rohanpillai20 / image-classification-by-keras-and-tensorflow

Image Classification using Keras as well as Tensorflow.

Image-Classification-by-Keras-and-Tensorflow

Image Classification using Keras as well as Tensorflow. Image Classification is a Machine Learning module that trains itself from an existing dataset of multiclass images and develops a model for future prediction of similar images not encountered during training. Developed using Convolutional Neural Network (CNN).

This repository contains implementation for multiclass image classification using Keras as well as Tensorflow. These two codes have no interdependecy on each other.

Convolutional Neural Network (CNN)

To be updated soon.

Modules Required

You can download the modules in the respective requirements.txt for each implementation.

Implementation using Keras

sys
os
time
keras==2.2.0
numpy==1.14.5

Implementation using Tensorflow

os
time
cv2
glob
dataset==1.1.0
tensorflow==1.15.0
numpy==1.14.5
sklearn==0.19.1

Steps

1. Install the modules required based on the type of implementation.
2. Download the dataset you want to train and predict your system with. For sample data, you can download the 101_ObjectCategories (131Mbytes).
3. Construct the folder sub-structure required.

Keras

Tensorflow

image-classification-by-keras-and-tensorflow's Issues

How can I plot a confusion matrix?

How can I plot a confusion matrix?

Learning Transfer

Hello,

How to do if I want to add classes to an existing training?

Thanks,

2 Errors

Error 1:
WARNING:tensorflow:From :41: calling argmax (from tensorflow.python.ops.math_ops) with dimension is deprecated and will be removed in a future version.
Instructions for updating:
Use the axis argument instead

Error 2:
saver.save(session, 'C:\

Does not save anything, so when calling Prediction.py it says:
Exception: File models/trained_model.meta does not exist.

Error 3:
Warning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses
import imp

Prediction

if answer == 1:
print("Predicted: chair")
elif answer == 0:
print("Predicted: Motorbikes")
elif answer == 2:
print("Predicted: soccer_ball")

How can you mention (if 1 it is chair, if 2 its ball?). How do you categorize it?

ValueError: Error when checking target: expected dense_2 to have shape (1,) but got array with shape (6,)

ValueError: Error when checking target: expected dense_2 to have shape (1,) but got array with shape (6,)

Fine tuning

Can you help me to fine tuning this to get better result because i try to use grayscale images

Image Labels

why keras code doesn't have labels for images ?
how the model this is chair or somthing else without labeling , i mean in traning set ?