Commit eda8ed1a authored by miaecle's avatar miaecle
Browse files

update sampels

parent 64d593ee

examples/delaney/delaney_tf_progressive.py

+1 −7
Original line number Diff line number Diff line
@@ -33,16 +33,10 @@ model = dc.models.ProgressiveMultitaskRegressor( 
    weight_init_stddevs=[.02] * n_layers,

    bias_init_consts=[1.] * n_layers,

    learning_rate=.001,

    penalty=.0001,

    penalty_type="l2",

    optimizer="adam",

    batch_size=100,

    seed=123,

    verbosity="high")

    batch_size=100)



# Fit trained model

model.fit(train_dataset)

model.save()



print("Evaluating model")

train_scores = model.evaluate(train_dataset, [metric], transformers)

examples/factors/FACTORS_tf_progressive.py

+13 −11
Original line number Diff line number Diff line
@@ -33,12 +33,15 @@ for trial in range(num_trials): 
  n_layers = 3

  nb_epoch = 50

  model = dc.models.ProgressiveMultitaskRegressor(

      len(FACTORS_tasks), train_dataset.get_data_shape()[0],

      layer_sizes=[750]*n_layers, dropouts=[.25]*n_layers,

      alpha_init_stddevs=[.02]*n_layers, weight_init_stddevs=[.02]*n_layers,

      bias_init_consts=[1.]*n_layers, learning_rate=.0003,

      penalty=.0001, penalty_type="l2", optimizer="adam", batch_size=100,

      logdir="FACTORS_tf_progressive")

      len(FACTORS_tasks),

      train_dataset.get_data_shape()[0],

      layer_sizes=[750] * n_layers,

      dropouts=[.25] * n_layers,

      alpha_init_stddevs=[.02] * n_layers,

      weight_init_stddevs=[.02] * n_layers,

      bias_init_consts=[1.] * n_layers,

      learning_rate=.0003,

      batch_size=100)



  #Use R2 classification metric

  metric = dc.metrics.Metric(dc.metrics.pearson_r2_score, task_averager=np.mean)
@@ -54,9 +57,8 @@ for trial in range(num_trials): 
  test_score, test_task_scores = model.evaluate(

      test_dataset, [metric], transformers, per_task_metrics=True)



  all_results.append((train_score, train_task_scores,

                      valid_score, valid_task_scores,

                      test_score, test_task_scores))

  all_results.append((train_score, train_task_scores, valid_score,

                      valid_task_scores, test_score, test_task_scores))



  print("Scores for trial %d" % trial)

  print("----------------------------------------------------------------")
@@ -76,8 +78,8 @@ for trial in range(num_trials): 
print("####################################################################")



for trial in range(num_trials):

  (train_score, train_task_scores, valid_score, valid_task_scores,

   test_score, test_task_scores) = all_results[trial]

  (train_score, train_task_scores, valid_score, valid_task_scores, test_score,

   test_task_scores) = all_results[trial]



  print("Scores for trial %d" % trial)

  print("----------------------------------------------------------------")

examples/hopv/hopv_tf_progressive.py

+0 −1
Original line number Diff line number Diff line
@@ -36,7 +36,6 @@ model = dc.models.ProgressiveMultitaskRegressor( 


# Fit trained model

model.fit(train_dataset, nb_epoch=25)

model.save()



print("Evaluating model")

train_scores = model.evaluate(train_dataset, metric, transformers)

examples/kaggle/KAGGLE_tf_progressive.py

+1 −5
Original line number Diff line number Diff line
@@ -41,11 +41,7 @@ for trial in range(num_trials): 
      weight_init_stddevs=[.02] * n_layers,

      bias_init_consts=[1.] * n_layers,

      learning_rate=.0003,

      penalty=.0001,

      penalty_type="l2",

      optimizer="adam",

      batch_size=100,

      logdir="KAGGLE_tf_progressive")

      batch_size=100)



  #Use R2 classification metric

  metric = dc.metrics.Metric(dc.metrics.pearson_r2_score, task_averager=np.mean)

examples/kinase/KINASE_tf_progressive.py

+13 −11
Original line number Diff line number Diff line
@@ -34,12 +34,15 @@ for trial in range(num_trials): 
  n_layers = 3

  nb_epoch = 50

  model = dc.models.ProgressiveMultitaskRegressor(

      len(KINASE_tasks), train_dataset.get_data_shape()[0],

      layer_sizes=[50]*n_layers, dropouts=[.25]*n_layers,

      alpha_init_stddevs=[.02]*n_layers, weight_init_stddevs=[.02]*n_layers,

      bias_init_consts=[1.]*n_layers, learning_rate=.0003,

      penalty=.0001, penalty_type="l2", optimizer="adam", batch_size=100,

      logdir="KINASE_tf_progressive")

      len(KINASE_tasks),

      train_dataset.get_data_shape()[0],

      layer_sizes=[50] * n_layers,

      dropouts=[.25] * n_layers,

      alpha_init_stddevs=[.02] * n_layers,

      weight_init_stddevs=[.02] * n_layers,

      bias_init_consts=[1.] * n_layers,

      learning_rate=.0003,

      batch_size=100)



  #Use R2 classification metric

  metric = dc.metrics.Metric(dc.metrics.pearson_r2_score, task_averager=np.mean)
@@ -55,9 +58,8 @@ for trial in range(num_trials): 
  test_score, test_task_scores = model.evaluate(

      test_dataset, [metric], transformers, per_task_metrics=True)



  all_results.append((train_score, train_task_scores,

                      valid_score, valid_task_scores,

                      test_score, test_task_scores))

  all_results.append((train_score, train_task_scores, valid_score,

                      valid_task_scores, test_score, test_task_scores))



  print("Scores for trial %d" % trial)

  print("----------------------------------------------------------------")
@@ -77,8 +79,8 @@ for trial in range(num_trials): 
print("####################################################################")



for trial in range(num_trials):

  (train_score, train_task_scores, valid_score, valid_task_scores,

   test_score, test_task_scores) = all_results[trial]

  (train_score, train_task_scores, valid_score, valid_task_scores, test_score,

   test_task_scores) = all_results[trial]



  print("Scores for trial %d" % trial)

  print("----------------------------------------------------------------")

CRA Git | Maintained and supported by SUSTech CRA and CCSE