Adding a conda environment and migrating to tf 1.x

environment.yml

@@ -0,0 +1,74 @@
+name: emoji2vec
+channels:
+  - defaults
+dependencies:
+  - _libgcc_mutex=0.1=main
+  - _openmp_mutex=5.1=1_gnu
+  - _tflow_select=2.3.0=mkl
+  - absl-py=0.15.0=pyhd3eb1b0_0
+  - astor=0.8.1=py36h06a4308_0
+  - blas=1.0=mkl
+  - c-ares=1.18.1=h7f8727e_0
+  - ca-certificates=2022.07.19=h06a4308_0
+  - certifi=2021.5.30=py36h06a4308_0
+  - coverage=5.5=py36h27cfd23_2
+  - cython=0.29.24=py36h295c915_0
+  - dataclasses=0.8=pyh4f3eec9_6
+  - gast=0.5.3=pyhd3eb1b0_0
+  - google-pasta=0.2.0=pyhd3eb1b0_0
+  - grpcio=1.36.1=py36h2157cd5_1
+  - h5py=2.10.0=py36hd6299e0_1
+  - hdf5=1.10.6=hb1b8bf9_0
+  - importlib-metadata=4.8.1=py36h06a4308_0
+  - intel-openmp=2022.0.1=h06a4308_3633
+  - joblib=1.0.1=pyhd3eb1b0_0
+  - keras-applications=1.0.8=py_1
+  - keras-preprocessing=1.1.2=pyhd3eb1b0_0
+  - ld_impl_linux-64=2.38=h1181459_1
+  - libffi=3.3=he6710b0_2
+  - libgcc-ng=11.2.0=h1234567_1
+  - libgfortran-ng=7.5.0=ha8ba4b0_17
+  - libgfortran4=7.5.0=ha8ba4b0_17
+  - libgomp=11.2.0=h1234567_1
+  - libprotobuf=3.17.2=h4ff587b_1
+  - libstdcxx-ng=11.2.0=h1234567_1
+  - markdown=3.3.4=py36h06a4308_0
+  - mkl=2018.0.3=1
+  - mkl_fft=1.0.6=py36h7dd41cf_0
+  - mkl_random=1.0.1=py36h4414c95_1
+  - ncurses=6.3=h5eee18b_3
+  - numpy=1.15.4=py36h1d66e8a_0
+  - numpy-base=1.15.4=py36h81de0dd_0
+  - openssl=1.1.1q=h7f8727e_0
+  - pandas=1.1.5=py36h2531618_0
+  - pip=21.2.2=py36h06a4308_0
+  - protobuf=3.17.2=py36h295c915_0
+  - python=3.6.13=h12debd9_1
+  - python-dateutil=2.8.2=pyhd3eb1b0_0
+  - pytz=2021.3=pyhd3eb1b0_0
+  - readline=8.1.2=h7f8727e_1
+  - scikit-learn=0.23.2=py36h0573a6f_0
+  - scipy=0.19.1=py36h9976243_3
+  - setuptools=58.0.4=py36h06a4308_0
+  - six=1.16.0=pyhd3eb1b0_1
+  - sqlite=3.39.2=h5082296_0
+  - tbb=2021.5.0=hd09550d_0
+  - tbb4py=2021.3.0=py36hd09550d_0
+  - tensorboard=1.14.0=py36hf484d3e_0
+  - tensorflow=1.14.0=mkl_py36h2526735_0
+  - tensorflow-base=1.14.0=mkl_py36h7ce6ba3_0
+  - tensorflow-estimator=1.14.0=py_0
+  - termcolor=1.1.0=py36h06a4308_1
+  - threadpoolctl=2.2.0=pyh0d69192_0
+  - tk=8.6.12=h1ccaba5_0
+  - typing_extensions=4.1.1=pyh06a4308_0
+  - werkzeug=2.0.3=pyhd3eb1b0_0
+  - wheel=0.37.1=pyhd3eb1b0_0
+  - wrapt=1.12.1=py36h7b6447c_1
+  - xz=5.2.5=h7f8727e_1
+  - zipp=3.6.0=pyhd3eb1b0_0
+  - zlib=1.2.12=h7f8727e_2
+  - pip:
+    - gensim==0.13.4.1
+    - smart-open==6.0.0
+prefix: /home/lhk/anaconda3/envs/emoji2vec

model.py

@@ -119,13 +119,13 @@ def __init__(self, model_params, num_emoji, embeddings_array, use_embeddings=Tru
         v_col = tf.nn.dropout(v_col, (1 - model_params.dropout))
 
         # Calculate the predicted score, a.k.a. dot product (here)
-        self.score = tf.reduce_sum(tf.mul(v_row, v_col), 1)
+        self.score = tf.reduce_sum(tf.multiply(v_row, v_col), 1)
 
         # Probability of match
         self.prob = tf.sigmoid(self.score)
 
         # Calculate the cross-entropy loss
-        self.loss = tf.nn.sigmoid_cross_entropy_with_logits(self.score, self.y)
+        self.loss = tf.nn.sigmoid_cross_entropy_with_logits(logits=self.score, labels=self.y)
 
     # train the model using the appropriate parameters
     def train(self, kb, hooks, session):

naga/shared/trainer.py

@@ -20,7 +20,7 @@ def __call__(self, batcher, placeholders, loss, model=None, session=None):
             session = tf.Session()
             close_session_after_training = True  # no session existed before, we provide a temporary session
 
-        init = tf.initialize_all_variables()
+        init = tf.global_variables_initializer()
         session.run(init)
         epoch = 1
         iteration = 1

tfrnn/hooks.py

@@ -25,8 +25,8 @@ def __call__(self, sess, epoch, iteration, model, loss):
         raise NotImplementedError
 
     def update_summary(self, sess, current_step, title, value):
-        cur_summary = tf.scalar_summary(title, value)
-        merged_summary_op = tf.merge_summary([cur_summary])  # if you are using some summaries, merge them
+        cur_summary = tf.summary.scalar(title, value)
+        merged_summary_op = tf.summary.merge([cur_summary])  # if you are using some summaries, merge them
         summary_str = sess.run(merged_summary_op)
         self.summary_writer.add_summary(summary_str, current_step)
 

train.py

@@ -101,7 +101,7 @@ def train_save_evaluate(params, kb, train_set, dev_set, ind2emoji, embeddings_ar
 
         else:
             # For visualizing using tensorboard
-            summary_writer = tf.train.SummaryWriter(model_folder + '/board', graph=sess.graph)
+            summary_writer = tf.summary.FileWriter(model_folder + '/board', graph=sess.graph)
 
             # Keep track of how the model is training
             hooks = [