According to the fsns curriculum, the attribute problem that needs to be adjusted

[MS-MA]

hello Bartzi,I have a new problem about train_fsns.py
1.if i created a curriculum learning strategy,like the follows:

[
{
"train":"/home/data/fsns/image/train/train_swap_max2words.csv",
"validation":"/home/data/fsns/image/validation/validation_swap_max2words.csv"
},
{
"train":"/home/data/fsns/image/train/train_swap_3words.csv",
"validation":"/home/data/fsns/image/validation/validation_swap_3words.csv"
},
{ "train":"/home/data/fsns/image/train/train_swap_4words.csv",
"validation":"/home/data/fsns/image/validation/validation_swap_4words.csv"
},
{ "train":"/home/data/fsns/image/train/train_swap_5words.csv",
"validation":"/home/data/fsns/image/validation/validation_swap_5words.csv"
},
{ "train":"/home/data/fsns/image/train/train_swap_6words.csv",
"validation":"/home/data/fsns/image/validation/validation_swap_6words.csv"
}
]

nextly,how should i adjust the attributes like the below in train_fsns.py ?

attributes_to_adjust = [
('num_timesteps', ['predictor', 'localization_net']),
('num_timesteps', ['predictor', 'recognition_net']),
('num_timesteps', ['lossfun', 'self']),
('num_labels', ['predictor', 'recognition_net']),
]

I am confused,could you give me a case?
2. Can you provide the curcululum learning strategy you used in the model_35000.npz model training?

very thanks@Bartzi

[Bartzi]

Hmm, I'll have to look around whether I can still find it. But I think it did not really use the curriculum for this. I rather first created a model with just one word, then I used this as a basis for creating a second model with two words, and so on and so on. I did all the curriculum steps by myself.
That's all I can say right now.