stratospark/food-101-mobile

Deep Learning Food Classifier for iOS using Keras and Tensorflow

food-101-mobile's Issues

How to get the independent Class Probability?

How to get the independent Probability?
I am trying to detect the food images in composite image using CNN,In the final layer I have done the Softmax classifier to get the prediction,I am getting the result as sum of all probabilities equals to one.But I need to get the independent class probabilities. So can you guide me is there any function that can give the independent class probability.So can you guide me how to get the independent class Probability,Using which function.

Thank you.

iOS example app

Can you suggest me an iOS example to proceed after the final steps? New to iOS :)
Your example base is non-existent now?

Getting this Error. Making Same app for Android and Open source that. Help wanted.

Traceback (most recent call last):
File "", line 1, in
NameError: name 'model' is not defined

y = model(x)
Traceback (most recent call last):
File "", line 1, in
NameError: name 'model' is not defined

You must feed a value for placeholder tensor 'Placeholder' with dtype float and shape [?,784] [[Node: Placeholder = Placeholder[dtype=DT_FLOAT, shape=[?,784]

I am getting this error when trying to read from .pb file. You can see complete error report in this link:
https://stackoverflow.com/questions/47492130/you-must-feed-a-value-for-placeholder-tensor-placeholder-with-dtype-float-and?noredirect=1#comment82006554_47492130

Suggestion on Creating a Deep Learning iOS App with Keras and Tensorflow¶

I am really happy finding the tutorial about how to transfer a Keras model to a Pb file. It works good and gives me much inspiration. In my experiment, I also try many other methods to simplify the procedure. Here are my suggestions for your reference:

1. In[6] we would better assign a name for input which may make it easier in [84] to get tensor by name
2. At the very beginning, before import keras model we can use K.set_learning_phase(0) to make things much more easier. By introducing K.set_learning_phase(0), we do not need to delete switch node and also do not need to modify the learning phase. This seems great I think

load_model failing with custom Keras model

Hello,

A really great tutorial!

Just what I was looking for. When I am trying this with a custom Keras model,the load_model() seems to be failing. I have a custom model trained in Keras that I am trying to bundle it as an android app.

model = load_model('my_weights_tensorflow.h5')

ValueError Traceback (most recent call last)
in ()
----> 1 model = load_model('my_weights_tensorflow.h5')

/usr/local/lib/python2.7/dist-packages/keras/models.pyc in load_model(filepath, custom_objects)
138 model_config = f.attrs.get('model_config')
139 if model_config is None:
--> 140 raise ValueError('No model found in config file.')
141 model_config = json.loads(model_config.decode('utf-8'))
142 model = model_from_config(model_config, custom_objects=custom_objects)

ValueError: No model found in config file.

I could load your model without any problem.

Any input would be of great assistance.

Thanks in advance!