I'm using Pycharm with python 3.9, torch 1.8.1+cu111 and tensorboard 2.6.0.
When I try to import SummaryWriter by:
from torch.utils.tensorboard import SummaryWriter
I get the following error:
*Traceback (most recent call last):
File "/data/heisery/Yaronhome/PycharmProjects/CSSC_pytorch/tensorboard.py", line 18, in <module>
from torch.utils.tensorboard import SummaryWriter
File "/usr/local/lib/python3.8/dist-packages/torch/utils/tensorboard/__init__.py", line 1, in <module>
import tensorboard
File "/data/heisery/Yaronhome/PycharmProjects/CSSC_pytorch/tensorboard.py", line 18, in <module>
from torch.utils.tensorboard import SummaryWriter*
**ImportError: cannot import name 'SummaryWriter' from partially initialized module 'torch.utils.tensorboard' (most likely due to a circular import) (/usr/local/lib/python3.8/dist-packages/torch/utils/tensorboard/__init__.py)**
This is my import list:
import torch
import torch.nn as nn
from torch.utils.data import DataLoader
import torchvision
import torchvision.transforms as transforms
import matplotlib.pyplot as plt
import sys
from torch.utils.tensorboard import SummaryWriter
writer = SummaryWriter('runs/mnist')
device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
Bro
I have just solved a similar problem like yours. Just ensure that your filename is not set as 'tensorboard'.
Related
Trying to run some machine learning examples so am importing some packages
sklearn version 0.22.1
tensorflow version 2.3.0
keras version 2.3.0
in the following way:
import os
import sys
import math
import numpy as np
import pandas as pd
from sklearn.preprocessing import MinMaxScaler
from keras.models import Sequential
from keras.layers import Dense, LSTM
import matplotlip.pyplot as plt
I am getting the import error:
ImportError: cannot import name 'Sequence'
Full traceback:
File "C:\Anaconda\lib\site-packages\keras\__init__.py", line 12, in <module>
from . import callbacks
File "C:\Anaconda\lib\site-packages\keras\callbacks\__init__.py", line 3, in <module>
from .callbacks import Callback
File "C:\Anaconda\lib\site-packages\keras\callbacks\callbacks.py", line 23, in <module>
from ..engine.training_utils import standardize_input_data
File "C:\Anaconda\lib\site-packages\keras\engine\training_utils.py", line 18, in <module>
from ..utils import Sequence
ImportError: cannot import name 'Sequence'
I have looked on line for help trying a few different tensorflow and keras versions. I have also tried:
import tensorflow.keras as keras
and import tensorflow.keras.utils import Sequence
without success
when I am writing the code in PowerShell in vscode
from tensorflow.keras.preprocessing import image
from keras_vggface.utils import preprocess_input
from keras_vggface.vggface import VGGFace
I am getting this error. Help me with this
Traceback (most recent call last):
File "src/02_feature_extractor.py", line 7, in <module>
from keras_vggface.utils import preprocess_input
File "C:\Users\shana\Desktop\face_match\env\lib\site-packages\keras_vggface\__init__.py", line 1, in <module> from keras_vggface.vggface import VGGFace
File "C:\Users\shana\Desktop\face_match\env\lib\site-packages\keras_vggface\vggface.py", line 9, in <module>
from keras_vggface.models import RESNET50, VGG16, SENET50 File "C:\Users\shana\Desktop\face_match\env\lib\site-packages\keras_vggface\models.py", line 20, in <module>
from keras.engine.topology import get_source_inputs
ModuleNotFoundError: No module named 'keras.engine.topology'
I'm learning ObjectDetection from this website
I have installed ImageAI,Tensorflow and Keras.
Then when I run this in python
from imageai.Detection import ObjectDetection
I got
Traceback (most recent call last):
File "", line 1, in
File "/home/carl/python-environments/env/lib/python3.9/site-packages/imageai/Detection/init.py", line 17, in
from imageai.Detection.YOLOv3.models import yolo_main, tiny_yolo_main
File "/home/carl/python-environments/env/lib/python3.9/site-packages/imageai/Detection/YOLOv3/models.py", line 8, in
from keras.layers.normalization import BatchNormalization
ImportError: cannot import name 'BatchNormalization' from 'keras.layers.normalization' (/home/carl/python-environments/env/lib/python3.9/site-packages/keras/layers/normalization/init.py)
Already tried:
from keras.layers.normalization.batch_normalization import BatchNormalization
from tensorflow.keras.layers import BatchNormalization
But still gave me the same error.
Tensorflow v2.8.0 support tf.keras.layers.BatchNormalization()
import tensorflow as tf
tf.keras.layers.BatchNormalization()
I am attempting to build an exe file that utilizes statsmodels via PyInstaller, from the terminal. The exe file is built by PyInstaller without error--however when run I get the following error:
ModuleNotFoundError: No module named 'statsmodels.__init__._version'
I have tried installing the statsmodels hook in the PyInstaller "hooks" directory, as described here, however that does not work. I have also attempted to explicitly import statsmodels.init._version in my Python code, but that has also not worked. Below is the top of my code where I import the packages:
import pandas as pd
from sklearn import preprocessing
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score
from sklearn.neighbors import NearestNeighbors, KNeighborsClassifier
import numpy as np
import matplotlib.pylab as plt
from sklearn.linear_model import LogisticRegression
import statsmodels.api as sm
from scipy import stats
from sklearn import datasets, neighbors
from mlxtend.plotting import plot_decision_regions
import seaborn as sns
from tqdm import tqdm
import matplotlib.ticker as mtick
from tkinter import *
from tkinter.filedialog import askopenfilename
import sklearn.utils._cython_blas
import sklearn.neighbors.typedefs
import statsmodels.tsa.statespace._filters
import statsmodels.tsa.statespace._filters._conventional
import statsmodels.tsa.statespace._filters._univariate
import statsmodels.tsa.statespace._filters._univariate_diffuse
import statsmodels.tsa.statespace._filters._inversions
import statsmodels.tsa.statespace._smoothers
import statsmodels.tsa.statespace._smoothers._conventional
import statsmodels.tsa.statespace._smoothers._univariate
import statsmodels.tsa.statespace._smoothers._univariate_diffuse
import statsmodels.tsa.statespace._smoothers._classical
import statsmodels.tsa.statespace._smoothers._alternative
import statsmodels.__init__._version
Here is the traceback of the error:
File "LR.py", line 9, in <module>
File "/opt/anaconda3/lib/python3.7/site-packages/PyInstaller/loader/pyimod03_importers.py", line 493, in exec_module
exec(bytecode, module.__dict__)
File "statsmodels/api.py", line 32, in <module>
File "/opt/anaconda3/lib/python3.7/site-packages/PyInstaller/loader/pyimod03_importers.py", line 493, in exec_module
exec(bytecode, module.__dict__)
File "statsmodels/__init__.py", line 2, in <module>
ModuleNotFoundError: No module named 'statsmodels.__init__._version'
What can I do to fix this so that the exe file includes the statsmodels package and can be run?
statsmodels version is accessed through the __version__ attribute.
import statsmodels
print(statsmodels.__version__)
which shows
v0.12.0rc0+20.g98dfc0073
on my system.
You should replace import statsmodels.__init__._version with from statsmodels.__init__ import __version__
I'm using Tensorflow v2.1.0 which is compiled from source. How can I import any function from conv_utils module? E.g. convert_data_format(), conv_output_length(), normalize_tuple(), etc.
I wanna create my own convolution/pooling layer. "Everything" almost looks different when I wanna migrate from import keras to from tensorflow import keras.
This is what I did
> python -c "from tensorflow.keras.utils import conv_utils"
Traceback (most recent call last):
File "<string>", line 1, in <module>
ImportError: cannot import name 'conv_utils' from 'tensorflow.keras.utils' ...
Thank you.
Try this:
from tensorflow.python.keras.utils import conv_utils
This is the new way to import conv_utils in Tensorflow 2.1.0:
import tensorflow.keras.utils as conv_utils