zmjjmz

import sys
from awsglue.transforms import *
from awsglue.utils import getResolvedOptions
from pyspark.context import SparkContext
from awsglue.context import GlueContext
from awsglue.job import Job

from pyspark.sql import functions as sf
from pyspark.sql import types as st
from awsglue.dynamicframe import DynamicFrame

zmjjmz

import os
import shutil
import tensorflow as tf
import tensorflow_datasets as tfds
import tabnet
import pandas as pd
import numpy as np

train_size = 125
BATCH_SIZE = 50

zmjjmz

W0401 13:47:33.782922 140134609663744 basic_session_run_hooks.py:732] It seems that global step (tf.train.get_global_step) 
has not been increased. Current value (could be stable): 0 vs previous value: 0. You could increase the global step by pass
ing tf.train.get_global_step() to Optimizer.apply_gradients or Optimizer.minimize.                                         
*** Error in `python': malloc(): memory corruption (fast): 0x00007f6f78326510 ***                                          
======= Backtrace: =========                                                                                               
/lib/x86_64-linux-gnu/libc.so.6(+0x70bfb)[0x7f73a0777bfb]                                                                  
/lib/x86_64-linux-gnu/libc.so.6(+0x76fc6)[0x7f73a077dfc6]                                                                  
/lib/x86_64-linux-gnu/libc.so.6(+0x79491)[0x7f73a0780491]                                                                  
/lib/x86_64-linux-gnu/libc.so.6(

zmjjmz

### Diagnostics

<details>
<summary>Diagnostics output</summary>

``````
--- check: autoidentify
INFO: diagnose_tensorboard.py version d515ab103e2b1cfcea2b096187741a0eeb8822ef

--- check: general

zmjjmz

import tensorflow as tf
from tensorflow import feature_column as fc
import pandas as pd
import numpy as np


FC_DISPATCH = {
    'numeric': lambda key, training_vals, **kwargs: fc.numeric_column(key, **kwargs),
    'categorical': lambda key, training_vals, **kwargs: fc.indicator_column(
        fc.categorical_column_with_vocabulary_list(

zmjjmz

   jax_jit_correlation  jax_jit_time  jax_nojit_correlation  jax_nojit_time  np_copy_correlation  np_copy_time  np_inplace_correlation  np_inplace_time      size
0            -0.087945      0.520224              -0.207465        0.515517            -0.086613      0.000577               -0.048701         0.000638       100
1             0.009647      0.409452              -0.009721        0.407500            -0.016213      0.000776               -0.032633         0.000694      1000
2             0.005357      0.652246               0.010250        0.633742             0.002910      0.003117               -0.001414         0.001872     10000
3            -0.002672      1.441106               0.000594        1.425177             0.004118      0.023031               -0.004778         0.017846    100000
4             0.001152     11.028948               0.000698       10.984871            -0.002028      0.228054                0.001034         0.163242   1000000
5             0.000007    250.970225              -0

zmjjmz

import numpy as onp
import jax.numpy as jnp
import jax.random as jrand
from jax import jit

@jit
def jax_shuffler(all_inputs, key):
    shuffled_input = jrand.shuffle(key, all_inputs, axis=0)
    return shuffled_input

zmjjmz

Traceback (most recent call last):
  File "jax_models.py", line 232, in <module>
    shuffle=True,
  File "jax_models.py", line 181, in fit
    voter_indices, target_indices, ratings, batch_size, batched_dataset_size)
  File "/home/u1/zach/proj/dataplayground3/lib/python3.5/site-packages/jax/api.py", line 150, in f_jitted
    out = xla.xla_call(flat_fun, *args_flat, device=device, backend=backend)
  File "/home/u1/zach/proj/dataplayground3/lib/python3.5/site-packages/jax/core.py", line 592, in call_bind
    outs = primitive.impl(f, *args, **params)
  File "/home/u1/zach/proj/dataplayground3/lib/python3.5/site-packages/jax/interpreters/xla.py", line 400, in _xla_call_impl

zmjjmz

{'error-code': 'INVALID_APPLICATION_PACKAGE',
 'message': 'Invalid application package: default.default: Error loading '
            'model: Could not import TensorFlow model from directory '
            "'/opt/vespa/var/db/vespa/config_server/serverdb/tenants/default/sessions/175/.preprocessed/models/plike_test/tf114_export': "
            "_output_shapes attribute of 'init_1' does not exist"}

zmjjmz

import keras                                                                           
import numpy as np                                                                     
                                                                                       
input_l = keras.Input(shape=(1,), name='input')                                        
layer_1 = keras.layers.Dense(1, activation='relu', name='layer_1')(input_l)            
output_l = keras.layers.Dense(1, activation='linear', name='output')(layer_1)          
model = keras.Model(inputs=[input_l], outputs=[output_l])                              
model.compile(loss='mean_absolute_error', optimizer='rmsprop')                         
                                                                                       
x = np.array(np.arange(1, 100000))