chaitjo

/

all-atom-diffusion-transformer

like 3

• "src.models.denoisers.dit.Mlp",
• "torch.optim.adamw.AdamW",
• "torch._utils._rebuild_parameter",
• "torch.nn.modules.linear.Identity",
• "torch.device",
• "functools.partial",
• "__builtin__.set",
• "torch.nn.modules.linear.NonDynamicallyQuantizableLinear",
• "__builtin__.dict",
• "collections.OrderedDict",
• "omegaconf.nodes.AnyNode",
• "src.models.denoisers.dit.FinalLayer",
• "omegaconf.base.Metadata",
• "torch.nn.modules.linear.Linear",
• "src.models.interpolants.flow_matching.FlowMatchingInterpolant",
• "torch.nn.modules.activation.SiLU",
• "torch.nn.modules.sparse.Embedding",
• "collections.defaultdict",
• "src.models.denoisers.dit.TimestepEmbedder",
• "src.models.denoisers.dit.DiTBlock",
• "src.models.denoisers.dit.DiT",
• "torch.nn.modules.container.Sequential",
• "omegaconf.dictconfig.DictConfig",
• "torch.nn.modules.activation.GELU",
• "typing.Any",
• "src.models.denoisers.dit.LabelEmbedder",
• "torch.FloatStorage",
• "torch._utils._rebuild_tensor_v2",
• "omegaconf.base.ContainerMetadata",
• "torch.nn.modules.normalization.LayerNorm",
• "torch.nn.modules.dropout.Dropout",
• "torch.nn.modules.container.ModuleList",
• "torch.nn.modules.activation.MultiheadAttention"

How to fix it?