search

sassoftware / python-dlpy

The SAS Deep Learning Python (DLPy) package provides the high-level Python APIs to deep learning methods in SAS Visual Data Mining and Machine Learning. It allows users to build deep learning models using friendly Keras-like APIs.
Apache License 2.0
224 stars 131 forks source link

Error while displaying images #400

Closed myuser-git closed 10 months ago

myuser-git commented 1 year ago

The following code runs on an instance of JupyterHub with Python 3.10 and dlpy-dev (1.2.1-dev):

Do not modify this file!

import swat conn = swat.CAS('https://iot.viya-azure-3.unx.sas.com:443/cas-shared-default-http', username='macazz',password='xxxxxx') out = conn.serverstatus() out import dlpy from dlpy.utils import from dlpy.images import ImageTable from dlpy.model import from dlpy import Model, Sequential from dlpy.layers import from dlpy.applications import from dlpy.splitting import two_way_split from dlpy.blocks import * import PIL.Image print(dlpy.version)

s.tableinfo() s.loadactionset('image') s.loadactionset('deepLearn') s.loadactionset('astore') modelPath='/home/jukhar/Common_Data/Juthika/Giraffe_Dolphin/' s.table.addcaslib(activeonadd=False,datasource={'srctype':'path'},name='dnfs',path=modelPath,subdirectories=True) s.setsessopt(caslib='dnfs') my_images = s.image.loadimages(casout=dict(name='vt_data', replace=True), caslib='dnfs', recurse=True, labellevels=2) trainImgTbl = s.CASTable('vt_data') Img = ImageTable.from_table(trainImgTbl) Img.show()

Output:

NOTE: Grid node action status report: 1 nodes, 9 total actions executed. 1.2.1-dev NOTE: No tables are available in caslib CASUSER(macazz) of Cloud Analytic Services. NOTE: Added action set 'image'. NOTE: Added action set 'deepLearn'. NOTE: Added action set 'astore'. NOTE: Cloud Analytic Services added the caslib 'dnfs'. NOTE: 'dnfs' is now the active caslib. NOTE: Loaded 257 images from /home/jukhar/Common_Data/Juthika/Giraffe_Dolphin/ into Cloud Analytic Services table vt_data. **--------------------------------------------------------------------------- AttributeError Traceback (most recent call last) Cell In[6], line 11 9 trainImgTbl = s.CASTable('vt_data') 10 Img = ImageTable.from_table(trainImgTbl) ---> 11 Img.show()

File /opt/conda/lib/python3.10/site-packages/dlpy/images.py:514, in ImageTable.show(self, nimages, ncol, randomize, figsize, where, id) 512 else: 513 ax.set_title('{}'.format(label)) --> 514 if len(image.size) == 2: 515 plt.imshow(np.array(image), cmap='Greys_r') 516 else:

AttributeError: 'bytes' object has no attribute 'size'**

Any insight would be greatly appreciated.

dxq77dxq commented 1 year ago

Hello, would you please share the version of your DLPy and cas?

myuser-git commented 1 year ago

Viya version is 2023.08 Dlpy version is 1.2.1-dev

dxq77dxq commented 1 year ago

Can you try ImgTbl = ImageTable.load_files(s, path='path_to_your_images', casout=dict(name='vt_data')) ImgTbl.show()

dxq77dxq commented 10 months ago

This was resolved by changing the way of connecting to Viya. The user ended up using binary connectivity instead of REST.