↓ 8 callersMethod__init__(self, hidden_size, num_layers, ffn_kernel_size=9, dropout=0.0,
num_heads=2, use_pos_embed=Tr
video_generation/MegaTTS3/tts/modules/ar_dur/commons/transformer.py:677
↓ 7 callersMethod_get_per_token_logps(self, model, input_ids, attention_mask, pixel_values, image_grid_thw)
train/src/trainer/grpo_trainer.py:467
↓ 6 callersMethod__init__(self, c, num_heads, dropout, attention_dropout=0.1, relu_dropout=0.1,
kernel_size=9, ffn_hid
video_generation/MegaTTS3/tts/modules/ar_dur/commons/rot_transformer.py:544
↓ 5 callersMethod__init__(self, in_channels, out_channels, filter_channels, kernel_size, p_dropout=0., activation=None)
video_generation/MegaTTS3/tts/modules/ar_dur/commons/rel_transformer.py:255