I am trying to download a pre-trained tensorflow model. I am using the below code
import numpy as np
import time
import PIL.Image
import IPython.display as display
import matplotlib.pylab as plt
import tensorflow as tf
import tensorflow_hub as hub
import datetime
from tensorflow.keras.preprocessing import image
from dateutil import parser
from keras.applications.inception_v3 import InceptionV3
model = InceptionV3()
model.summary()
I am getting the following error
AttributeError: module 'dateutil' has no attribute 'parser'
I am using python -3.7 and TF-2.7, python-dateutil-2.8.1
Please help me fix this. Thank You :)
The correct import syntax is:
import dateutil.parser
and then:
parser.parse(time_string)
or:
from dateutil.parser import parse
parse(time_string)
Documentation: https://dateutil.readthedocs.io/en/stable/parser.html
Related
import torch
from torch import nn
from torch.nn import functional as F
from .utils import (
round_filters,
round_repeats,
drop_connect,
get_same_padding_conv2d,
get_model_params,
efficientnet_params,
load_pretrained_weights,
Swish,
MemoryEfficientSwish,
calculate_output_image_size
)
I tried a lot but I could'nt find the solution to fix this error
from .utils import round_filters
It gives error: relative import with no known parent package. See discussion for this in stackoverflow thread.
You need to install EfficientNet-PyTorch package. About this package
!pip install EfficientNet-PyTorch
And add parent package as from efficientnet_pytorch.utils import round_filters
So your code should be
import torch
from torch import nn
from torch.nn import functional as F
from efficientnet_pytorch.utils import (
round_filters,
round_repeats,
drop_connect,
get_same_padding_conv2d,
get_model_params,
efficientnet_params,
load_pretrained_weights,
Swish,
MemoryEfficientSwish,
calculate_output_image_size
)
When I am running datasets_utils.py from '/usr/local/lib/python3.7/dist-packages/torchtext/data/datasets_utils.py' in Google Colab, the following error occurs even with the most updated versions of Python packages:
ImportError: cannot import name 'functional_datapipe' from 'torch.utils.data' (/usr/local/lib/python3.7/dist-packages/torch/utils/data/init.py)
Are there any solutions to solve such errors, as I could not find functional_datapipe even in the official torch.utils.data documentation? The following is excerpt from datasets._utils.py in the Google Colab environment
import functools
import inspect
import os
import io
import torch
from torchtext.utils import (
validate_file,
download_from_url,
extract_archive,
)
from torch.utils.data import functional_datapipe, IterDataPipe
from torch.utils.data.datapipes.utils.common import StreamWrapper
import codecs
It might be available only on torchdata.datapipes
I am trying to use the pretrained SciBERT model (https://huggingface.co/allenai/scibert_scivocab_uncased) from Huggingface to evaluate masked words in scientific/biomedical text for bias using CrowS-Pairs (https://github.com/nyu-mll/crows-pairs/). The CrowS-Pairs code works great with the built in models like BERT.
I modified the code of metric.py with the goal of allowing an option of using the SciBERT model -
import os
import csv
import json
import math
import torch
import argparse
import difflib
import logging
import numpy as np
import pandas as pd
from transformers import BertTokenizer, BertForMaskedLM
from transformers import AlbertTokenizer, AlbertForMaskedLM
from transformers import RobertaTokenizer, RobertaForMaskedLM
from transformers import AutoTokenizer, AutoModelForMaskedLM
and get the following error
2021-06-21 17:11:38.626413: I tensorflow/stream_executor/platform/default/dso_loader.cc:53] Successfully opened dynamic library libcudart.so.11.0
Traceback (most recent call last):
File "metric.py", line 15, in <module>
from transformers import AutoTokenizer, AutoModelForMaskedLM
ImportError: cannot import name 'AutoModelForMaskedLM' from 'transformers' (/usr/local/lib/python3.7/dist-packages/transformers/__init__.py)
Later in the Python file, the AutoTokenizer and AutoModelForMaskedLM are defined as
tokenizer = AutoTokenizer.from_pretrained("allenai/scibert_scivocab_uncased")
model = AutoModelForMaskedLM.from_pretrained("allenai/scibert_scivocab_uncased")
Libraries
huggingface-hub-0.0.8
sacremoses-0.0.45
tokenizers-0.10.3
transformers-4.7.0
The error occurs with and without GPU support.
Try this:
tokenizer = BertTokenizer.from_pretrained("allenai/scibert_scivocab_uncased", do_lower_case=True)
model = BertForMaskedLM.from_pretrained("allenai/scibert_scivocab_uncased")
I had a problem running my http://127.0.0.1:5000/iris/1/1/1/1. It keeps getting an internal server error. I don't know why and It kept doing it. I tried many ways but it still didn't work. Its kept saying "modulenotfounderror: no module named 'sklearn.linear_model.logistic' ." Is there a way to resolve this.
from flask import Flask, render_template
from flask import jsonify
from datetime import datetime
import numpy as np
import pandas as pd
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
from sklearn.linear_model import LinearRegression
from sklearn.linear_model import LogisticRegression
from mpl_toolkits import mplot3d
from sklearn.datasets import load_iris
from sklearn import datasets
import csv
import joblib
app = Flask(__name__)
def predict(sepal_length, sepal_width, petal_length, petal_width):
test_data = np.array([sepal_length, sepal_width, petal_length,petal_width])
test_data = test_data.reshape(1,-1)
file = open("data/iris.pkl","rb")
trained_model = joblib.load(file)
prediction = trained_model.predict(test_data)
return prediction
#app.route("/iris/<sepal_length>/<sepal_width>/<petal_length>/<petal_width>")
def iris(sepal_length, sepal_width, petal_length, petal_width):
result=predict(np.double(sepal_length),
np.double(sepal_width),
np.double(petal_length),
np.double(petal_width))
if result[0]==0:
hasil='Setosa'
elif result[0]==1:
hasil='Versicolor'
else:
hasil='Virginica'
return hasil
Check the version of scikit-learn, maybe it is 0.21.x. upgrade the version.
Here is scikit-learn stable:
https://github.com/scikit-learn/scikit-learn/tree/master/sklearn/linear_model
i have the following python imports with in Jupyter Notebook.
import numpy as np
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras.layers import Dense, Activation
from tensorflow.keras.optimizers import Adam
from tensorflow.keras.metrics import categorical_crossentropy
from tensorflow.keras.preprocessing.image import ImageDataGenerator
from tensorflow.keras.preprocessing import image
from tensorflow.keras.models import Model
from tensorflow.keras.applications import imagenet_utils
from sklearn.metrics import confusion_matrix
import itertools
import os
import shutil
import random
import matplotlib.pyplot as plt
%matplotlib inline
But i keep getting the following error
ImportError: cannot import name 'imagenet_utils' from 'tensorflow.keras.applications' (C:\ProgramData\Anaconda3\lib\site-packages\tensorflow\python\keras\api_v2\keras\applications_init_.py)
when i search for **cannot import name 'imagenet_utils' from 'tensorflow.keras.applications' **in google i dont get much helpful information.
Has anyone come across this at all?
change
from tensorflow.keras.applications import imagenet_utils
to
from keras.applications import imagenet_utils
i managed to solve my issue.
first i ran the following to update all modules.
conda update --all
Then i used 'from keras.applications import imagenet_utils'
instead of '#from tensorflow.keras import imagenet_utils'