Not able to resolve the error ModuleNotFoundError: No module named 'tensorflow.compat' in tensorflow and flask

Question

I am trying to serve a TensorFlow model I built with flask. While I was running the flask code, it came up with this error I reinstalled conda but the error persisted. The thing is now even if I import tensorflow, this error comes up. I tried on another device which didn't have conda but just vanilla python. The same error came up.

this thread doesn't help

I will post the entire error here:

>>> import tensorflow as tf
2021-05-18 13:20:02.804699: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'cudart64_110.dll'; dlerror: cudart64_110.dll not found
2021-05-18 13:20:02.830901: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "G:\AiDEV\Kid\CNN\myEnv\lib\site-packages\tensorflow\__init__.py", line 41, in <module>
    from tensorflow.python.tools import module_util as _module_util
  File "G:\AiDEV\Kid\CNN\myEnv\lib\site-packages\tensorflow\python\__init__.py", line 46, in <module>
    from tensorflow.python import data
  File "G:\AiDEV\Kid\CNN\myEnv\lib\site-packages\tensorflow\python\data\__init__.py", line 25, in <module>
    from tensorflow.python.data import experimental
  File "G:\AiDEV\Kid\CNN\myEnv\lib\site-packages\tensorflow\python\data\experimental\__init__.py", line 99, in <module>
    from tensorflow.python.data.experimental import service
  File "G:\AiDEV\Kid\CNN\myEnv\lib\site-packages\tensorflow\python\data\experimental\service\__init__.py", line 140, in <module>
    from tensorflow.python.data.experimental.ops.data_service_ops import distribute
  File "G:\AiDEV\Kid\CNN\myEnv\lib\site-packages\tensorflow\python\data\experimental\ops\data_service_ops.py", line 25, in <module>
    from tensorflow.python.data.experimental.ops import compression_ops
  File "G:\AiDEV\Kid\CNN\myEnv\lib\site-packages\tensorflow\python\data\experimental\ops\compression_ops.py", line 20, in <module>
    from tensorflow.python.data.util import structure
  File "G:\AiDEV\Kid\CNN\myEnv\lib\site-packages\tensorflow\python\data\util\structure.py", line 26, in <module>
    from tensorflow.python.data.util import nest
  File "G:\AiDEV\Kid\CNN\myEnv\lib\site-packages\tensorflow\python\data\util\nest.py", line 40, in <module>
    from tensorflow.python.framework import sparse_tensor as _sparse_tensor
  File "G:\AiDEV\Kid\CNN\myEnv\lib\site-packages\tensorflow\python\framework\sparse_tensor.py", line 28, in <module>
    from tensorflow.python.framework import constant_op
  File "G:\AiDEV\Kid\CNN\myEnv\lib\site-packages\tensorflow\python\framework\constant_op.py", line 29, in <module>
    from tensorflow.python.eager import execute
  File "G:\AiDEV\Kid\CNN\myEnv\lib\site-packages\tensorflow\python\eager\execute.py", line 28, in <module>
    from tensorflow.python.framework import ops
  File "G:\AiDEV\Kid\CNN\myEnv\lib\site-packages\tensorflow\python\framework\ops.py", line 26, in <module>
    from absl import app
  File "G:\AiDEV\Kid\CNN\myEnv\lib\site-packages\absl\app.py", line 35, in <module>
    import pdb
  File "c:\users\adi\appdata\local\programs\python\python39\lib\pdb.py", line 77, in <module>
    import code
  File "G:\AiDEV\Kid\CNN\New folder\code.py", line 4, in <module>
    from keras.preprocessing.image import ImageDataGenerator
  File "G:\AiDEV\Kid\CNN\myEnv\lib\site-packages\keras\__init__.py", line 22, in <module>
    from keras import distribute
  File "G:\AiDEV\Kid\CNN\myEnv\lib\site-packages\keras\distribute\__init__.py", line 18, in <module>
    from keras.distribute import sidecar_evaluator
  File "G:\AiDEV\Kid\CNN\myEnv\lib\site-packages\keras\distribute\sidecar_evaluator.py", line 18, in <module>
    import tensorflow.compat.v2 as tf
ModuleNotFoundError: No module named 'tensorflow.compat'

The flask code I am trying to run is this:

from flask import Flask, render_template, request
from werkzeug.utils import secure_filename
from werkzeug.datastructures import  FileStorage
from keras.preprocessing.image import ImageDataGenerator
import tensorflow as tf
import numpy as np
import os

try:
    import shutil
    shutil.rmtree('uploaded / image')
    # % cd uploaded % mkdir image % cd ..
    print()
except:
    pass

model = tf.keras.models.load_model('model')
app = Flask(__name__)

app.config['UPLOAD_FOLDER'] = 'uploaded / image'

@app.route('/')
def upload_f():
    return render_template('upload.html')

def finds():
    test_datagen = ImageDataGenerator(rescale = 1./255)
    vals = ['Positive', 'Negative'] 
    test_dir = 'uploaded'
    test_generator = test_datagen.flow_from_directory(
            test_dir,
            target_size =(224, 224),
            color_mode ="rgb",
            shuffle = False,
            class_mode ='categorical',
            batch_size = 1)

    pred = model.predict_generator(test_generator)
    print(pred)
    return str(vals[np.argmax(pred)])

@app.route('/uploader', methods = ['GET', 'POST'])
def upload_file():
    if request.method == 'POST':
        f = request.files['file']
        f.save(os.path.join(app.config['UPLOAD_FOLDER'], secure_filename(f.filename)))
        val = finds()
        return render_template('pred.html', ss = val)

if __name__ == '__main__':
    app.run()

Thanks in advance.

Answer 1

Let's look at your traceback in detail, particularly here:

import pdb
File "c:\users\adi\appdata\local\programs\python\python39\lib\pdb.py", line 77, in <module>
import code
File "G:\AiDEV\Kid\CNN\New folder\code.py", line 4, in <module>
from keras.preprocessing.image import ImageDataGenerator

You can clearly see here that import code inside pdb is jumping back to your own code, and that is your have a file called code.py, never name one of your scripts the same as a known package, since this will shadow the real python package.

The solution here is to rename your code.py file to something else.