search

ikhlestov / rbm_based_autoencoders_with_tensorflow

MIT License
5 stars 4 forks source link

Running on Python 2.7 #1

Open dbl001 opened 7 years ago

dbl001 commented 7 years ago

I am trying to run this code on Python 2.7. Is this a big mistake?

Here's some issues I encountered:

  1. tf.mul() should now be tf.multiply()
  2. tf_sub() should now be tf.subtract()
  3. sigmoid_cross_entropy_with_logits has been renamed to: tf.nn.sigmoid_cross_entropy_with_logits,
  4. scalar_summary has been renamed to: tf.summary.scalar(),
  5. merge_all has been renamed to: tf.summary.merge_all

https://stackoverflow.com/questions/41066244/tensorflow-module-object-has-no-attribute-scalar-summary

How should I deal with os.makedirs(settings.STATIC_ROOT, exist_ok=True) in Python 2.7?

try: os.makedirs(settings.STATIC_ROOT) except OSError as e: if e.errno != errno.EEXIST: raise

ikhlestov commented 7 years ago

Hi! Yes, unfortunately code was created with python 3.4+ and tensorflow 0.10. First 5 issues points to tensorflow version, not python - at tensorflow 1.0 a lot of API were changed. I've not updated the code, because of not a lot of users. You may send a pull request if you want. Default tensorflow version mentioned in requirements.txt file.

Regarding os.makedirs(dir_name, exist_ok=True) - yes, such api was provided at Python 3 only. It can be replaced with few lines, as:

if not os.path.exists(dir_name):
    os.makedirs(dir_name)
dbl001 commented 7 years ago

Understood about tensorflow version.

I made the required changes and ran the script. Any ideas about this error?

===== Evaluate embeddings though SVM =====

Traceback (most recent call last): File "results_validation/visualize_distribution.py", line 34, in notes = [get_notes_from_filepath(path) for path in handled_pathes] File "results_validation/visualize_distribution.py", line 28, in get_notes_from_filepath log_name = [l for l in alllogs if l.startswith(idx + '')][0] IndexError: list index out of range

btw - tflearn should probably be added to requirments.txt

On Jun 28, 2017, at 1:33 AM, Illarion notifications@github.com wrote:

Hi! Yes, unfortunately code was created with python 3.4+ and tensorflow 0.10. First 5 issues points to tensorflow version, not python - at tensorflow 1.0 a lot of API were changed. I've not updated the code, because of not a lot of users. You may send a pull request if you want. Default tensorflow version mentioned in requirements.txt file.

Regarding os.makedirs(dir_name, exist_ok=True) - yes, such api was provided at Python 3 only. It can be replaced with few lines, as:

if not os.path.exists(dir_name): os.makedirs(dir_name) — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/ikhlestov/rbm_based_autoencoders_with_tensorflow/issues/1#issuecomment-311593844, or mute the thread https://github.com/notifications/unsubscribe-auth/AC9i27FwQiM_RpA_ZjEE7naJT3z6Zy41ks5sIg_MgaJpZM4OHGnt.

dbl001 commented 7 years ago

Does this look wrong?

…

0 [07:08<00:00, 20.63it/s] rbm: train_layers_by_pairs__last_layer_binarized Hamming: accuracy - 0, time consumption 141.291084051 seconds Dot product: accuracy - 0, time consumption 0 seconds

...

autoencoder: new_initialized_model__without_Gaussian_noise Hamming: accuracy - 0, time consumption 132.102096081 seconds Dot product: accuracy - 0, time consumption 0 seconds

Results were saved to /tmp/find_similar_accuracy_results.csv

On Jun 28, 2017, at 11:40 AM, David Laxer davidl@softintel.com wrote:

Understood about tensorflow version.

I made the required changes and ran the script. Any ideas about this error?

===== Evaluate embeddings though SVM =====

Traceback (most recent call last): File "results_validation/visualize_distribution.py", line 34, in notes = [get_notes_from_filepath(path) for path in handled_pathes] File "results_validation/visualize_distribution.py", line 28, in get_notes_from_filepath log_name = [l for l in alllogs if l.startswith(idx + '')][0] IndexError: list index out of range

btw - tflearn should probably be added to requirments.txt

On Jun 28, 2017, at 1:33 AM, Illarion <notifications@github.com mailto:notifications@github.com> wrote:

Hi! Yes, unfortunately code was created with python 3.4+ and tensorflow 0.10. First 5 issues points to tensorflow version, not python - at tensorflow 1.0 a lot of API were changed. I've not updated the code, because of not a lot of users. You may send a pull request if you want. Default tensorflow version mentioned in requirements.txt file.

Regarding os.makedirs(dir_name, exist_ok=True) - yes, such api was provided at Python 3 only. It can be replaced with few lines, as:

if not os.path.exists(dir_name): os.makedirs(dir_name) — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/ikhlestov/rbm_based_autoencoders_with_tensorflow/issues/1#issuecomment-311593844, or mute the thread https://github.com/notifications/unsubscribe-auth/AC9i27FwQiM_RpA_ZjEE7naJT3z6Zy41ks5sIg_MgaJpZM4OHGnt.

ikhlestov commented 7 years ago

Yes, this is incorrect results. Now I try make a version update at least for tensorflow 0.12 but I don't know how long may it take

dbl001 commented 7 years ago

Here’s the full output. Does anything else look wrong?

David-Laxers-MacBook-Pro:rbm_based_autoencoders_with_tensorflow davidlaxer$ ./run_all.sh

===== Train RBM ===== Train model by pair layers Only last layer is binarized Extracting /tmp/MNIST_data/train-images-idx3-ubyte.gz Extracting /tmp/MNIST_data/train-labels-idx1-ubyte.gz Extracting /tmp/MNIST_data/t10k-images-idx3-ubyte.gz Extracting /tmp/MNIST_data/t10k-labels-idx1-ubyte.gz Train layers pair 0 and 1 for 2 epochs Training model no: 0 Initialize new variables WARNING:tensorflow:From /Users/davidlaxer/anaconda/lib/python2.7/site-packages/tensorflow/python/util/tf_should_use.py:170: initialize_all_variables (from tensorflow.python.ops.variables) is deprecated and will be removed after 2017-03-02. Instructions for updating: Use tf.global_variables_initializer instead. Epoch: 0, time consumption: 0:00:34.575373 Epoch: 1, time consumption: 0:00:33.535722 Train layers pair 1 and 2 for 2 epochs Training model no: 0 Restore variables from previous run: bias_0 bias_1 W_0_1 Initialize not restored variables: W_1_2:0 bias_2:0 WARNING:tensorflow:From /Users/davidlaxer/anaconda/lib/python2.7/site-packages/tensorflow/python/util/tf_should_use.py:170: initialize_variables (from tensorflow.python.ops.variables) is deprecated and will be removed after 2017-03-02. Instructions for updating: Use tf.variables_initializer instead. Epoch: 0, time consumption: 0:00:22.907593 Epoch: 1, time consumption: 0:00:22.378566 Train layers pair 2 and 3 for 2 epochs Training model no: 0 Restore variables from previous run: bias_0 bias_1 bias_2 W_0_1 W_1_2 Initialize not restored variables: W_2_3:0 bias_3:0 WARNING:tensorflow:From /Users/davidlaxer/anaconda/lib/python2.7/site-packages/tensorflow/python/util/tf_should_use.py:170: initialize_variables (from tensorflow.python.ops.variables) is deprecated and will be removed after 2017-03-02. Instructions for updating: Use tf.variables_initializer instead. Epoch: 0, time consumption: 0:00:19.697597 Epoch: 1, time consumption: 0:00:20.018516 Get embeddings for test set Get embeddings for train set

===== Train AutoEncoder initialized from RBM ===== Extracting /tmp/MNIST_data/train-images-idx3-ubyte.gz Extracting /tmp/MNIST_data/train-labels-idx1-ubyte.gz Extracting /tmp/MNIST_data/t10k-images-idx3-ubyte.gz Extracting /tmp/MNIST_data/t10k-labels-idx1-ubyte.gz Such variables were created: W_0_1:0: (784, 484) bias_1:0: (484,) W_1_2:0: (484, 196) bias_2:0: (196,) W_2_3:0: (196, 100) bias_3:0: (100,) W_3_4:0: (196, 100) bias_4:0: (196,) W_4_5:0: (484, 196) bias_5:0: (484,) W_5_6:0: (784, 484) bias_6:0: (784,) Training model no: 0 !!!RBM model save file .ckpt with run_no: 0 not exists!!!

===== Train AutoEncoder initialized with new variables ===== Extracting /tmp/MNIST_data/train-images-idx3-ubyte.gz Extracting /tmp/MNIST_data/train-labels-idx1-ubyte.gz Extracting /tmp/MNIST_data/t10k-images-idx3-ubyte.gz Extracting /tmp/MNIST_data/t10k-labels-idx1-ubyte.gz Such variables were created: W_0_1:0: (784, 484) bias_1:0: (484,) W_1_2:0: (484, 196) bias_2:0: (196,) W_2_3:0: (196, 100) bias_3:0: (100,) W_3_4:0: (196, 100) bias_4:0: (196,) W_4_5:0: (484, 196) bias_5:0: (484,) W_5_6:0: (784, 484) bias_6:0: (784,) Training model no: 0 WARNING:tensorflow:From /Users/davidlaxer/anaconda/lib/python2.7/site-packages/tensorflow/python/util/tf_should_use.py:170: initialize_all_variables (from tensorflow.python.ops.variables) is deprecated and will be removed after 2017-03-02. Instructions for updating: Use tf.global_variables_initializer instead. Epoch: 0, time consumption: 0:00:38.965459 Epoch: 1, time consumption: 0:00:39.121245 Epoch: 2, time consumption: 0:00:40.023313 Epoch: 3, time consumption: 0:00:39.448177 Epoch: 4, time consumption: 0:00:38.664667 Epoch: 5, time consumption: 0:00:39.490704 Such variables were created: W_0_1:0: (784, 484) bias_1:0: (484,) W_1_2:0: (484, 196) bias_2:0: (196,) W_2_3:0: (196, 100) bias_3:0: (100,) W_3_4:0: (196, 100) bias_4:0: (196,) W_4_5:0: (484, 196) bias_5:0: (484,) W_5_6:0: (784, 484) bias_6:0: (784,) Get embeddings for test set Get embeddings for train set

===== Train AutoEncoder initialized from RBM without Gaussian noise===== Extracting /tmp/MNIST_data/train-images-idx3-ubyte.gz Extracting /tmp/MNIST_data/train-labels-idx1-ubyte.gz Extracting /tmp/MNIST_data/t10k-images-idx3-ubyte.gz Extracting /tmp/MNIST_data/t10k-labels-idx1-ubyte.gz Such variables were created: W_0_1:0: (784, 484) bias_1:0: (484,) W_1_2:0: (484, 196) bias_2:0: (196,) W_2_3:0: (196, 100) bias_3:0: (100,) W_3_4:0: (196, 100) bias_4:0: (196,) W_4_5:0: (484, 196) bias_5:0: (484,) W_5_6:0: (784, 484) bias_6:0: (784,) Training model no: 1 !!!RBM model save file .ckpt with run_no: 0 not exists!!!

===== Train AutoEncoder initialized with new variables without Gaussian noise ===== Extracting /tmp/MNIST_data/train-images-idx3-ubyte.gz Extracting /tmp/MNIST_data/train-labels-idx1-ubyte.gz Extracting /tmp/MNIST_data/t10k-images-idx3-ubyte.gz Extracting /tmp/MNIST_data/t10k-labels-idx1-ubyte.gz Such variables were created: W_0_1:0: (784, 484) bias_1:0: (484,) W_1_2:0: (484, 196) bias_2:0: (196,) W_2_3:0: (196, 100) bias_3:0: (100,) W_3_4:0: (196, 100) bias_4:0: (196,) W_4_5:0: (484, 196) bias_5:0: (484,) W_5_6:0: (784, 484) bias_6:0: (784,) Training model no: 1 WARNING:tensorflow:From /Users/davidlaxer/anaconda/lib/python2.7/site-packages/tensorflow/python/util/tf_should_use.py:170: initialize_all_variables (from tensorflow.python.ops.variables) is deprecated and will be removed after 2017-03-02. Instructions for updating: Use tf.global_variables_initializer instead. Epoch: 0, time consumption: 0:00:39.791340 Epoch: 1, time consumption: 0:00:41.863832 Epoch: 2, time consumption: 0:00:39.197791 Epoch: 3, time consumption: 0:00:38.405157 Epoch: 4, time consumption: 0:00:38.756205 Epoch: 5, time consumption: 0:00:39.403081 Such variables were created: W_0_1:0: (784, 484) bias_1:0: (484,) W_1_2:0: (484, 196) bias_2:0: (196,) W_2_3:0: (196, 100) bias_3:0: (100,) W_3_4:0: (196, 100) bias_4:0: (196,) W_4_5:0: (484, 196) bias_5:0: (484,) W_5_6:0: (784, 484) bias_6:0: (784,) Get embeddings for test set Get embeddings for train set

===== Display embeddings distribution from autoencoders(as background process) =====

===== Evaluate embeddings though SVM ===== Traceback (most recent call last): File "results_validation/visualize_distribution.py", line 34, in notes = [get_notes_from_filepath(path) for path in handled_pathes] File "results_validation/visualize_distribution.py", line 28, in get_notes_from_filepath log_name = [l for l in alllogs if l.startswith(idx + '')][0] IndexError: list index out of range Extracting /tmp/MNIST_data/train-images-idx3-ubyte.gz Extracting /tmp/MNIST_data/train-labels-idx1-ubyte.gz Extracting /tmp/MNIST_data/t10k-images-idx3-ubyte.gz Extracting /tmp/MNIST_data/t10k-labels-idx1-ubyte.gz Time cons: 206.56s, type: default mnist dataset Accuracy: 0.44660 precision recall f1-score support

      0       0.96      0.69      0.81       980
      1       0.62      0.01      0.03      1135
      2       0.73      0.30      0.42      1032
      3       0.78      0.30      0.43      1010
      4       0.78      0.41      0.54       982
      5       0.50      0.38      0.43       892
      6       0.76      0.75      0.76       958
      7       0.74      0.33      0.45      1028
      8       0.39      0.47      0.42       974
      9       0.19      0.91      0.32      1009

avg / total 0.65 0.45 0.45 10000

Time cons: 41.23s, type: rbm: train_layers_by_pairs__last_layer_binarized Accuracy: 0.42620 precision recall f1-score support

      0       0.08      0.06      0.07       980
      1       0.72      0.47      0.57      1135
      2       0.46      0.40      0.42      1032
      3       0.41      0.38      0.40      1010
      4       0.40      0.31      0.35       982
      5       0.28      0.32      0.30       892
      6       0.59      0.85      0.70       958
      7       0.72      0.63      0.67      1028
      8       0.33      0.33      0.33       974
      9       0.30      0.50      0.38      1009

avg / total 0.43 0.43 0.42 10000

Time cons: 42.56s, type: autoencoder: new_initialized_model__with_Gaussian_noise Accuracy: 0.09780 precision recall f1-score support

      0       0.10      0.15      0.12       980
      1       0.11      0.13      0.12      1135
      2       0.10      0.10      0.10      1032
      3       0.10      0.08      0.09      1010
      4       0.08      0.05      0.06       982
      5       0.08      0.09      0.08       892
      6       0.12      0.08      0.10       958
      7       0.09      0.08      0.08      1028
      8       0.11      0.10      0.10       974
      9       0.09      0.11      0.10      1009

avg / total 0.10 0.10 0.10 10000

Time cons: 45.38s, type: autoencoder: new_initialized_model__without_Gaussian_noise Accuracy: 0.08760 precision recall f1-score support

      0       0.00      0.00      0.00       980
      1       0.01      0.00      0.01      1135
      2       0.00      0.00      0.00      1032
      3       0.00      0.00      0.00      1010
      4       0.00      0.00      0.00       982
      5       0.06      0.29      0.10       892
      6       0.03      0.01      0.01       958
      7       0.00      0.00      0.00      1028
      8       0.12      0.62      0.21       974
      9       0.00      0.00      0.00      1009

avg / total 0.02 0.09 0.03 10000

Results were saved to /tmp/svm_clusterizationn_test_results.csv

===== Evaluate embeddings though most similar accuracy Extracting /tmp/MNIST_data/train-images-idx3-ubyte.gz Extracting /tmp/MNIST_data/train-labels-idx1-ubyte.gz Extracting /tmp/MNIST_data/t10k-images-idx3-ubyte.gz Extracting /tmp/MNIST_data/t10k-labels-idx1-ubyte.gz hamming_dist/default mnist dataset: 100%|█| 10000/10000 [16:57<00:00, 15.02it/s] dot_product/default mnist dataset: 100%|█| 10000/10000 [1:08:26<00:00, 5.00it/s] default mnist dataset Hamming: accuracy - 0, time consumption 1017.47319913 seconds Dot product: accuracy - 0, time consumption 0 seconds

hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 0%| | 0/10000 [hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 0%| | 10/10000 hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 0%| | 19/10000 hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 0%| | 30/10000 hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 0%| | 42/10000 hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 0%| | 50/10000 hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 1%| | 58/10000 hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 1%| | 65/10000 hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 1%| | 72/10000 hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 1%| | 79/10000 hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 1%| | 86/10000 hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 1%| | 92/10000 hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 1%| | 98/10000 hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 1%| | 106/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 1%| | 112/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 1%| | 118/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 1%| | 124/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 1%| | 130/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 1%| | 135/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 1%| | 140/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 1%| | 144/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 1%| | 148/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 2%| | 152/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 2%| | 159/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 2%| | 163/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 2%| | 167/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 2%| | 171/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 2%| | 176/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 2%| | 180/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 2%| | 185/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 2%| | 190/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 2%| | 197/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 2%| | 202/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 2%| | 209/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 2%| | 216/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 2%| | 222/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 2%| | 229/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 2%| | 235/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 2%| | 241/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 2%| | 246/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 3%| | 255/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 3%| | 264/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 3%| | 271/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 3%| | 278/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 3%| | 285/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 3%| | 292/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 3%| | 299/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 3%| | 305/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 3%| | 313/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 3%| | 320/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 3%| | 326/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 3%| | 334/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 3%| | 341/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 3%| | 348/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 4%| | 354/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 4%| | 362/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 4%| | 368/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 4%| | 373/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 4%| | 378/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 4%| | 382/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 4%| | 386/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 4%| | 393/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 4%| | 398/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 4%| | 404/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 4%| | 409/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 4%| | 419/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 4%| | 425/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 4%| | 436/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 4%| | 443/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 4%| | 450/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 5%| | 458/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 5%| | 466/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 5%| | 474/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 5%| | 481/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 5%| | 488/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 5%| | 499/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 5%| | 512/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 5%| | 526/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 5%| | 541/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 6%| | 554/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 6%| | 565/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 6%| | 579/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 6%| | 593/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 6%| | 605/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 6%| | 618/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 6%| | 630/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 6%| | 644/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 7%| | 657/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 7%| | 670/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 7%| | 683/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 7%| | 696/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 7%| | 707/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 7%| | 717/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 7%| | 726/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 7%| | 734/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 7%| | 742/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 8%| | 751/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 8%| | 758/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 8%| | 764/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 8%| | 770/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 8%| | 776/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 8%| | 782/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 8%| | 787/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 8%| | 794/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 8%| | 800/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 8%| | 806/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 8%| | 812/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 8%| | 817/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 8%| | 822/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 8%| | 828/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 8%| | 838/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 8%| | 846/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 9%| | 853/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 9%| | 859/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 9%| | 865/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 9%| | 872/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 9%| | 878/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 9%| | 888/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 9%| | 899/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 9%| | 911/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 9%| | 922/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 9%| | 934/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 9%| | 946/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 10%| | 958/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 10%| | 969/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 10%| | 982/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 10%| | 994/10000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 10%| | 1009/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 10%| | 1022/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 10%| | 1036/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 10%| | 1050/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 11%| | 1064/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 11%| | 1077/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 11%| | 1089/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 11%| | 1099/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 11%| | 1108/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 11%| | 1117/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 11%| | 1125/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 11%| | 1132/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 11%| | 1139/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 11%| | 1146/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 12%| | 1153/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 12%| | 1159/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 12%| | 1165/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 12%| | 1171/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 12%| | 1177/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 12%| | 1183/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 12%| | 1191/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 12%| | 1197/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 12%| | 1206/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 12%| | 1213/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 12%| | 1220/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 12%| | 1226/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 12%| | 1234/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 12%| | 1240/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 12%| | 1246/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 13%|▏| 1254/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 13%|▏| 1262/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 13%|▏| 1271/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 13%|▏| 1279/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 13%|▏| 1287/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 13%|▏| 1300/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 13%|▏| 1311/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 13%|▏| 1323/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 13%|▏| 1336/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 13%|▏| 1349/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 14%|▏| 1361/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 14%|▏| 1373/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 14%|▏| 1385/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 14%|▏| 1400/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 14%|▏| 1413/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 14%|▏| 1426/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 14%|▏| 1440/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 15%|▏| 1453/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 15%|▏| 1466/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 15%|▏| 1477/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 15%|▏| 1487/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 15%|▏| 1496/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 15%|▏| 1504/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 15%|▏| 1512/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 15%|▏| 1519/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 15%|▏| 1526/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 15%|▏| 1535/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 15%|▏| 1542/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 15%|▏| 1548/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 16%|▏| 1555/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 16%|▏| 1561/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 16%|▏| 1567/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 16%|▏| 1573/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 16%|▏| 1579/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 16%|▏| 1586/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 16%|▏| 1592/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 16%|▏| 1598/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 16%|▏| 1604/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 16%|▏| 1613/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 16%|▏| 1619/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 16%|▏| 1626/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 16%|▏| 1632/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 16%|▏| 1640/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 16%|▏| 1647/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 17%|▏| 1653/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 17%|▏| 1659/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 17%|▏| 1671/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 17%|▏| 1679/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 17%|▏| 1691/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 17%|▏| 1703/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 17%|▏| 1713/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 17%|▏| 1725/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 17%|▏| 1738/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 18%|▏| 1751/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 18%|▏| 1764/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 18%|▏| 1776/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 18%|▏| 1789/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 18%|▏| 1803/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 18%|▏| 1816/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 18%|▏| 1829/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 18%|▏| 1842/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 19%|▏| 1855/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 19%|▏| 1865/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 19%|▏| 1874/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 19%|▏| 1882/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 19%|▏| 1890/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 19%|▏| 1898/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 19%|▏| 1906/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 19%|▏| 1913/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 19%|▏| 1921/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 19%|▏| 1929/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 19%|▏| 1936/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 19%|▏| 1943/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 20%|▏| 1950/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 20%|▏| 1957/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 20%|▏| 1964/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 20%|▏| 1971/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 20%|▏| 1978/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 20%|▏| 1984/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 20%|▏| 1990/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 20%|▏| 1998/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 20%|▏| 2004/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 20%|▏| 2009/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 20%|▏| 2018/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 20%|▏| 2025/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 20%|▏| 2034/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 20%|▏| 2043/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 21%|▏| 2054/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 21%|▏| 2067/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 21%|▏| 2078/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 21%|▏| 2088/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 21%|▏| 2102/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 21%|▏| 2115/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 21%|▏| 2129/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 21%|▏| 2143/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 22%|▏| 2157/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 22%|▏| 2170/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 22%|▏| 2184/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 22%|▏| 2198/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 22%|▏| 2212/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 22%|▏| 2226/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 22%|▏| 2240/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 23%|▏| 2253/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 23%|▏| 2264/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 23%|▏| 2274/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 23%|▏| 2283/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 23%|▏| 2292/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 23%|▏| 2300/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 23%|▏| 2308/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 23%|▏| 2316/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 23%|▏| 2324/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 23%|▏| 2333/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 23%|▏| 2340/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 23%|▏| 2348/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 24%|▏| 2355/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 24%|▏| 2362/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 24%|▏| 2368/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 24%|▏| 2375/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 24%|▏| 2381/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 24%|▏| 2387/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 24%|▏| 2396/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 24%|▏| 2401/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 24%|▏| 2410/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 24%|▏| 2416/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 24%|▏| 2424/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 24%|▏| 2430/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 24%|▏| 2437/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 24%|▏| 2444/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 25%|▏| 2457/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 25%|▏| 2470/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 25%|▏| 2483/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 25%|▏| 2494/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 25%|▎| 2505/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 25%|▎| 2518/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 25%|▎| 2528/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 25%|▎| 2539/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 26%|▎| 2550/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 26%|▎| 2563/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 26%|▎| 2577/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 26%|▎| 2590/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 26%|▎| 2604/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 26%|▎| 2617/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 26%|▎| 2630/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 26%|▎| 2641/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 27%|▎| 2652/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 27%|▎| 2661/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 27%|▎| 2670/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 27%|▎| 2678/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 27%|▎| 2686/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 27%|▎| 2694/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 27%|▎| 2701/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 27%|▎| 2713/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 27%|▎| 2720/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 27%|▎| 2727/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 27%|▎| 2734/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 27%|▎| 2742/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 27%|▎| 2749/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 28%|▎| 2755/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 28%|▎| 2762/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 28%|▎| 2768/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 28%|▎| 2774/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 28%|▎| 2780/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 28%|▎| 2786/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 28%|▎| 2796/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 28%|▎| 2803/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 28%|▎| 2809/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 28%|▎| 2818/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 28%|▎| 2825/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 28%|▎| 2835/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 28%|▎| 2845/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 29%|▎| 2858/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 29%|▎| 2870/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 29%|▎| 2880/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 29%|▎| 2892/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 29%|▎| 2904/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 29%|▎| 2917/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 29%|▎| 2929/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 29%|▎| 2943/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 30%|▎| 2957/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 30%|▎| 2970/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 30%|▎| 2981/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 30%|▎| 2993/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 30%|▎| 3004/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 30%|▎| 3016/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 30%|▎| 3027/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 30%|▎| 3036/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 30%|▎| 3045/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 31%|▎| 3054/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 31%|▎| 3063/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 31%|▎| 3071/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 31%|▎| 3078/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 31%|▎| 3085/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 31%|▎| 3091/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 31%|▎| 3097/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 31%|▎| 3103/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 31%|▎| 3110/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 31%|▎| 3116/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 31%|▎| 3122/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 31%|▎| 3130/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 31%|▎| 3137/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 31%|▎| 3144/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 32%|▎| 3150/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 32%|▎| 3155/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 32%|▎| 3165/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 32%|▎| 3171/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 32%|▎| 3176/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 32%|▎| 3183/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 32%|▎| 3190/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 32%|▎| 3201/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 32%|▎| 3213/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 32%|▎| 3225/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 32%|▎| 3235/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 32%|▎| 3249/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 33%|▎| 3262/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 33%|▎| 3275/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 33%|▎| 3288/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 33%|▎| 3301/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 33%|▎| 3315/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 33%|▎| 3328/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 33%|▎| 3341/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 34%|▎| 3353/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 34%|▎| 3367/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 34%|▎| 3380/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 34%|▎| 3391/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 34%|▎| 3400/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 34%|▎| 3409/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 34%|▎| 3418/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 34%|▎| 3427/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 34%|▎| 3435/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 34%|▎| 3442/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 34%|▎| 3449/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 35%|▎| 3456/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 35%|▎| 3463/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 35%|▎| 3470/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 35%|▎| 3476/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 35%|▎| 3483/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 35%|▎| 3490/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 35%|▎| 3497/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 35%|▎| 3506/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 35%|▎| 3513/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 35%|▎| 3519/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 35%|▎| 3525/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 35%|▎| 3536/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 35%|▎| 3543/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 36%|▎| 3550/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 36%|▎| 3556/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 36%|▎| 3564/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 36%|▎| 3573/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 36%|▎| 3580/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 36%|▎| 3588/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 36%|▎| 3601/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 36%|▎| 3614/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 36%|▎| 3625/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 36%|▎| 3635/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 36%|▎| 3648/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 37%|▎| 3661/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 37%|▎| 3673/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 37%|▎| 3687/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 37%|▎| 3699/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 37%|▎| 3712/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 37%|▎| 3725/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 37%|▎| 3737/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 37%|▎| 3749/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 38%|▍| 3762/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 38%|▍| 3774/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 38%|▍| 3783/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 38%|▍| 3792/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 38%|▍| 3801/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 38%|▍| 3810/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 38%|▍| 3818/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 38%|▍| 3825/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 38%|▍| 3833/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 38%|▍| 3840/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 38%|▍| 3848/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 39%|▍| 3855/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 39%|▍| 3861/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 39%|▍| 3868/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 39%|▍| 3874/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 39%|▍| 3882/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 39%|▍| 3891/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 39%|▍| 3898/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 39%|▍| 3904/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 39%|▍| 3910/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 39%|▍| 3916/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 39%|▍| 3922/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 39%|▍| 3932/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 39%|▍| 3939/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 39%|▍| 3946/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 40%|▍| 3953/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 40%|▍| 3961/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 40%|▍| 3968/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 40%|▍| 3974/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 40%|▍| 3981/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 40%|▍| 3993/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 40%|▍| 4001/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 40%|▍| 4015/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 40%|▍| 4028/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 40%|▍| 4041/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 41%|▍| 4054/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 41%|▍| 4067/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 41%|▍| 4080/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 41%|▍| 4093/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 41%|▍| 4106/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 41%|▍| 4118/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 41%|▍| 4130/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 41%|▍| 4142/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 42%|▍| 4154/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 42%|▍| 4163/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 42%|▍| 4172/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 42%|▍| 4180/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 42%|▍| 4189/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 42%|▍| 4197/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 42%|▍| 4204/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 42%|▍| 4211/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 42%|▍| 4218/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 42%|▍| 4225/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 42%|▍| 4232/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 42%|▍| 4239/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 42%|▍| 4245/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 43%|▍| 4252/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 43%|▍| 4258/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 43%|▍| 4264/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 43%|▍| 4270/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 43%|▍| 4275/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 43%|▍| 4283/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 43%|▍| 4293/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 43%|▍| 4303/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 43%|▍| 4315/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 43%|▍| 4326/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 43%|▍| 4339/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 44%|▍| 4351/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 44%|▍| 4364/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 44%|▍| 4376/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 44%|▍| 4389/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 44%|▍| 4401/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 44%|▍| 4413/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 44%|▍| 4426/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 44%|▍| 4440/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 45%|▍| 4453/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 45%|▍| 4465/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 45%|▍| 4477/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 45%|▍| 4487/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 45%|▍| 4496/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 45%|▍| 4505/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 45%|▍| 4513/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 45%|▍| 4521/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 45%|▍| 4528/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 45%|▍| 4535/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 45%|▍| 4542/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 45%|▍| 4548/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 46%|▍| 4554/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 46%|▍| 4560/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 46%|▍| 4567/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 46%|▍| 4576/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 46%|▍| 4589/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 46%|▍| 4602/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 46%|▍| 4614/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 46%|▍| 4625/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 46%|▍| 4637/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 46%|▍| 4647/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 47%|▍| 4659/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 47%|▍| 4672/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 47%|▍| 4685/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 47%|▍| 4697/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 47%|▍| 4711/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 47%|▍| 4725/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 47%|▍| 4738/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 48%|▍| 4751/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 48%|▍| 4762/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 48%|▍| 4772/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 48%|▍| 4781/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 48%|▍| 4789/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 48%|▍| 4797/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 48%|▍| 4806/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 48%|▍| 4813/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 48%|▍| 4820/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 48%|▍| 4828/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 48%|▍| 4835/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 48%|▍| 4841/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 48%|▍| 4847/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 49%|▍| 4854/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 49%|▍| 4860/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 49%|▍| 4868/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 49%|▍| 4875/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 49%|▍| 4882/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 49%|▍| 4889/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 49%|▍| 4896/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 49%|▍| 4902/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 49%|▍| 4909/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 49%|▍| 4917/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 49%|▍| 4924/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 49%|▍| 4932/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 49%|▍| 4941/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 49%|▍| 4948/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 50%|▍| 4957/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 50%|▍| 4964/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 50%|▍| 4971/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 50%|▍| 4983/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 50%|▍| 4995/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 50%|▌| 5006/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 50%|▌| 5018/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 50%|▌| 5031/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 50%|▌| 5042/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 51%|▌| 5055/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 51%|▌| 5067/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 51%|▌| 5079/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 51%|▌| 5091/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 51%|▌| 5105/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 51%|▌| 5119/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 51%|▌| 5132/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 51%|▌| 5145/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 52%|▌| 5155/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 52%|▌| 5164/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 52%|▌| 5173/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 52%|▌| 5181/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 52%|▌| 5189/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 52%|▌| 5197/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 52%|▌| 5204/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 52%|▌| 5211/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 52%|▌| 5218/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 52%|▌| 5225/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 52%|▌| 5232/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 52%|▌| 5239/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 52%|▌| 5246/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 53%|▌| 5253/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 53%|▌| 5259/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 53%|▌| 5265/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 53%|▌| 5274/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 53%|▌| 5283/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 53%|▌| 5290/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 53%|▌| 5296/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 53%|▌| 5303/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 53%|▌| 5309/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 53%|▌| 5316/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 53%|▌| 5322/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 53%|▌| 5328/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 53%|▌| 5334/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 53%|▌| 5346/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 54%|▌| 5359/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 54%|▌| 5371/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 54%|▌| 5383/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 54%|▌| 5396/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 54%|▌| 5409/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 54%|▌| 5423/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 54%|▌| 5437/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 55%|▌| 5451/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 55%|▌| 5464/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 55%|▌| 5477/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 55%|▌| 5490/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 55%|▌| 5503/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 55%|▌| 5515/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 55%|▌| 5525/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 55%|▌| 5534/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 55%|▌| 5543/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 56%|▌| 5552/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 56%|▌| 5560/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 56%|▌| 5568/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 56%|▌| 5575/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 56%|▌| 5583/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 56%|▌| 5590/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 56%|▌| 5596/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 56%|▌| 5602/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 56%|▌| 5608/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 56%|▌| 5614/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 56%|▌| 5620/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 56%|▌| 5626/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 56%|▌| 5632/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 56%|▌| 5638/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 56%|▌| 5644/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 57%|▌| 5654/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 57%|▌| 5661/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 57%|▌| 5669/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 57%|▌| 5675/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 57%|▌| 5682/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 57%|▌| 5689/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 57%|▌| 5698/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 57%|▌| 5705/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 57%|▌| 5714/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 57%|▌| 5726/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 57%|▌| 5739/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 58%|▌| 5752/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 58%|▌| 5763/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 58%|▌| 5777/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 58%|▌| 5790/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 58%|▌| 5803/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 58%|▌| 5815/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 58%|▌| 5828/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 58%|▌| 5841/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 59%|▌| 5854/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 59%|▌| 5867/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 59%|▌| 5880/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 59%|▌| 5892/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 59%|▌| 5902/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 59%|▌| 5911/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 59%|▌| 5919/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 59%|▌| 5926/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 59%|▌| 5935/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 59%|▌| 5943/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 60%|▌| 5950/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 60%|▌| 5959/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 60%|▌| 5967/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 60%|▌| 5974/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 60%|▌| 5982/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 60%|▌| 5989/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 60%|▌| 5996/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 60%|▌| 6002/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 60%|▌| 6009/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 60%|▌| 6017/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 60%|▌| 6025/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 60%|▌| 6032/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 60%|▌| 6038/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 60%|▌| 6050/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 61%|▌| 6057/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 61%|▌| 6064/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 61%|▌| 6071/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 61%|▌| 6077/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 61%|▌| 6084/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 61%|▌| 6093/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 61%|▌| 6101/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 61%|▌| 6108/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 61%|▌| 6117/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 61%|▌| 6124/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 61%|▌| 6135/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 61%|▌| 6147/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 62%|▌| 6158/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 62%|▌| 6169/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 62%|▌| 6179/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 62%|▌| 6193/1000hamming_dist/rbm: train_layers_by_pairslast_layer_binarized: 62%|▌| 6204/1000hamming_dist/rbm: train_layers_by_pairslast_layer_b

ikhlestov commented 7 years ago

It seems that first RBM training pass. After autoencoders try to initialize from saved RBM, but tensorflow have another API/names for saves and two training of the autoencoders are failed. This caused by line #412 at the autoencoder.py

ikhlestov commented 7 years ago

And after all there are some troubles with names of stored tensor/variables

dbl001 commented 7 years ago

So, this error means the RBM training phase didn’t save the model properly?

!!!RBM model save file .ckpt with run_no: 0 not exists!!!

On Jun 28, 2017, at 1:01 PM, Illarion notifications@github.com wrote:

And after all there are some troubles with names of stored tensor/variables

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/ikhlestov/rbm_based_autoencoders_with_tensorflow/issues/1#issuecomment-311771409, or mute the thread https://github.com/notifications/unsubscribe-auth/AC9i23VeMHWBomLexmQEuhPc5jBEi7U4ks5sIrEygaJpZM4OHGnt.

ikhlestov commented 7 years ago

No, this means that RBM was trained/saved as expected. Now it saved as model.ckpt.index file. So you can just add 'index' at the end of this line. But later this lines in the autoencoder will fail, because tensorflow change once again naming convetions for saved layers inside checkpoint.

dbl001 commented 7 years ago

Is this the issue you alluded to with the contents/format of the checkpoint file?

https://github.com/tensorflow/tensorflow/issues/3967

On Jun 28, 2017, at 1:24 PM, Illarion notifications@github.com wrote:

No, this means that RBM was trained/saved as expected. Now it saved as model.ckpt.index file. So you can just add 'index' at the end of this line. But later this lines in the autoencoder will fail, because tensorflow change once again naming convetions for saved layers inside checkpoint.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.

dbl001 commented 7 years ago

And this one ...

https://github.com/tensorflow/tensorflow/blob/e0344f1e7ceff9c192369000d244c4ca47d7a1cc/tensorflow/python/tools/inspect_checkpoint.py

On Jun 28, 2017, at 1:24 PM, Illarion notifications@github.com wrote:

No, this means that RBM was trained/saved as expected. Now it saved as model.ckpt.index file. So you can just add 'index' at the end of this line. But later this lines in the autoencoder will fail, because tensorflow change once again naming convetions for saved layers inside checkpoint.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.

ikhlestov commented 7 years ago

Yes, the second link may help in the debugging

dbl001 commented 7 years ago

I installed a python 3.4 virtual environment, installed tdqm, installed tensorflow 1.1.0 and replaced a few old tf functions mul, sub, merge_summaries, ...

Does this look correct? screen shot 2017-07-09 at 11 08 47 am screen shot 2017-07-09 at 11 08 24 am

notes,accuracy,prec,f_score,recall default mnist dataset,0.4466,0.647146339198,0.460551796425,0.454023693039 rbm: train_layers_by_pairslast_layer_binarized,0.413,0.43410044028,0.413626653203,0.411131796451 autoencoder: new_initialized_modelwith_Gaussian_noise,0.0968,0.0957067341317,0.0791412829712,0.0962315867979 autoencoder: new_initialized_model__without_Gaussian_noise,0.1009,0.01009,0.0183304568989,0.1

autoencoder: new_initialized_model__without_Gaussian_noise Hamming: accuracy - 0.1012, time consumption 68.1845691204071 seconds Dot product: accuracy - 0.1012, time consumption 0.1012 seconds

Results were saved to /tmp/find_similar_accuracy_results.csv

ikhlestov commented 7 years ago

No, at least you should not have negative loss