About Autoencoder TensorFlow --Example.dxp

[nttcom online]

 I downloaded autoencoder_tensorflow_v1.0.0.zip from the following URL and used Autoencoder TensorFlow --Example.dxp.

​

URL : https://community.spotfire.com/s/exchange/aCv4z0000008OckCAE/autoencoder-tensorflow-python-data-function-for-tibco-spotfire

​

On the "Build and Evaluate Model" page, it says that even if the File Path is empty, it will be executed in the default state.

​

However, when I set File Path to empty and clicked "Create Autoencoder Model", the following error message appeared.

​

How can I resolve this error？

If anyone knows please let me know.

​

------------------------Error Message------------------------

Could not execute function call '[Modeling] Autoencoder TensorFlow'

​

2022-07-29 11:38:50.877476: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'cudart64_110.dll'; dlerror: cudart64_110.dll not found

2022-07-29 11:38:50.878140: 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.

2022-07-29 11:39:02.595189: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'nvcuda.dll'; dlerror: nvcuda.dll not found

2022-07-29 11:39:02.595235: W tensorflow/stream_executor/cuda/cuda_driver.cc:269] failed call to cuInit: UNKNOWN ERROR (303)

2022-07-29 11:39:02.609277: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:169] retrieving CUDA diagnostic information for host: RNW21151

2022-07-29 11:39:02.609595: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:176] hostname: RNW21151

2022-07-29 11:39:02.609949: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX AVX2

To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.

​

Error executing Python script:

​

NameError: name 'fitted_model' is not defined

​

Traceback (most recent call last):

File "data_function.py", line 341, in _execute_script

exec(compiled_script, self.globals)

File "<data_function>", line 335, in <module>

​

Standard output:

TensorFlow Version: 2.9.1

Categorical Variables: ['ProductLine']

Numerical Variables: ['ProdPerMinute', 'Pressure', 'BaroPressure', 'WasteGas', 'LossOfEfficiency', 'SR', 'PowerIn', '1dp', '2dp', '3dp', '4dp', '5dp', '6dp']

Model file path: best_model.h5

Model: "sequential"

_________________________________________________________________

Layer (type) Output Shape Param #

=================================================================

dense (Dense) (None, 64) 1088

_________________________________________________________________

dense_1 (Dense) (None, 32) 2080

_________________________________________________________________

dense_2 (Dense) (None, 5) 165

_________________________________________________________________

dense_3 (Dense) (None, 32) 192

_________________________________________________________________

dense_4 (Dense) (None, 64) 2112

_________________________________________________________________

dense_5 (Dense) (None, 16) 1040

=================================================================

Total params: 6,677

Trainable params: 6,677

Non-trainable params: 0

_________________________________________________________________

None

Error in compiling Keras model: Could not interpret optimizer identifier: <keras.optimizers.optimizer_v2.adam.Adam object at 0x00000275396BBE80>

​

​

​

at Spotfire.Dxp.Data.DataFunctions.Executors.LocalPythonFunctionClient.<RunFunction>d__8.MoveNext()

at Spotfire.Dxp.Data.DataFunctions.Executors.PythonScriptExecutor.<ExecuteFunction>d__11.MoveNext()

at Spotfire.Dxp.Data.DataFunctions.DataFunctionExecutorService.<ExecuteFunction>d__8.MoveNext() 

[Sweta Kotha]

Hi, I fixed a few lines of code in the DXP. Please try this version; we will update the Exchange component soon.

(fixes: using packages tf.keras instead of tf.python.keras, inserting optional model optimizer)

[nttcom online]

The DXP file you gave me ran without any errors.

Thank you so much.

[nttcom online]

Sorry for asking so many questions.

If you know how to replace the training data and test data with data prepared at hand, please let me know.

Thank you.

[Sweta Kotha]

The instructions on the Exchange component are still relevant for how to configure the data function with new data: https://community.spotfire.com/s/exchange/aCv4z0000008OckCAE/autoencoder-tensorflow-python-data-function-for-tibco-spotfire

please reference the links for setting up data functions and the readme for the data function

[MASUHIRO ISHITOBI]

My Spotfire version is 10.10.3. So i couldn't load fixed version1.1.​

It would be appreciated if you could give me DXP for this version or fixed python script as a text file.　　ーー＞　its done. Solved by myself.