ledas

/

ttm-re-pretrain

like 0

• "transformers.models.roberta.modeling_roberta.RobertaEmbeddings",
• "torch._utils._rebuild_tensor_v2",
• "transformers.models.roberta.modeling_roberta.RobertaSelfOutput",
• "__builtin__.set",
• "torch._utils._rebuild_parameter",
• "torch.nn.modules.normalization.LayerNorm",
• "torch.nn.modules.container.ModuleList",
• "collections.OrderedDict",
• "torch.float32",
• "torch._C._nn.gelu",
• "transformers.models.roberta.configuration_roberta.RobertaConfig",
• "transformers.models.roberta.modeling_roberta.RobertaSdpaSelfAttention",
• "transformers.models.roberta.modeling_roberta.RobertaOutput",
• "transformers.models.roberta.modeling_roberta.RobertaEncoder",
• "torch.nn.modules.dropout.Dropout",
• "transformers.models.roberta.modeling_roberta.RobertaPooler",
• "torch.nn.modules.activation.Tanh",
• "torch.nn.modules.sparse.Embedding",
• "torch.FloatStorage",
• "torch.LongStorage",
• "transformers.models.roberta.modeling_roberta.RobertaAttention",
• "torch.nn.modules.linear.Linear",
• "transformers.activations.GELUActivation",
• "transformers.models.roberta.modeling_roberta.RobertaModel",
• "transformers.models.roberta.modeling_roberta.RobertaLayer",
• "transformers.models.roberta.modeling_roberta.RobertaIntermediate"

How to fix it?

• "torch.FloatStorage",
• "torch.nn.modules.container.ModuleList",
• "ttm.TokenTuringMachineEncoder",
• "ttm.TokenAddEraseWrite",
• "ttm.MLPBlock",
• "ttm.TokenLearnerModuleV11",
• "torch.nn.modules.dropout.Dropout",
• "torch.nn.modules.transformer.TransformerEncoderLayer",
• "torch.nn.modules.transformer.TransformerEncoder",
• "torch._utils._rebuild_parameter",
• "torch.nn.modules.activation.GELU",
• "torch._utils._rebuild_tensor_v2",
• "torch.nn.functional.relu",
• "torch.nn.modules.linear.Linear",
• "torch.nn.modules.linear.NonDynamicallyQuantizableLinear",
• "__builtin__.set",
• "ttm.TokenTuringMachineUnit",
• "collections.OrderedDict",
• "torch.nn.modules.normalization.LayerNorm",
• "torch.nn.modules.activation.MultiheadAttention"

How to fix it?

• "__builtin__.set",
• "torch._utils._rebuild_parameter",
• "collections.OrderedDict",
• "torch._utils._rebuild_tensor_v2",
• "torch.nn.modules.linear.Linear",
• "torch.FloatStorage"

How to fix it?