T.E.D.D. 1140 XXL

T.E.D.D. 1104 S: 138M Parameters. 1.6GB

Hyper Parameters

control_mode: keyboard
cnn_model_name: efficientnet_v2_l
pretrained_cnn: true
encoder_type: transformer
embedded_size: 896
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	🌞	☀️	53.2	84.4	46.2
City	🌞	☔	51.4	83.4	46.3
City	🌛	☀️	54.3	85.6	46.3
City	🌛	☔	47.3	82.3	49.9
Highway	🌞	☀️	72.7	97.7	40.6
Highway	🌞	☔	70.6	99.3	39.6
Highway	🌛	☀️	77.9	99.3	45.7
Highway	🌛	☔	70.9	97.6	30.8

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

T.E.D.D. 1140 S

T.E.D.D. 1104 S: 26M Parameters. 260MB

Hyper Parameters

control_mode: keyboard
cnn_model_name: efficientnet_v2_s
pretrained_cnn: true
encoder_type: transformer
embedded_size: 384
nhead: 8
num_layers_encoder: 2
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	🌞	☀️	51.0	83.0	46.3
City	🌞	☔	49.0	82.5	45.2
City	🌛	☀️	56.3	86.6	49.0
City	🌛	☔	49.4	81.4	42.5
Highway	🌞	☀️	70.3	100	68.5
Highway	🌞	☔	71.2	100	37.6
Highway	🌛	☀️	80.9	100	49.1
Highway	🌛	☔	69.3	100	61.1

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

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.

T.E.D.D. 1140 Base (DEPRECATED)

THIS MODEL WILL NOT WORK WITH THE CURRENT VERSION OF TEDD1104

T.E.D.D. 1104 Base: 34.6M Parameters. 415.9Mb

Hyper Parameters

cnn_model_name: efficientnet_b4
control_mode: keyboard
dropout_cnn_out: 0.3
dropout_encoder: 0.1
dropout_encoder_features: 0.3
embedded_size: 512
encoder_type: transformer
learning_rate: 1.0e-05
lstm_hidden_size: 512
mask_prob: 0.2
nhead: 8
num_layers_encoder: 4
positional_embeddings_dropout: 0.1
pretrained_cnn: true
sequence_size: 5
weight_decay: 0.001
weights: null
num_epochs: 12
batch_size: 64

Performance

	Time	Weather	Micro-Acc K@1	Micro-Acc k@3	Macro-Acc K@1
City	🌞	☀️	49.8	83.8	44.1
City	🌞	☔	52.1	84.7	46.1
City	🌛	☀️	54.5	86.9	48.0
City	🌛	☔	48.8	82.5	43.2
Highway	🌞	☀️	65.6	100.0	53.2
Highway	🌞	☔	70.6	98.0	54.2
Highway	🌛	☀️	71.3	100.0	52.3
Highway	🌛	☔	67.7	100.0	50.9