huaxiuyao / metast Goto Github PK

Traceback (most recent call last):
File "maml/train_model.py", line 138, in
main()
File "maml/train_model.py", line 71, in main
model.construct_model()
File "/home/sao/MetaST/maml/maml.py", line 332, in construct_model
parallel_iterations=self.meta_batch_size)
File "/home/sao/anaconda3/envs/tf/lib/python3.6/site-packages/tensorflow/python/ops/functional_ops.py", line 497, in map_fn
maximum_iterations=n)
File "/home/sao/anaconda3/envs/tf/lib/python3.6/site-packages/tensorflow/python/ops/control_flow_ops.py", line 3556, in while_loop
return_same_structure)
File "/home/sao/anaconda3/envs/tf/lib/python3.6/site-packages/tensorflow/python/ops/control_flow_ops.py", line 3087, in BuildLoop
pred, body, original_loop_vars, loop_vars, shape_invariants)
File "/home/sao/anaconda3/envs/tf/lib/python3.6/site-packages/tensorflow/python/ops/control_flow_ops.py", line 3022, in _BuildLoop
body_result = body(*packed_vars_for_body)
File "/home/sao/anaconda3/envs/tf/lib/python3.6/site-packages/tensorflow/python/ops/control_flow_ops.py", line 3525, in
body = lambda i, lv: (i + 1, orig_body(*lv))
File "/home/sao/anaconda3/envs/tf/lib/python3.6/site-packages/tensorflow/python/ops/functional_ops.py", line 486, in compute
packed_fn_values = fn(packed_values)
File "/home/sao/MetaST/maml/maml.py", line 284, in task_metalearn
fast_weights = self.update(self.combine_loss(task_lossa, task_clossa), weights)
File "/home/sao/MetaST/maml/maml.py", line 272, in combine_loss
return loss + self.cluster_loss_weight * closs
File "/home/sao/anaconda3/envs/tf/lib/python3.6/site-packages/tensorflow/python/ops/math_ops.py", line 839, in r_binary_op_wrapper
x = ops.convert_to_tensor(x, dtype=y.dtype.base_dtype, name="x")
File "/home/sao/anaconda3/envs/tf/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 1039, in convert_to_tensor
return convert_to_tensor_v2(value, dtype, preferred_dtype, name)
File "/home/sao/anaconda3/envs/tf/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 1097, in convert_to_tensor_v2
as_ref=False)
File "/home/sao/anaconda3/envs/tf/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 1175, in internal_convert_to_tensor
ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
File "/home/sao/anaconda3/envs/tf/lib/python3.6/site-packages/tensorflow/python/framework/constant_op.py", line 304, in _constant_tensor_conversion_function
return constant(v, dtype=dtype, name=name)
File "/home/sao/anaconda3/envs/tf/lib/python3.6/site-packages/tensorflow/python/framework/constant_op.py", line 245, in constant
allow_broadcast=True)
File "/home/sao/anaconda3/envs/tf/lib/python3.6/site-packages/tensorflow/python/framework/constant_op.py", line 283, in _constant_impl
allow_broadcast=allow_broadcast))
File "/home/sao/anaconda3/envs/tf/lib/python3.6/site-packages/tensorflow/python/framework/tensor_util.py", line 454, in make_tensor_proto
raise ValueError("None values not supported.")
ValueError: None values not supported.

I am getting this error while trying to train the model. How to resolve this ? Any suggestions?
I have tried with Tensorflow version 1.13 and 1.14. Same error persists.

Thank you for sharing the code. I have a question about test_model.py.
Why, in test_model.py, is testing done first on the test set of the target city (line28-39) and then fine-tuning on the training set of the target city(line49-60)? Shouldn't it be fine-tuning first and then testing?

Hey, I have time-series data of speed over time. Could you please share the script to prepare the data in the format used in your model.

Hello, I've been looking at the testing procedure (test_model.py) and there are some things which are not clear to me. I can see that you fine-tune the model on the source city training dataset (line 49 to 60) which is ok, but then you are validating (line 28 to 38) on the test dataset? From my understanding you should not use the testing dataset until you're done fine-tuning. How do you get the final results which are presented in the paper? Select the best model or?