Pretrained ENAS for 12layers

[MichailChatzianastasis]

Hey,
Is there a chance to upload the pretrained ENAS model, for the 12 layers?

Thanks in advance.

[muhanzhang]

Thanks for the question. Yes, I have updated the README to include the link to the 12-layer model: https://drive.google.com/drive/folders/18GU9g5DNiHn2MOVKOiF1fCwNQMTA-mnH?usp=sharing

[MichailChatzianastasis]

Thanks for your reply. I downloaded the pretrained 12-layer model , but i have the following error. As i observed, in some point , the program requires a tensor of shape (21,) which is the length of the representation of the 6-layer architecture, but now i pass a shape of (78,) which corresponds to the length of the representation of the 12-layer architecture. Is there any extra configuration, in order to make it accept the 12-layer representations?

Error:

Current idx is 0
Eval at 352352
2020-09-24 15:18:10.185416: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1618] Found device 0 with properties:
name: Tesla K80 major: 3 minor: 7 memoryClockRate(GHz): 0.8235
pciBusID: 067e:00:00.0
2020-09-24 15:18:10.186298: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1618] Found device 1 with properties:
name: Tesla K80 major: 3 minor: 7 memoryClockRate(GHz): 0.8235
pciBusID: e1e6:00:00.0
2020-09-24 15:18:10.186432: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.0
2020-09-24 15:18:10.186475: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcublas.so.10.0
2020-09-24 15:18:10.186526: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcufft.so.10.0
2020-09-24 15:18:10.186565: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcurand.so.10.0
2020-09-24 15:18:10.186609: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusolver.so.10.0
2020-09-24 15:18:10.186647: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcusparse.so.10.0
2020-09-24 15:18:10.186671: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7
2020-09-24 15:18:10.189854: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1746] Adding visible gpu devices: 0, 1
2020-09-24 15:18:10.189932: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1159] Device interconnect StreamExecutor with strength 1 edge matrix:
2020-09-24 15:18:10.189947: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1165] 0 1
2020-09-24 15:18:10.189958: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1178] 0: N N
2020-09-24 15:18:10.189967: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1178] 1: N N
2020-09-24 15:18:10.192358: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1304] Created TensorFlow device (/device:GPU:0 with 10671 MB memory) -> physical GPU (device: 0, name: Tesla K80, pci bus id: 067e:00:00.0, compute capability: 3.7)
2020-09-24 15:18:10.193297: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1304] Created TensorFlow device (/device:GPU:1 with 10671 MB memory) -> physical GPU (device: 1, name: Tesla K80, pci bus id: e1e6:00:00.0, compute capability: 3.7)
HERE /device:GPU:0
Traceback (most recent call last):
File "bo.py", line 267, in
score = -eva.eval(arc)
File "/home/mchatzi/my_projects/D-VAE/bayesian_optimization/../software/enas/src/cifar10/evaluation.py", line 262, in eval
return self.ops["eval_func"](self.sess, "valid", feed_dict={self.ops["controller"]["sample_arc3"]: np.asarray(arch_str)})
File "/home/mchatzi/my_projects/D-VAE/bayesian_optimization/../software/enas/src/cifar10/eval_child.py", line 717, in customized_eval_once
acc = self.eval_once(sess, eval_set, feed_dict, verbose)
File "/home/mchatzi/my_projects/D-VAE/bayesian_optimization/../software/enas/src/cifar10/models.py", line 181, in eval_once
acc = sess.run(acc_op, feed_dict=feed_dict)
File "/anaconda/envs/azureml_py36_automl/lib/python3.6/site-packages/tensorflow_core/python/training/monitored_session.py", line 754, in run
run_metadata=run_metadata)
File "/anaconda/envs/azureml_py36_automl/lib/python3.6/site-packages/tensorflow_core/python/training/monitored_session.py", line 1360, in run
raise six.reraise(*original_exc_info)
File "/anaconda/envs/azureml_py36_automl/lib/python3.6/site-packages/six.py", line 703, in reraise
raise value
File "/anaconda/envs/azureml_py36_automl/lib/python3.6/site-packages/tensorflow_core/python/training/monitored_session.py", line 1345, in run
return self._sess.run(*args, **kwargs)
File "/anaconda/envs/azureml_py36_automl/lib/python3.6/site-packages/tensorflow_core/python/training/monitored_session.py", line 1418, in run
run_metadata=run_metadata)
File "/anaconda/envs/azureml_py36_automl/lib/python3.6/site-packages/tensorflow_core/python/training/monitored_session.py", line 1176, in run
return self._sess.run(*args, **kwargs)
File "/anaconda/envs/azureml_py36_automl/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 956, in run
run_metadata_ptr)
File "/anaconda/envs/azureml_py36_automl/lib/python3.6/site-packages/tensorflow_core/python/client/session.py", line 1156, in _run
(np_val.shape, subfeed_t.name, str(subfeed_t.get_shape())))
ValueError: Cannot feed value of shape (78,) for Tensor 'Placeholder:0', which has shape '(21,)'

[muhanzhang]

Sorry. I forgot to mention there are some necessary changes to enable 12-layer ENAS evaluation.

First, change "outputs_6" to "outputs_12" in this line.

Second, change "child_num_layers" to 12 in this line.

Third, change "sample_structures6.txt" to "sample_structures12.txt" in this line.

Then it should work.