I have a trouble after trying to parallelize data using nn.Dataparallel - pytorch

I didn't have any probelm without data parallelization, but after I put JUST A ONE LINE "model = nn.DataParallel(model)" the error message "TypeError: 'list' object is not callable" comes. If I push out that damn line the source works clean. plz help me
I can't do anything but just mad. I search that google and there are some ways to solve that error message but I can't do anything. Because nn.Dataparallel is already used by other coders. Sorry English is not my mothertongue.
if use_cuda:
[[[model = nn.DataParallel(model)]]]
model = model.cuda()
criterion = criterion.cuda()
print('cuda is used')
I just put model = nn.DataParallel(model) and error comes right after.
Traceback (most recent call last):
File "/home/scrcdeep2/YBJ/espnet_myself/Main.py", line 119, in
train(loader, model_, criterion_, optimizer_, use_cuda_, pretrained=None)
File "/home/scrcdeep2/YBJ/espnet_myself/Main.py", line 83, in train
outputs = model(inputs)
File "/home/scrcdeep2/YBJ/lib/python3.5/site-packages/torch/nn/modules/module.py", line 493, in call
result = self.forward(*input, **kwargs)
File "/home/scrcdeep2/YBJ/lib/python3.5/site-packages/torch/nn/parallel/data_parallel.py", line 151, in forward
replicas = self.replicate(self.module, self.device_ids[:len(inputs)])
File "/home/scrcdeep2/YBJ/lib/python3.5/site-packages/torch/nn/parallel/data_parallel.py", line 156, in replicate
return replicate(module, device_ids)
File "/home/scrcdeep2/YBJ/lib/python3.5/site-packages/torch/nn/parallel/replicate.py", line 114, in replicate
modules = list(network.modules())
TypeError: 'list' object is not callable

I met the same problem, because I've set moules = [list].
I changed my code like this:
def __init__(self, embedding_size, activation_function='relu'):
super().__init__()
self.act_fn = getattr(F, activation_function)
self.embedding_size = embedding_size
self.conv1 = nn.Conv2d(3, 32, 4, stride=2)
self.conv2 = nn.Conv2d(32, 64, 4, stride=2)
self.conv3 = nn.Conv2d(64, 128, 4, stride=2)
self.conv4 = nn.Conv2d(128, 256, 4, stride=2)
self.fc = nn.Identity() if embedding_size == 1024 else nn.Linear(1024, embedding_size)
if not args.MultiGPU:
self.modules = [self.conv1, self.conv2, self.conv3, self.conv4]

Related

[Pytorch]Error when using DistributedDataParallel in the broadcasting stage of initialization

I'm currently working on GroupFormer which used DistributedDataParallel for trainning. The error message is listed below and it shows that the error is caused by tensor size mismatch while broadcasting in the initialization stage.
This error first occurred when I set --nnodes=1 and --nproc_per_node=2 (train with 2 GPU on 1 computer), but even when I set --nnodes=1 and --nproc_per_node=1 (train with 1 GPU on 1 computer) , the same error still occurred. As far as I know, these broadcasting functions(_sync_params_and_buffers,dist._broadcast_coalesced) are designed to broadcast parameters from main GPU to others, it doesn't make sense that this error still occurred when training with 1 GPU.
Traceback (most recent call last):
File "main.py", line 53, in <module>
main()
File "main.py", line 43, in main
group_helper = Group(config, work_dir=config['basedir'])
File "/home/disk1/wgf/project/GroupFormer/group/group.py", line 54, in __init__
self._build()
File "/home/disk1/wgf/project/GroupFormer/group/group.py", line 59, in _build
self._build_model()
File "/home/disk1/wgf/project/GroupFormer/group/group.py", line 104, in _build_model
self.model = DistributedDataParallel(model.cuda(), device_ids=[self.rank % torch.cuda.device_count()],
File "/home/disk1/wgf/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/nn/parallel/distributed.py", line 648, in __init__
_sync_module_states(
File "/home/disk1/wgf/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/distributed/utils.py", line 113, in _sync_module_states
_sync_params_and_buffers(
File "/home/disk1/wgf/anaconda3/envs/pytorch/lib/python3.8/site-packages/torch/distributed/utils.py", line 131, in _sync_params_and_buffers
dist._broadcast_coalesced(
RuntimeError: The size of tensor a (64) must match the size of tensor b (0) at non-singleton dimension 3
The command I used for training is also listed below.
python -m torch.distributed.run --nnodes=1 --nproc_per_node=1 --node_rank=0 --master_port=22332 main.py
I tried using a simple model as below, and this error did not occur. How could this happen? What could be wrong with the original model in GroupFormer? And how can I fix it?
class ToyModel(nn.Module):
def __init__(self):
super(ToyModel, self).__init__()
self.conv1 = nn.Conv2d(3, 6, 5)
self.pool = nn.MaxPool2d(2, 2)
self.conv2 = nn.Conv2d(6, 16, 5)
self.fc1 = nn.Linear(16 * 5 * 5, 120)
self.fc2 = nn.Linear(120, 84)
self.fc3 = nn.Linear(84, 10)
def forward(self, x):
x = self.pool(F.relu(self.conv1(x)))
x = self.pool(F.relu(self.conv2(x)))
x = x.view(-1, 16 * 5 * 5)
x = F.relu(self.fc1(x))
x = F.relu(self.fc2(x))
x = self.fc3(x)
return x

Why does pytorch lightnings configure_optimizer throw AssertionError: param group must be a dict?

I have set up multiple pytorch lightning projects in the past and while setting up a new quick demo project, I stumbled across this weird error and somehow I cannot get rid of it.
Here are the relevant sections of my model file..
class TSModel(pl.LightningModule):
def __init__(self):
super().__init__()
self.backbone = nn.Sequential(
nn.Conv2d(3, 10, kernel_size=(3, 3), padding=(1, 1)),
nn.ReLU(),
nn.MaxPool2d(kernel_size=(2, 2), stride=(2, 2))
)
self.classifier = nn.Sequential(
nn.Linear(10*16*16, 128),
nn.ReLU(),
nn.Linear(128, 10)
)
self.criterion = nn.CrossEntropyLoss()
def forward(self, x):
N = x.shape[0]
x = self.backbone(x)
x = x.view(N, -1)
return self.classifier(x)
def configure_optimizers(self):
params = [p for p in self.parameters() if p.requires_grad]
return torch.optim.AdamW(self.parameters())
However, when starting the training process, the program exits and the following is thrown:
Traceback (most recent call last):
File "/torchserve-example/main.py", line 25, in <module>
ts_train()
File "/torchserve-example/main.py", line 21, in ts_train
trainer.fit(model, datamodule)
File ".local/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 458, in fit
self._run(model)
File ".local/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 715, in _run
self.accelerator.setup(self, model) # note: this sets up self.lightning_module
File ".local/lib/python3.8/site-packages/pytorch_lightning/accelerators/cpu.py", line 39, in setup
return super().setup(trainer, model)
File ".local/lib/python3.8/site-packages/pytorch_lightning/accelerators/accelerator.py", line 92, in setup
self.setup_optimizers(trainer)
File ".local/lib/python3.8/site-packages/pytorch_lightning/accelerators/accelerator.py", line 374, in setup_optimizers
optimizers, lr_schedulers, optimizer_frequencies = self.training_type_plugin.init_optimizers(
File ".local/lib/python3.8/site-packages/pytorch_lightning/plugins/training_type/training_type_plugin.py", line 190, in init_optimizers
return trainer.init_optimizers(model)
File ".local/lib/python3.8/site-packages/pytorch_lightning/trainer/optimizers.py", line 34, in init_optimizers
optim_conf = model.configure_optimizers()
File "/torchserve-example/model.py", line 52, in configure_optimizers
return torch.optim.AdamW(self.parameters())
File ".local/lib/python3.8/site-packages/torch/optim/adamw.py", line 47, in __init__
super(AdamW, self).__init__(params, defaults)
File ".local/lib/python3.8/site-packages/torch/optim/optimizer.py", line 55, in __init__
self.add_param_group(param_group)
File ".local/lib/python3.8/site-packages/torch/optim/optimizer.py", line 242, in add_param_group
assert isinstance(param_group, dict), "param group must be a dict"
AssertionError: param group must be a dict
When I execute print(type(params[0])) in the configure_optimizers, it prints <class 'torch.nn.parameter.Parameter'> to stdout. Any idea what went wrong here?
Note: As this error occurs during initialization of the optimizer, this is probably not directly related to pytorch lightning which is why I included pytorch as a tag as well.
In the library code, I found:
# if not isinstance(param_groups[0], dict):
# param_groups = [{'params': param_groups}]
When commenting this in, everything works normally.
I leave the question open because it does not feel like a good solution to change the underlying library or to just copy this code section to my file.
Actually, this line is wrong in your code:
def configure_optimizers(self):
params = [p for p in self.parameters() if p.requires_grad]
return torch.optim.AdamW(self.parameters())
You are passing params not self.parameters() as that would work fine.
With params created like this you are essentially passing list with generator inside, which is not an instance of dict.
In PyTorch it is possible to pass multiple different parameters with different learning rates etc. via dicts contained in lists. This is what your params looks like to PyTorch’s API.

Invalid placeholder in tensorflow

I am trying to write a custom loss function as follows.
def vgg16_feature_model(flayers, weights='imagenet'):
"""
Feature exctraction VGG16 model.
# Arguments
flayers: list of strings with names of layers to get the features for.
The length of `flayers` should be > 1, otherwise the output shape
is one axis less.
weights: ether "imagenet" or path to the file with weights.
# Returns
features_model: keras.models.Model instance to extract the features.
# Raises
AssertionError: in case of `flayers` is not a list.
AssertionError: in case of length of 'flayers' < 2.
"""
assert isinstance(flayers,list), "First argument 'flayers' must be a list"
assert len(flayers) > 1, "Length of 'flayers' must be > 1."
base_model = VGG16(include_top=False, weights=weights)
vgg16_outputs = [base_model.get_layer(flayers[i]).output for i in range(len(flayers))]
features_model = Model(inputs=[base_model.input], outputs=vgg16_outputs, name='vgg16_features')
features_model.trainable = False
features_model.compile(loss='mse', optimizer='adam')
return features_model
# Losses:
# -------
def total_loss(mask, vgg16_weights='imagenet'):
"""
Total loss defined in Eq 7 of Liu et al 2018 with:
y_true = I_gt,
y_pred = I_out,
y_comp = I_comp.
"""
vgg16_lnames = ['block1_pool', 'block2_pool', 'block3_pool']
vgg_model = vgg16_feature_model(vgg16_lnames, weights=vgg16_weights)
def loss(y_true, y_pred):
mask_inv = 1 - mask
y_comp = mask * y_true + mask_inv * y_pred
print("y_pred", y_pred)
print(y_comp)
input()
vgg_out = vgg_model(y_pred)
vgg_gt = vgg_model(y_true)
print("abc-----------------------------------")
vgg_comp = vgg_model(y_comp)
print("abc")
l_valid = loss_per_pixel(y_true, y_pred, mask)
l_hole = loss_per_pixel(y_true, y_pred, mask_inv)
l_perc = loss_perc(vgg_out, vgg_gt, vgg_comp)
l_style = loss_style(vgg_out, vgg_gt, vgg_comp)
l_tv = loss_tv(y_comp, mask_inv)
return l_valid + 6.*l_hole + 0.05*l_perc + 120.*l_style + 0.1*l_tv
return loss
I am getting an error as
Traceback (most recent call last):
File "inpainter_main.py", line 46, in <module>
model = pconv_model(lr=LR_STAGE1, image_size=IMAGE_SIZE, vgg16_weights=VGG16_WEIGHTS)
File "/home/bitsy-chuck/Downloads/PConv2D-2ndimp/inpainter_utils/pconv2d_model.py", line 118, in pconv_model
model.compile(Adam(lr=lr), loss=total_loss(mask_input, vgg16_weights=vgg16_weights))
File "/home/bitsy-chuck/anaconda3/lib/python3.7/site-packages/tensorflow/python/training/tracking/base.py", line 456, in _method_wrapper
result = method(self, *args, **kwargs)
File "/home/bitsy-chuck/anaconda3/lib/python3.7/site-packages/tensorflow/python/keras/engine/training_v1.py", line 446, in compile
self._compile_weights_loss_and_weighted_metrics()
File "/home/bitsy-chuck/anaconda3/lib/python3.7/site-packages/tensorflow/python/training/tracking/base.py", line 456, in _method_wrapper
result = method(self, *args, **kwargs)
File "/home/bitsy-chuck/anaconda3/lib/python3.7/site-packages/tensorflow/python/keras/engine/training_v1.py", line 1515, in _compile_weights_loss_and_weighted_metrics
self.total_loss = self._prepare_total_loss(masks)
File "/home/bitsy-chuck/anaconda3/lib/python3.7/site-packages/tensorflow/python/keras/engine/training_v1.py", line 1575, in _prepare_total_loss
per_sample_losses = loss_fn.call(y_true, y_pred)
File "/home/bitsy-chuck/anaconda3/lib/python3.7/site-packages/tensorflow/python/keras/losses.py", line 246, in call
return self.fn(y_true, y_pred, **self._fn_kwargs)
File "/home/bitsy-chuck/Downloads/PConv2D-2ndimp/inpainter_utils/pconv2d_loss.py", line 58, in loss
vgg_comp = vgg_model(y_comp)
File "/home/bitsy-chuck/anaconda3/lib/python3.7/site-packages/tensorflow/python/keras/engine/base_layer_v1.py", line 737, in __call__
base_layer_utils.create_keras_history(inputs)
File "/home/bitsy-chuck/anaconda3/lib/python3.7/site-packages/tensorflow/python/keras/engine/base_layer_utils.py", line 186, in create_keras_history
_, created_layers = _create_keras_history_helper(tensors, set(), [])
File "/home/bitsy-chuck/anaconda3/lib/python3.7/site-packages/tensorflow/python/keras/engine/base_layer_utils.py", line 249, in _create_keras_history_helper
layer_inputs, processed_ops, created_layers)
File "/home/bitsy-chuck/anaconda3/lib/python3.7/site-packages/tensorflow/python/keras/engine/base_layer_utils.py", line 246, in _create_keras_history_helper
constants[i] = backend.function([], op_input)([])
File "/home/bitsy-chuck/anaconda3/lib/python3.7/site-packages/tensorflow/python/keras/backend.py", line 3632, in __call__
run_metadata=self.run_metadata)
File "/home/bitsy-chuck/anaconda3/lib/python3.7/site-packages/tensorflow/python/client/session.py", line 1472, in __call__
run_metadata_ptr)
tensorflow.python.framework.errors_impl.InvalidArgumentError: You must feed a value for placeholder tensor 'pconv2d_dec_16_target' with dtype float and shape [?,?,?,?]
[[{{node pconv2d_dec_16_target}}]]
I first thought that y_comp is not correct, but
y_pred ---> Tensor("pconv2d_dec_16/BiasAdd:0", shape=(None, 512, 512, 3), dtype=float32)
y_comp ---> Tensor("loss_1/pconv2d_dec_16_loss/add:0", shape=(None, 512, 512, 3), dtype=float32)
They both appear the same to me and it should work, according to me.
error is at line vgg_comp = vgg_model(y_comp)
Can anyone also explain why am I getting an error of placeholder?
Tf version 1.3
keras 2.2.4
placeholder errors are usually due to tensorflow versions. I had the exact same error and it was fixed when I installed keras first and then tensorflow first. Using anaconda might help as they cache all the files when you uninstall so it is easy to install again without having to download the entire thing again.
There might be some other fix, I believe, but this fixed mine.

Chainer CNN- TypeError: forward() missing 1 required positional argument: 'x'

I'm trying to run a classifier on Chainer, but failed due to the following error.
I have no idea about the error, because I confirmed that the iterator actually sent a batch to the trainer.
Is there a problem with the neural network model? Or, the way the data has been fed into the model is wrong?
Input.py
from chainer.datasets import split_dataset_random
from chainer.iterators import SerialIterator
from chainer.optimizers import Adam
from chainer.training import Trainer
from chainer.training.updaters import StandardUpdater
from chainer import functions as F, links as L
from chainer import Sequential
import numpy as np
batch_size = 3
X_train = np.ones((9957, 60, 80, 3), dtype=np.float32)
X_train, _ = split_dataset_random(X_train, 8000, seed=0)
train_iter = SerialIterator(X_train, batch_size)
model = Sequential(
L.Convolution2D(None, 64, 3, 2),
F.relu,
L.Convolution2D(64, 32, 3, 2),
F.relu,
L.Linear(None, 16),
F.dropout,
L.Linear(16, 4)
)
model_loss = L.Classifier(model)
optimizer = Adam()
optimizer.setup(model_loss)
updater = StandardUpdater(train_iter, optimizer)
trainer = Trainer(updater, (25, 'epoch'))
trainer.run()
Stacktrace.py
Exception in main training loop: forward() missing 1 required positional argument: 'x'
Traceback (most recent call last):
File "/home/user/miniconda3/lib/python3.7/site-packages/chainer/training/trainer.py", line 315, in run
update()
File "/home/user/miniconda3/lib/python3.7/site-packages/chainer/training/updaters/standard_updater.py", line 165, in update
self.update_core()
File "/home/user/miniconda3/lib/python3.7/site-packages/chainer/training/updaters/standard_updater.py", line 181, in update_core
optimizer.update(loss_func, in_arrays)
File "/home/user/miniconda3/lib/python3.7/site-packages/chainer/optimizer.py", line 680, in update
loss = lossfun(*args, **kwds)
File "/home/user/miniconda3/lib/python3.7/site-packages/chainer/link.py", line 242, in __call__
out = forward(*args, **kwargs)
File "/home/user/miniconda3/lib/python3.7/site-packages/chainer/links/model/classifier.py", line 143, in forward
self.y = self.predictor(*args, **kwargs)
File "/home/user/miniconda3/lib/python3.7/site-packages/chainer/link.py", line 242, in __call__
out = forward(*args, **kwargs)
File "/home/user/miniconda3/lib/python3.7/site-packages/chainer/sequential.py", line 210, in forward
x = layer(*x)
File "/home/user/miniconda3/lib/python3.7/site-packages/chainer/link.py", line 242, in __call__
out = forward(*args, **kwargs)
Will finalize trainer extensions and updater before reraising the exception.
Traceback (most recent call last):
File "/home/user/deploy/aaa.py", line 33, in <module>
trainer.run()
File "/home/user/miniconda3/lib/python3.7/site-packages/chainer/training/trainer.py", line 348, in run
six.reraise(*exc_info)
File "/home/user/miniconda3/lib/python3.7/site-packages/six.py", line 693, in reraise
raise value
File "/home/user/miniconda3/lib/python3.7/site-packages/chainer/training/trainer.py", line 315, in run
update()
File "/home/user/miniconda3/lib/python3.7/site-packages/chainer/training/updaters/standard_updater.py", line 165, in update
self.update_core()
File "/home/user/miniconda3/lib/python3.7/site-packages/chainer/training/updaters/standard_updater.py", line 181, in update_core
optimizer.update(loss_func, in_arrays)
File "/home/user/miniconda3/lib/python3.7/site-packages/chainer/optimizer.py", line 680, in update
loss = lossfun(*args, **kwds)
File "/home/user/miniconda3/lib/python3.7/site-packages/chainer/link.py", line 242, in __call__
out = forward(*args, **kwargs)
File "/home/user/miniconda3/lib/python3.7/site-packages/chainer/links/model/classifier.py", line 143, in forward
self.y = self.predictor(*args, **kwargs)
File "/home/user/miniconda3/lib/python3.7/site-packages/chainer/link.py", line 242, in __call__
out = forward(*args, **kwargs)
File "/home/user/miniconda3/lib/python3.7/site-packages/chainer/sequential.py", line 210, in forward
x = layer(*x)
File "/home/user/miniconda3/lib/python3.7/site-packages/chainer/link.py", line 242, in __call__
out = forward(*args, **kwargs)
TypeError: forward() missing 1 required positional argument: 'x'
Is there a problem with the neural network model or the way the data has been fed into the model? Please let me know if you need to see the whole code
All you had to do is to give a tuple of ndarray and int to the model, because this is the specification of L.Classifier.
Is there a problem with the neural network model? Or, the way the data has been fed into the model is wrong?
Therefore, the absolute answer is "the way the data has been fed in to the model is wrong".
In the following code, I defined a class inheriting DatasetMixin to feed a tuple of ndarray and int. (This is a conventional way how Chainer goes)
It should be noted that the input argument of L.Convolution2D must be an ndarray whose shape is (batch, channel, width, height). So I transpose the array in the dataset.
Solution.py
from chainer.datasets import split_dataset_random
from chainer.iterators import SerialIterator
from chainer.optimizers import Adam
from chainer.training import Trainer
from chainer.training.updaters import StandardUpdater
from chainer import functions as F, links as L
from chainer import Sequential
from chainer.dataset import DatasetMixin
import numpy as np
class MyDataset(DatasetMixin):
def __init__(self, X, labels):
super(MyDataset, self).__init__()
self.X_ = X
self.labels_ = labels
self.size_ = X.shape[0]
def __len__(self):
return self.size_
def get_example(self, i):
return np.transpose(self.X_[i, ...], (2, 0, 1)), self.labels_[i]
batch_size = 3
X_train = np.ones((9957, 60, 80, 3), dtype=np.float32)
label_train = np.random.randint(0, 4, (9957,), dtype=np.int32)
dataset = MyDataset(X_train, label_train)
dataset_train, _ = split_dataset_random(dataset, 8000, seed=0)
train_iter = SerialIterator(dataset_train, batch_size)
model = Sequential(
L.Convolution2D(None, 64, 3, 2),
F.relu,
L.Convolution2D(64, 32, 3, 2),
F.relu,
L.Linear(None, 16),
F.dropout,
L.Linear(16, 4)
)
model_loss = L.Classifier(model)
optimizer = Adam()
optimizer.setup(model_loss)
updater = StandardUpdater(train_iter, optimizer)
trainer = Trainer(updater, (25, 'epoch'))
trainer.run()

In Tensorflow I can't use any MultiRNNCell instance in dynamic decode, but a single RNNCell instance can work on it

I make a seq2seq model using tensorflow and meet a problem that my program throws an error when I use MultiRNNCell in tf.contrib.seq2seq.dynamic_decode.
The problem happens over here:
defw_rnn=tf.nn.rnn_cell.MultiRNNCell([
tf.nn.rnn_cell.LSTMCell(num_units=self.FLAGS.rnn_units,
initializer=tf.orthogonal_initializer)
for _ in range(self.FLAGS.rnn_layer_size)])
training_helper = tf.contrib.seq2seq.TrainingHelper(inputs=decoder_inputs,
sequence_length=self.decoder_targets_length,
time_major=False)
training_decoder = \
tf.contrib.seq2seq.BasicDecoder(
defw_rnn, training_helper,
encoder_final_state,
output_layer)
training_decoder_output, _, training_decoder_output_length = \
tf.contrib.seq2seq.dynamic_decode(
training_decoder,
impute_finished=True,
maximum_iterations=self.FLAGS.max_len)
When I run this code,the console shows this Error message:
C:\Users\TopView\AppData\Local\Programs\Python\Python36\python.exe E:/PycharmProject/cikm_transport/CIKM/CIKM/translate_model/train.py
WARNING:tensorflow:From C:\Users\TopView\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\ops\rnn.py:417: calling reverse_sequence (from tensorflow.python.ops.array_ops) with seq_dim is deprecated and will be removed in a future version.
Instructions for updating:
seq_dim is deprecated, use seq_axis instead
WARNING:tensorflow:From C:\Users\TopView\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\util\deprecation.py:432: calling reverse_sequence (from tensorflow.python.ops.array_ops) with batch_dim is deprecated and will be removed in a future version.
Instructions for updating:
batch_dim is deprecated, use batch_axis instead
encoder_final_state shpe
LSTMStateTuple(c=<tf.Tensor 'encoder/bidirectional_rnn/fw/fw/while/Exit_5:0' shape=(?, 24) dtype=float32>, h=<tf.Tensor 'encoder/bidirectional_rnn/fw/fw/while/Exit_6:0' shape=(?, 24) dtype=float32>)
decoder_inputs shape before embedded
(128, 10)
decoder inputs shape after embedded
(128, 10, 5)
Traceback (most recent call last):
File "E:/PycharmProject/cikm_transport/CIKM/CIKM/translate_model/train.py", line 14, in <module>
len(embedding_matrix['embedding'][0]))
File "E:\PycharmProject\cikm_transport\CIKM\CIKM\translate_model\model.py", line 109, in __init__
maximum_iterations=self.FLAGS.max_len)
File "C:\Users\TopView\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\contrib\seq2seq\python\ops\decoder.py", line 323, in dynamic_decode
swap_memory=swap_memory)
File "C:\Users\TopView\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\ops\control_flow_ops.py", line 3209, in while_loop
result = loop_context.BuildLoop(cond, body, loop_vars, shape_invariants)
File "C:\Users\TopView\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\ops\control_flow_ops.py", line 2941, in BuildLoop
pred, body, original_loop_vars, loop_vars, shape_invariants)
File "C:\Users\TopView\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\ops\control_flow_ops.py", line 2878, in _BuildLoop
body_result = body(*packed_vars_for_body)
File "C:\Users\TopView\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\ops\control_flow_ops.py", line 3179, in <lambda>
body = lambda i, lv: (i + 1, orig_body(*lv))
File "C:\Users\TopView\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\contrib\seq2seq\python\ops\decoder.py", line 266, in body
decoder_finished) = decoder.step(time, inputs, state)
File "C:\Users\TopView\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\contrib\seq2seq\python\ops\basic_decoder.py", line 137, in step
cell_outputs, cell_state = self._cell(inputs, state)
File "C:\Users\TopView\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\ops\rnn_cell_impl.py", line 232, in __call__
return super(RNNCell, self).__call__(inputs, state)
File "C:\Users\TopView\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\layers\base.py", line 329, in __call__
outputs = super(Layer, self).__call__(inputs, *args, **kwargs)
File "C:\Users\TopView\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\keras\engine\base_layer.py", line 703, in __call__
outputs = self.call(inputs, *args, **kwargs)
File "C:\Users\TopView\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\ops\rnn_cell_impl.py", line 1325, in call
cur_inp, new_state = cell(cur_inp, cur_state)
File "C:\Users\TopView\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\ops\rnn_cell_impl.py", line 339, in __call__
*args, **kwargs)
File "C:\Users\TopView\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\layers\base.py", line 329, in __call__
outputs = super(Layer, self).__call__(inputs, *args, **kwargs)
File "C:\Users\TopView\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\keras\engine\base_layer.py", line 703, in __call__
outputs = self.call(inputs, *args, **kwargs)
File "C:\Users\TopView\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\ops\rnn_cell_impl.py", line 846, in call
(c_prev, m_prev) = state
File "C:\Users\TopView\AppData\Local\Programs\Python\Python36\lib\site-packages\tensorflow\python\framework\ops.py", line 436, in __iter__
"Tensor objects are not iterable when eager execution is not "
TypeError: Tensor objects are not iterable when eager execution is not enabled. To iterate over this tensor use tf.map_fn.
Process finished with exit code 1
But when I change the instance of defw_rnn, make it a single RNN instance like LSTMCell, the Error disappears:
defw_rnn=tf.nn.rnn_cell.LSTMCell(num_units=self.FLAGS.rnn_units,
initializer=tf.orthogonal_initializer)
And the code works well. However, I've found that most of the code about seq2seq model on the Internet using MultiRNNCell and they also use tensorflow, so it really confuse me that what is wrong with my program.
Here is the entire code:
import tensorflow as tf
import numpy as np
class Seq2SeqModel(object):
def bw_fw_rnn(self):
with tf.name_scope("forward_rnn"):
fw = tf.nn.rnn_cell.MultiRNNCell([
tf.nn.rnn_cell.LSTMCell(num_units=self.FLAGS.rnn_units,
initializer=tf.orthogonal_initializer) for _ in
range(self.FLAGS.rnn_layer_size)])
fw = tf.nn.rnn_cell.DropoutWrapper(fw, output_keep_prob=self.FLAGS.keep_prob)
with tf.name_scope("backward_rnn"):
bw = tf.nn.rnn_cell.MultiRNNCell([
tf.nn.rnn_cell.LSTMCell(num_units=self.FLAGS.rnn_units,
initializer=tf.orthogonal_initializer) for _ in
range(self.FLAGS.rnn_layer_size)])
bw = tf.nn.rnn_cell.DropoutWrapper(bw, output_keep_prob=self.FLAGS.keep_prob)
return (fw, bw)
def decode_inputs_preprocess(self, data, id_matrix):
ending=tf.strided_slice(data,[0,0],[self.batch_size,-1],[1,1])
decoder_input=tf.concat([tf.fill([self.batch_size,1],id_matrix.index('<go>')),ending],1)
return decoder_input
def __init__(self, FLAGS, english_id_matrix, spanish_id_matrix, english_vocab_size,spanish_vocab_size, embedding_size):
self.FLAGS = FLAGS
self.english_vocab_size = english_vocab_size
self.embedding_size = embedding_size
self.encoder_input = tf.placeholder(shape=[None, self.FLAGS.max_len], dtype=tf.int32, name='encoder_inputs')
self.decoder_targets = tf.placeholder(shape=[None, self.FLAGS.max_len], dtype=tf.int32, name='decoder_targets')
self.encoder_input_sequence_length = tf.placeholder(shape=[None], dtype=tf.int32, name='encoder_inputs_length')
self.decoder_targets_length = tf.placeholder(shape=[None], dtype=tf.int32, name='decoder_targets_length')
self.batch_size = self.FLAGS.batch_size
with tf.name_scope('embedding_look_up'):
spanish_embeddings = tf.Variable(
tf.random_uniform([english_vocab_size,
embedding_size], -1.0, 1.0),
dtype=tf.float32)
english_embeddings = tf.Variable(
tf.random_uniform([english_vocab_size,
embedding_size], -1.0, 1.0),
dtype=tf.float32)
self.spanish_embeddings_inputs = tf.placeholder(
dtype=tf.float32, shape=[english_vocab_size, embedding_size],
name='spanish_embeddings_inputs')
self.english_embeddings_inputs = tf.placeholder(
dtype=tf.float32, shape=[english_vocab_size, embedding_size],
name='spanish_embeddings_inputs')
self.spanish_embeddings_inputs_op = spanish_embeddings.assign(self.spanish_embeddings_inputs)
self.english_embeddings_inputs_op = english_embeddings.assign(self.english_embeddings_inputs)
encoder_inputs = tf.nn.embedding_lookup(spanish_embeddings, self.encoder_input)
with tf.name_scope('encoder'):
enfw_rnn, enbw_rnn = self.bw_fw_rnn()
encoder_outputs, encoder_final_state = \
tf.nn.bidirectional_dynamic_rnn(enfw_rnn, enbw_rnn, encoder_inputs
, sequence_length=self.encoder_input_sequence_length, dtype=tf.float32)
print("encoder_final_state shpe")
# final_state_c=tf.concat([encoder_final_state[0][-1].c,encoder_final_state[1][-1].c],1)
# final_state_h=tf.concat([encoder_final_state[0][-1].h,encoder_final_state[1][-1].h],1)
# encoder_final_state=tf.contrib.rnn.LSTMStateTuple(c=final_state_c,
# h=final_state_h)
encoder_final_state=encoder_final_state[0][-1]
print(encoder_final_state)
with tf.name_scope('dense_layer'):
output_layer = tf.layers.Dense(english_vocab_size,
kernel_initializer=tf.truncated_normal_initializer(
mean=0.0, stddev=0.1
))
# training decoder
with tf.name_scope('decoder'), tf.variable_scope('decode'):
decoder_inputs=self.decode_inputs_preprocess(self.decoder_targets,english_id_matrix)
print('decoder_inputs shape before embedded')
print(decoder_inputs.shape)
decoder_inputs = tf.nn.embedding_lookup(english_embeddings,decoder_inputs)
print('decoder inputs shape after embedded')
print(decoder_inputs.shape)
defw_rnn=tf.nn.rnn_cell.MultiRNNCell([
tf.nn.rnn_cell.LSTMCell(num_units=self.FLAGS.rnn_units,
initializer=tf.orthogonal_initializer)
for _ in range(self.FLAGS.rnn_layer_size)])
training_helper = tf.contrib.seq2seq.TrainingHelper(inputs=decoder_inputs,
sequence_length=self.decoder_targets_length,
time_major=False)
training_decoder = \
tf.contrib.seq2seq.BasicDecoder(
defw_rnn, training_helper,
encoder_final_state,
output_layer)
training_decoder_output, _, training_decoder_output_length = \
tf.contrib.seq2seq.dynamic_decode(
training_decoder,
impute_finished=True,
maximum_iterations=self.FLAGS.max_len)
training_logits = tf.identity(training_decoder_output.rnn_output, 'logits')
print("training logits shape")
print(training_logits.shape)
# predicting decoder
with tf.variable_scope('decode', reuse=True):
start_tokens = tf.tile(tf.constant([english_id_matrix.index('<go>')], dtype=tf.int32),
[self.batch_size], name='start_tokens')
predicting_helper = tf.contrib.seq2seq.GreedyEmbeddingHelper(english_embeddings,
start_tokens,
english_id_matrix.index('<eos>'))
predicting_decoder = tf.contrib.seq2seq.BasicDecoder(defw_rnn,
predicting_helper,
encoder_final_state,
output_layer)
predicting_decoder_output, _, predicting_decoder_output_length =\
tf.contrib.seq2seq.dynamic_decode(
predicting_decoder,
impute_finished=True,
maximum_iterations=self.FLAGS.max_len)
self.predicting_logits = tf.identity(predicting_decoder_output.sample_id, name='predictions')
print("predicting logits shape")
print(self.predicting_logits.shape)
masks = tf.sequence_mask(self.decoder_targets_length, self.FLAGS.max_len, dtype=tf.float32, name='masks')
with tf.variable_scope('optimization'), tf.name_scope('optimization'):
# Loss
self.cost = tf.contrib.seq2seq.sequence_loss(training_logits, self.decoder_targets, masks)
# Optimizer
optimizer = tf.train.AdamOptimizer(self.FLAGS.alpha)
# Gradient Clipping
gradients = optimizer.compute_gradients(self.cost)
capped_gradients = [(tf.clip_by_value(grad, -5., 5.), var) for grad, var in gradients if grad is not None]
self.train_op = optimizer.apply_gradients(capped_gradients)
Well……I've figured out.The problem happened because I only sent the final state of the encoder to a decoder.

Resources