Intel® DevCloud
Help for those needing help starting or connecting to the Intel® DevCloud
1628 Discussions

kernal restart issue

Supriyo-ux
Beginner
1,905 Views

whenever i try to run any code it throws error.

0 Kudos
12 Replies
JaideepK_Intel
Moderator
1,889 Views

Hi,

 

Thank you for posting in Intel Communities.

Could you let us know whether you are using Intel DevCloud for oneAPI or Intel DevCloud for the Edge or DevCloud for FPGA ?

 

Thanks,

Jaideep


0 Kudos
Supriyo-ux
Beginner
1,883 Views

hi thanks, i am using intel devcloud one api kernal for python, let me know if we have a screen sharing session via google meet

0 Kudos
JaideepK_Intel
Moderator
1,872 Views

Hi,


Can you please answer the following questions:

1. How often do you face this issue? Do you face this issue immediately after you launch jupyter notebook?

2. Can you run a command line interface? (via remote ssh)


Can you please try the below workarounds:

1. Delete your browser's cache and cookies and try to run in incognito mode.

2. If your application is running for longer than 6 hours, increase the "wall time." (-l walltime=hh:mm:ss)


We will connect with you through private message to get your DevCloud ID & error screenshots.


Thanks,

Jaideep


0 Kudos
Supriyo-ux
Beginner
1,865 Views

i am getting on immediate effect .

here is my id-  (uxxxxx)

Supriyoux_0-1652362039236.png

 

one new error message also coming like that-

 

Supriyoux_1-1652362086829.png

it is unable to read also

 

0 Kudos
JaideepK_Intel
Moderator
1,854 Views

Hi,


Thanks for sharing the information. We are going to inform our admin team and get back to you with an update.


Thanks,

Jaideep


0 Kudos
Supriyo-ux
Beginner
1,801 Views

any update?

0 Kudos
JaideepK_Intel
Moderator
1,652 Views

Hi,


The issue is related to the Ray component is allocating a lot of memory which is killing the Jupyter kernel.

Workaround solution is to start Ray with memory arguments to stay within limits like :

ray.init(

  _memory=16000 * 1024 * 1024,

  object_store_memory=500 * 1024 * 1024,

  _driver_object_store_memory=500 * 1024 * 1024)


Thanks,

Jaideep


0 Kudos
Supriyo-ux
Beginner
1,517 Views

Hi i have added this:

import ray
ray.init(_memory=16000 * 1024 * 1024,object_store_memory=500 * 1024 * 1024,_driver_object_store_memory=500 * 1024 * 1024)

 

 

but still not working :

AttributeError                            Traceback (most recent call last)
Input In [5], in <cell line: 1>()
----> 1 import ray
      2 ray.init(_memory=16000 * 1024 * 1024,object_store_memory=500 * 1024 * 1024,_driver_object_store_memory=500 * 1024 * 1024)

File ~/.local/lib/python3.9/site-packages/ray/__init__.py:160, in <module>
    158 from ray.runtime_context import get_runtime_context  # noqa: E402
    159 from ray import autoscaler  # noqa:E402
--> 160 from ray import data  # noqa: E402,F401
    161 from ray import util  # noqa: E402
    162 from ray import _private  # noqa: E402,F401

File ~/.local/lib/python3.9/site-packages/ray/data/__init__.py:1, in <module>
----> 1 from ray.data.read_api import (
      2     from_items,
      3     range,
      4     range_arrow,
      5     range_tensor,
      6     read_parquet,
      7     read_json,
      8     read_csv,
      9     read_binary_files,
     10     from_dask,
     11     from_modin,
     12     from_mars,
     13     from_pandas,
     14     from_pandas_refs,
     15     from_numpy,
     16     from_arrow,
     17     from_arrow_refs,
     18     from_spark,
     19     read_datasource,
     20     read_numpy,
     21     read_text,
     22 )
     23 from ray.data.datasource import Datasource, ReadTask
     24 from ray.data.dataset import Dataset

File ~/.local/lib/python3.9/site-packages/ray/data/read_api.py:37, in <module>
     28 from ray.util.annotations import PublicAPI, DeveloperAPI
     29 from ray.data.block import (
     30     Block,
     31     BlockAccessor,
   (...)
     35     BlockPartitionMetadata,
     36 )
---> 37 from ray.data.context import DatasetContext
     38 from ray.data.dataset import Dataset
     39 from ray.data.datasource import (
     40     Datasource,
     41     RangeDatasource,
   (...)
     51     DefaultParquetMetadataProvider,
     52 )

File ~/.local/lib/python3.9/site-packages/ray/data/context.py:33, in <module>
     28 # Whether to furthermore fuse prior map tasks with shuffle stages.
     29 DEFAULT_OPTIMIZE_FUSE_SHUFFLE_STAGES = True
     32 @DeveloperAPI
---> 33 class DatasetContext:
     34     """Singleton for shared Dataset resources and configurations.
     35 
     36     This object is automatically propagated to workers and can be retrieved
     37     from the driver and remote workers via DatasetContext.get_current().
     38     """
     40     def __init__(
     41         self,
     42         block_owner: ray.actor.ActorHandle,
   (...)
     48         optimize_fuse_shuffle_stages: bool,
     49     ):

File ~/.local/lib/python3.9/site-packages/ray/data/context.py:42, in DatasetContext()
     32 @DeveloperAPI
     33 class DatasetContext:
     34     """Singleton for shared Dataset resources and configurations.
     35 
     36     This object is automatically propagated to workers and can be retrieved
     37     from the driver and remote workers via DatasetContext.get_current().
     38     """
     40     def __init__(
     41         self,
---> 42         block_owner: ray.actor.ActorHandle,
     43         block_splitting_enabled: bool,
     44         target_max_block_size: int,
     45         enable_pandas_block: bool,
     46         optimize_fuse_stages: bool,
     47         optimize_fuse_read_stages: bool,
     48         optimize_fuse_shuffle_stages: bool,
     49     ):
     50         """Private constructor (use get_current() instead)."""
     51         self.block_owner = block_owner

AttributeError: partially initialized module 'ray' has no attribute 'actor' (most likely due to a circular import)
0 Kudos
JaideepK_Intel
Moderator
1,508 Views

Hi,

 

I assume there is some issue while downloading the ray package.

Please follow these steps:

If you are using a Jupyter notebook use the below command:

!pip install -U ray

or

If you are using terminal use the below command:

pip install -U ray

JaideepK_Intel_0-1653050153202.png

 

Below command to verify ray is installed properly or not:

!pip3 list | grep -i ray (Jupyter notebook)
pip3 list | grep -i ray (terminal)

Now import ray with the below command:

import ray
ray.init(_memory=16000 * 1024 * 1024,object_store_memory=500 * 1024 * 1024,_driver_object_store_memory=500 * 1024 * 1024)

JaideepK_Intel_2-1653050432388.png

 

Thanks,

Jaideep

 

0 Kudos
Supriyo-ux
Beginner
1,114 Views

new issue: not able to read csv file which can read in other local jupyter tool

Supriyoux_0-1653374847966.png

 

 

0 Kudos
JaideepK_Intel
Moderator
1,102 Views

Hi,

 

Since you gave the wrong path, I assume you're getting this error.

correct path: /home/<user_id(example: u123456)>/nlp

 

So the command should look like this:

 

df=pd.read_csv("/home/<user_id(example: u123456)>/nlp/valid.csv")

Since you are in the same directory where valid.csv is present, you can also use the below command.

df=pd.read_csv("./valid.csv")

JaideepK_Intel_0-1653383943338.png

 

 

If this resolves your issue, make sure to accept this as a solution. This would help others with similar issue. Thank you!

 

Thanks,

Jaideep

 

0 Kudos
JaideepK_Intel
Moderator
933 Views

Hi,


We assume that your issue is resolved. If you need any additional information, please post a new question as this thread will no longer be monitored by Intel.


Thanks,

Jaideep


0 Kudos
Reply