 root(s), 793 node(s), and 2 input feature(s).
[INFO abstract_model.cc:993] Engine "GradientBoostedTreesGeneric" built
[INFO kernel.cc:848] Use fast generic engine
Traceback (most recent call last):
File "mre2.py", line 10, in <module>
results = boostedDecisionTree(evalData)
File "/usr/local/lib/python3.6/site-packages/keras/engine/base_layer.py", line 1037, in __call__
outputs = call_fn(inputs, *args, **kwargs)
File "/usr/local/lib/python3.6/site-packages/keras/saving/saved_model/utils.py", line 68, in return_outputs_and_add_losses
outputs, losses = fn(*args, **kwargs)
File "/usr/local/lib/python3.6/site-packages/keras/saving/saved_model/utils.py", line 166, in wrap_with_training_arg
lambda: replace_training_and_call(False))
File "/usr/local/lib/python3.6/site-packages/keras/utils/control_flow_util.py", line 106, in smart_cond
pred, true_fn=true_fn, false_fn=false_fn, name=name)
File "/usr/local/lib64/python3.6/site-packages/tensorflow/python/framework/smart_cond.py", line 58, in smart_cond
return false_fn()
File "/usr/local/lib/python3.6/site-packages/keras/saving/saved_model/utils.py", line 166, in <lambda>
lambda: replace_training_and_call(False))
File "/usr/local/lib/python3.6/site-packages/keras/saving/saved_model/utils.py", line 162, in replace_training_and_call
return wrapped_call(*args, **kwargs)
File "/usr/local/lib64/python3.6/site-packages/tensorflow/python/eager/def_function.py", line 885, in __call__
result = self._call(*args, **kwds)
File "/usr/local/lib64/python3.6/site-packages/tensorflow/python/eager/def_function.py", line 933, in _call
self._initialize(args, kwds, add_initializers_to=initializers)
File "/usr/local/lib64/python3.6/site-packages/tensorflow/python/eager/def_function.py", line 760, in _initialize
*args, **kwds))
File "/usr/local/lib64/python3.6/site-packages/tensorflow/python/eager/function.py", line 3066, in _get_concrete_function_internal_garbage_collected
graph_function, _ = self._maybe_define_function(args, kwargs)
File "/usr/local/lib64/python3.6/site-packages/tensorflow/python/eager/function.py", line 3463, in _maybe_define_function
graph_function = self._create_graph_function(args, kwargs)
File "/usr/local/lib64/python3.6/site-packages/tensorflow/python/eager/function.py", line 3308, in _create_graph_function
capture_by_value=self._capture_by_value),
File "/usr/local/lib64/python3.6/site-packages/tensorflow/python/framework/func_graph.py", line 1007, in func_graph_from_py_func
func_outputs = python_func(*func_args, **func_kwargs)
File "/usr/local/lib64/python3.6/site-packages/tensorflow/python/eager/def_function.py", line 668, in wrapped_fn
out = weak_wrapped_fn().__wrapped__(*args, **kwds)
File "/usr/local/lib64/python3.6/site-packages/tensorflow/python/saved_model/function_deserialization.py", line 294, in restored_function_body
"\n\n".join(signature_descriptions)))
ValueError: Could not find matching function to call loaded from the SavedModel. Got:
Positional arguments (2 total):
* <_VariantDataset shapes: {a: (None,), b: (None,)}, types: {a: tf.float32, b: tf.float32}>
* False
Keyword arguments: {}
Expected these arguments to match one of the following 4 option(s):
Option 1:
Positional arguments (2 total):
* {'b': TensorSpec(shape=(None, 1), dtype=tf.float32, name='inputs/b'), 'a': TensorSpec(shape=(None, 1), dtype=tf.float32, name='inputs/a')}
* False
Keyword arguments: {}
Option 2:
Positional arguments (2 total):
* {'a': TensorSpec(shape=(None, 1), dtype=tf.float32, name='inputs/a'), 'b': TensorSpec(shape=(None, 1), dtype=tf.float32, name='inputs/b')}
* True
Keyword arguments: {}
Option 3:
Positional arguments (2 total):
* {'a': TensorSpec(shape=(None, 1), dtype=tf.float32, name='a'), 'b': TensorSpec(shape=(None, 1), dtype=tf.float32, name='b')}
* False
Keyword arguments: {}
Option 4:
Positional arguments (2 total):
* {'a': TensorSpec(shape=(None, 1), dtype=tf.float32, name='a'), 'b': TensorSpec(shape=(None, 1), dtype=tf.float32, name='b')}
* True
Keyword arguments: {}