Commit 1f7470b3 authored by joegomes's avatar joegomes
Browse files

Update QM7 example and add QM7b

parent cc222811

examples/qm7/qm7_datasets.py

+64 −34
Original line number Diff line number Diff line
@@ -12,55 +12,85 @@ import deepchem as dc 
import scipy.io

import csv



def load_qm7_from_mat(featurizer=None, split=0):

def load_qm7_from_mat(featurizer=None, split='stratified'):



  if not os.path.exists('qm7.mat'): os.system('wget http://www.quantum-machine.org/data/qm7.mat')

  dataset = scipy.io.loadmat('qm7.mat')

  

  P = dataset['P'][list(range(0,split))+list(range(split+1,5))].flatten()

  X = dataset['X'][P]

  y = dataset['T'][0,P]

  X = dataset['X']

  y = dataset['T']

  w = np.ones_like(y)

  train_dataset = dc.data.NumpyDataset(X, y, w, ids=None)

  dataset = dc.data.DiskDataset.from_numpy(X, y, w, ids=None) 

  print(len(dataset))

  

  Ptest = dataset['P'][split]

  X = dataset['X'][Ptest]

  y = dataset['T'][0,Ptest]

  w = np.ones_like(y)

  test_dataset = dc.data.NumpyDataset(X, y, w, ids=None)

  current_dir = os.path.dirname(os.path.realpath(__file__))

  split_file = os.path.join(

      current_dir, "./qm7_splits.csv")



  split_indices = []

  with open(split_file, 'r') as f:

    reader = csv.reader(f)

    for row in reader:

      row_int = (np.asarray(list(map(int, row)))).tolist()

      split_indices.append(row_int)

  

  splitters = {'index': dc.splits.IndexSplitter(),

               'random': dc.splits.RandomSplitter(),

               'indice': dc.splits.IndiceSplitter(valid_indices=split_indices[1]),

               'stratified': dc.splits.SingletaskStratifiedSplitter(task_number=0)}

  splitter = splitters[split]

  train_dataset, valid_dataset, test_dataset = splitter.train_valid_test_split(dataset)

  print(len(train_dataset))

  print(len(valid_dataset))

  print(len(test_dataset))



  transformers = [dc.trans.NormalizationTransformer(transform_y=True, dataset=train_dataset)]



  for transformer in transformers:

    train_dataset = transformer.transform(train_dataset)

    valid_dataset = transformer.transform(valid_dataset)

    test_dataset = transformer.transform(test_dataset)



  qm7_tasks = ["atomization_energy"]

  return qm7_tasks, (train_dataset, test_dataset), transformers

  qm7_tasks = np.arange(y.shape[1])

  return qm7_tasks, (train_dataset, valid_dataset, test_dataset), transformers



def load_qm7b_from_mat(featurizer=None, split='stratified'):



  if not os.path.exists('qm7b.mat'): os.system('wget http://www.quantum-machine.org/data/qm7b.mat')

  dataset_b = scipy.io.loadmat('qm7b.mat')

  dataset = scipy.io.loadmat('qm7b.mat')

  

  X = dataset_b['X']

  y = dataset_b['T']

  X = dataset['X']

  y = dataset['T']

  w = np.ones_like(y)

  dataset = dc.data.NumpyDataset(X, y, w, ids=None)

  dataset = dc.data.DiskDataset.from_numpy(X, y, w, ids=None)



  transformers = [dc.trans.NormalizationTransformer(transform_y=True, dataset=dataset)]

  current_dir = os.path.dirname(os.path.realpath(__file__))

  split_file = os.path.join(

      current_dir, "./qm7_splits.csv")



  for transformer in transformers:

    dataset = transformer.transform(dataset)

  split_indices = []

  with open(split_file, 'r') as f:

    reader = csv.reader(f)

    for row in reader:

      row_int = (np.asarray(list(map(int, row)))).tolist()

      split_indices.append(row_int)

  

  splitters = {'index': dc.splits.IndexSplitter(),

               'random': dc.splits.RandomSplitter(),

               'stratified': dc.splits.SingletaskStratifiedSplitter()}

               'indice': dc.splits.IndiceSplitter(valid_indices=split_indices[1]),

               'stratified': dc.splits.SingletaskStratifiedSplitter(task_number=0)}

  splitter = splitters[split]

  train_dataset, test_dataset = splitter.train_test_split(dataset)

  train_dataset, valid_dataset, test_dataset = splitter.train_valid_test_split(dataset)



  transformers = [dc.trans.NormalizationTransformer(transform_y=True, dataset=train_dataset)]



  for transformer in transformers:

    train_dataset = transformer.transform(train_dataset)

    valid_dataset = transformer.transform(valid_dataset)

    test_dataset = transformer.transform(test_dataset)



  qm7_tasks = np.arange(y.shape[1])

  return qm7_tasks, (train_dataset, test_dataset), transformers

  return qm7_tasks, (train_dataset, valid_dataset, test_dataset), transformers



def load_qm7(featurizer=None, split='random'):
@@ -77,14 +107,6 @@ def load_qm7(featurizer=None, split='random'): 
                             mol_field="mol", featurizer=featurizer)

  dataset = loader.featurize(dataset_file)

 

  # Initialize transformers 

  transformers = [

      dc.trans.NormalizationTransformer(transform_y=True, dataset=dataset)]



  print("About to transform data")

  for transformer in transformers:

    dataset = transformer.transform(dataset)

  

  split_file = os.path.join(

      current_dir, "./qm7_splits.csv")
@@ -95,10 +117,18 @@ def load_qm7(featurizer=None, split='random'): 
      row_int = (np.asarray(list(map(int, row)))).tolist()

      split_indices.append(row_int)

  

  

  splitters = {'index': dc.splits.IndexSplitter(),

               'random': dc.splits.RandomSplitter(),

               'indice': dc.splits.IndiceSplitter(valid_indices=split_indices[1])}

               'indice': dc.splits.IndiceSplitter(valid_indices=split_indices[1]),

               'stratified': dc.splits.SingletaskStratifiedSplitter(task_number=0)}

  splitter = splitters[split]

  train, valid, test = splitter.train_valid_test_split(dataset)

  return qm7_tasks, (train, valid, test), transformers
 
  train_dataset, valid_dataset, test_dataset = splitter.train_valid_test_split(dataset)



  transformers = [dc.trans.NormalizationTransformer(transform_y=True, dataset=train_dataset)]



  for transformer in transformers:

    train_dataset = transformer.transform(train_dataset)

    valid_dataset = transformer.transform(valid_dataset)

    test_dataset = transformer.transform(test_dataset)



  return qm7_tasks, (train_dataset, valid_dataset, test_dataset), transformers

examples/qm7/qm7_tf_model.py

+6 −6
Original line number Diff line number Diff line
@@ -11,12 +11,8 @@ import numpy as np 
from qm7_datasets import load_qm7_from_mat



np.random.seed(123)

split = 0

num_atoms = 23



qm7_tasks, datasets, transformers = load_qm7_from_mat(split)

train_dataset, test_dataset = datasets



qm7_tasks, datasets, transformers = load_qm7_from_mat(split='stratified')

train_dataset, valid_dataset, test_dataset = datasets

fit_transformers = [dc.trans.CoulombFitTransformer(train_dataset)]

regression_metric = [dc.metrics.Metric(dc.metrics.mean_absolute_error, mode="regression"), 

              dc.metrics.Metric(dc.metrics.pearson_r2_score, mode="regression")]
@@ -34,6 +30,10 @@ train_scores = model.evaluate(train_dataset, regression_metric, transformers) 
print("Train scores [kcal/mol]")

print(train_scores)



valid_scores = model.evaluate(valid_dataset, regression_metric, transformers)

print("Valid scores [kcal/mol]")

print(valid_scores)



test_scores = model.evaluate(test_dataset, regression_metric, transformers)

print("Test scores [kcal/mol]")

print(test_scores)

examples/qm7/qm7b_tf_model.py

0 → 100644
+39 −0
Original line number Diff line number Diff line 
"""

Script that trains Tensorflow singletask models on QM7 dataset.

"""

from __future__ import print_function

from __future__ import division

from __future__ import unicode_literals



import os

import deepchem as dc

import numpy as np

from qm7_datasets import load_qm7b_from_mat



np.random.seed(123)

qm7_tasks, datasets, transformers = load_qm7b_from_mat(split='stratified')

train_dataset, valid_dataset, test_dataset = datasets

fit_transformers = [dc.trans.CoulombFitTransformer(train_dataset)]

regression_metric = [dc.metrics.Metric(dc.metrics.mean_absolute_error, mode="regression"), 

              dc.metrics.Metric(dc.metrics.pearson_r2_score, mode="regression")]

model = dc.models.TensorflowMultiTaskFitTransformRegressor(

    n_tasks=len(qm7_tasks), n_features=[23, 23], learning_rate=0.001 , momentum=.8, batch_size=25,

    weight_init_stddevs=[1/np.sqrt(400),1/np.sqrt(100),1/np.sqrt(100)],

    bias_init_consts=[0.,0.,0.], layer_sizes=[400,100,100], 

    dropouts=[0.01,0.01,0.01], fit_transformers=fit_transformers, n_evals=10, seed=123)



# Fit trained model

model.fit(train_dataset, nb_epoch=50)

model.save()



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

print("Train scores [kcal/mol]")

print(train_scores)



valid_scores = model.evaluate(valid_dataset, regression_metric, transformers)

print("Valid scores [kcal/mol]")

print(valid_scores)



test_scores = model.evaluate(test_dataset, regression_metric, transformers)

print("Test scores [kcal/mol]")

print(test_scores)

CRA Git | Maintained and supported by SUSTech CRA and CCSE