siavash-khodadadeh

In the readme file train-shots is 10, however, it seems that this should be 5-way 1-shot. Am I missing something? # transductive 1-shot 5-way Omniglot. python -u run_omniglot.py --shots 1...

Does this mean that during meta-learning we are training with 5-way 10-shot? but for the test, we evaluate on 5-way, 1-shot?

I have the same issue. More particularly, I get this error message: failed to execute: make -j16 -C \/Build > \/BuildOpenNI.txt

Hello, Yes, that would be possible. First, you need to subclass the Database class [here](https://github.com/siavash-khodadadeh/MetaLearning-TF2.0/blob/master/databases/data_bases.py#L11). Then you need to implement two abstract methods: First one is [this method](https://github.com/siavash-khodadadeh/MetaLearning-TF2.0/blob/master/databases/data_bases.py#L86) that returns...

I see. I believe this is because python interpreter does not find the models module. The way to resolve it is to add the path of the root of the...

So you need to write a function similar to [this](https://github.com/siavash-khodadadeh/MetaLearning-TF2.0/blob/b8f7010647e007bd902529365da27fdbd83b7d73/databases/parse_mixins.py#L6) in your database class. In fact `image = tf.image.decode_jpeg(tf.io.read_file(example_address))` this line loads your image assuming it is jpeg image. Note...

Yes, it is done in the algorithm part. As an example, you can look at [this](https://github.com/siavash-khodadadeh/MetaLearning-TF2.0/blob/master/models/maml/maml_airplane.py#L7). Just make an instance of your dataset class and pass it to MAML algorithm....

Can you please check your image shapes when you load them by TensorFlow? To do that you can add a line `tf.print(image.shape)` before returning the image in [this](https://github.com/siavash-khodadadeh/MetaLearning-TF2.0/blob/b8f7010647e007bd902529365da27fdbd83b7d73/databases/parse_mixins.py#L11) function.

I would definitely try to do something like this to see if it solves the problem. `image = image[:, :, :3]` The reason why these are `None` is because of...

Hello Samiur, Thank you for using this repository for your project. Glad that you were able to run the training. Just to make sure in the future people can search...