T.E.D.D. 1140 M

T.E.D.D. 1104 S: 68M Parameters. 685MB

Hyper Parameters

control_mode: keyboard
cnn_model_name: efficientnet_v2_m
pretrained_cnn: true
encoder_type: transformer
embedded_size: 512
nhead: 8
num_layers_encoder: 4
learning_rate: 1e-05
optimizer_name: adafactor
scheduler_name: cosine
warmup_factor: 0.05
max_epochs: 20
batch_size: 32
mask_prob: 0.2
dropout_cnn_out: 0.3
dropout_encoder: 0.1
dropout_encoder_features: 0.3
positional_embeddings_dropout: 0.1
weight_decay: 1e-3

Accuracy in the test datasets:

	Time	Weather	Micro-Acc K@1	Micro-Acc k@3	Macro-Acc K@1
City	🌞	☀️	52.9	84.1	43.1
City	🌞	☔	49.9	81.3	42.2
City	🌛	☀️	54.7	85.1	48.4
City	🌛	☔	49.5	81.1	41.1
Highway	🌞	☀️	62.5	99.2	43.1
Highway	🌞	☔	71.9	99.3	39.2
Highway	🌛	☀️	79.4	99.3	45.3
Highway	🌛	☔	63.0	97.2	47.2

The release includes the best epoch in the development set and the last epoch.