path
stringlengths 13
17
| screenshot_names
sequencelengths 1
873
| code
stringlengths 0
40.4k
| cell_type
stringclasses 1
value |
|---|---|---|---|
88102789/cell_11 | [
"text_plain_output_1.png"
] | import seaborn as sns
df = sns.load_dataset('titanic')
df.head() | code |
88102789/cell_19 | [
"text_plain_output_1.png",
"image_output_1.png"
] | import seaborn as sns
df = sns.load_dataset('titanic')
sns.countplot(data=df, x='alone', hue='alive', palette='deep') | code |
88102789/cell_7 | [
"text_plain_output_1.png",
"image_output_1.png"
] | import seaborn as sns
df = sns.load_dataset('titanic')
sns.distplot(df['fare'], kde=False, bins=30, color='y') | code |
88102789/cell_18 | [
"text_html_output_1.png"
] | import seaborn as sns
df = sns.load_dataset('titanic')
sns.countplot(data=df, x='alive', hue='alone', palette='deep') | code |
88102789/cell_8 | [
"text_plain_output_1.png",
"image_output_1.png"
] | import seaborn as sns
df = sns.load_dataset('titanic')
sns.kdeplot(df['fare'], shade=True, color='y') | code |
88102789/cell_15 | [
"text_plain_output_1.png",
"image_output_1.png"
] | import seaborn as sns
df = sns.load_dataset('titanic')
df['who'].unique() | code |
88102789/cell_16 | [
"text_plain_output_1.png",
"image_output_1.png"
] | import seaborn as sns
df = sns.load_dataset('titanic')
sns.countplot(data=df, x='alive', hue='who', palette='deep') | code |
88102789/cell_3 | [
"text_plain_output_1.png",
"image_output_1.png"
] | import seaborn as sns
df = sns.load_dataset('titanic')
df.head() | code |
88102789/cell_17 | [
"text_plain_output_1.png",
"image_output_1.png"
] | import seaborn as sns
df = sns.load_dataset('titanic')
sns.countplot(data=df, x='who', hue='alive', palette='deep') | code |
88102789/cell_24 | [
"text_plain_output_1.png",
"image_output_1.png"
] | import seaborn as sns
df = sns.load_dataset('titanic')
sns.countplot(data=df, x='alone', hue='sex') | code |
88102789/cell_14 | [
"text_plain_output_2.png",
"application_vnd.jupyter.stderr_output_1.png",
"image_output_1.png"
] | import seaborn as sns
df = sns.load_dataset('titanic')
sns.countplot(data=df, x='alive', hue='sex', palette='deep') | code |
88102789/cell_22 | [
"text_plain_output_1.png"
] | import seaborn as sns
df = sns.load_dataset('titanic')
sns.countplot(data=df, x='embark_town', hue='class', palette='deep') | code |
88102789/cell_10 | [
"text_html_output_1.png"
] | import seaborn as sns
df = sns.load_dataset('titanic')
sns.jointplot(data=df[df['fare'] <= 100], x='age', y='fare', kind='scatter', color='c') | code |
88102789/cell_12 | [
"text_plain_output_2.png",
"application_vnd.jupyter.stderr_output_1.png",
"image_output_1.png"
] | import seaborn as sns
df = sns.load_dataset('titanic')
sns.countplot(data=df, x='survived', palette='dark') | code |
88102789/cell_5 | [
"text_plain_output_1.png",
"image_output_1.png"
] | import seaborn as sns
df = sns.load_dataset('titanic')
sns.distplot(df['age'], kde=False, bins=30, color='m') | code |
104117096/cell_13 | [
"text_html_output_1.png"
] | import glob
import numpy as np
import pandas as pd
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import os
import glob
glob.glob('../input/skin-cancer-mnist-ham10000/*')
images_path = glob.glob('../input/skin-cancer-mnist-ham10000/HAM10000_images_part_1/*')
img_id = []
for i in images_path:
img_id.append(i.split('/')[4].split('.')[0])
data = pd.read_csv('../input/skin-cancer-mnist-ham10000/HAM10000_metadata.csv')
full_name_cancer_dict = {'akiec': "Bowen's disease", 'bcc': 'basal cell carcinoma', 'bkl': 'benign keratosis-like lesions', 'df': 'dermatofibroma', 'mel': 'melanoma', 'nv': 'melanocytic nevi', 'vasc': 'vascular lesions'}
img_with_class = data.loc[:, ['image_id', 'dx']].to_dict('list')
img_with_class.keys()
img_to_class = {img_id: full_name_cancer_dict[disease] for img_id, disease in zip(img_with_class['image_id'], img_with_class['dx'])}
random = np.random.choice(list(img_to_class.keys()), 10)
{i: img_to_class[i] for i in random}
inter_data_id = set(img_to_class.keys()).intersection(set(img_id))
len(inter_data) | code |
104117096/cell_4 | [
"text_plain_output_1.png"
] | import glob
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import os
import glob
glob.glob('../input/skin-cancer-mnist-ham10000/*')
images_path = glob.glob('../input/skin-cancer-mnist-ham10000/HAM10000_images_part_1/*')
img_id = []
for i in images_path:
img_id.append(i.split('/')[4].split('.')[0])
img_id[:10] | code |
104117096/cell_6 | [
"text_plain_output_1.png"
] | import pandas as pd
data = pd.read_csv('../input/skin-cancer-mnist-ham10000/HAM10000_metadata.csv')
data.info() | code |
104117096/cell_2 | [
"text_plain_output_1.png"
] | import glob
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import os
import glob
glob.glob('../input/skin-cancer-mnist-ham10000/*') | code |
104117096/cell_7 | [
"image_output_1.png"
] | import glob
import pandas as pd
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import os
import glob
glob.glob('../input/skin-cancer-mnist-ham10000/*')
images_path = glob.glob('../input/skin-cancer-mnist-ham10000/HAM10000_images_part_1/*')
img_id = []
for i in images_path:
img_id.append(i.split('/')[4].split('.')[0])
data = pd.read_csv('../input/skin-cancer-mnist-ham10000/HAM10000_metadata.csv')
len(set(data['image_id']).intersection(set(img_id))) | code |
104117096/cell_15 | [
"text_plain_output_1.png"
] | import glob
import numpy as np
import pandas as pd
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import os
import glob
glob.glob('../input/skin-cancer-mnist-ham10000/*')
images_path = glob.glob('../input/skin-cancer-mnist-ham10000/HAM10000_images_part_1/*')
img_id = []
for i in images_path:
img_id.append(i.split('/')[4].split('.')[0])
data = pd.read_csv('../input/skin-cancer-mnist-ham10000/HAM10000_metadata.csv')
full_name_cancer_dict = {'akiec': "Bowen's disease", 'bcc': 'basal cell carcinoma', 'bkl': 'benign keratosis-like lesions', 'df': 'dermatofibroma', 'mel': 'melanoma', 'nv': 'melanocytic nevi', 'vasc': 'vascular lesions'}
img_with_class = data.loc[:, ['image_id', 'dx']].to_dict('list')
img_with_class.keys()
img_to_class = {img_id: full_name_cancer_dict[disease] for img_id, disease in zip(img_with_class['image_id'], img_with_class['dx'])}
random = np.random.choice(list(img_to_class.keys()), 10)
{i: img_to_class[i] for i in random}
inter_data_id = set(img_to_class.keys()).intersection(set(img_id))
len(inter_data)
def compelete_path(id_images, path):
full_paths = [path + i + '.jpg' for i in id_images]
return full_paths
full_paths = compelete_path(inter_data_id, '../input/skin-cancer-mnist-ham10000/HAM10000_images_part_1/')
full_paths[:5] | code |
104117096/cell_16 | [
"text_plain_output_1.png"
] | import glob
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import os
import glob
glob.glob('../input/skin-cancer-mnist-ham10000/*')
images_path = glob.glob('../input/skin-cancer-mnist-ham10000/HAM10000_images_part_1/*')
img_id = []
for i in images_path:
img_id.append(i.split('/')[4].split('.')[0])
data = pd.read_csv('../input/skin-cancer-mnist-ham10000/HAM10000_metadata.csv')
full_name_cancer_dict = {'akiec': "Bowen's disease", 'bcc': 'basal cell carcinoma', 'bkl': 'benign keratosis-like lesions', 'df': 'dermatofibroma', 'mel': 'melanoma', 'nv': 'melanocytic nevi', 'vasc': 'vascular lesions'}
img_with_class = data.loc[:, ['image_id', 'dx']].to_dict('list')
img_with_class.keys()
img_to_class = {img_id: full_name_cancer_dict[disease] for img_id, disease in zip(img_with_class['image_id'], img_with_class['dx'])}
random = np.random.choice(list(img_to_class.keys()), 10)
{i: img_to_class[i] for i in random}
inter_data_id = set(img_to_class.keys()).intersection(set(img_id))
len(inter_data)
def compelete_path(id_images, path):
full_paths = [path + i + '.jpg' for i in id_images]
return full_paths
full_paths = compelete_path(inter_data_id, '../input/skin-cancer-mnist-ham10000/HAM10000_images_part_1/')
full_paths[:5]
plt.figure(figsize=(10, 8))
for i in range(25):
plt.subplot(5, 5, i + 1)
plt.imshow(plt.imread(full_paths[i]))
plt.axis('off') | code |
104117096/cell_10 | [
"text_plain_output_1.png"
] | import pandas as pd
data = pd.read_csv('../input/skin-cancer-mnist-ham10000/HAM10000_metadata.csv')
img_with_class = data.loc[:, ['image_id', 'dx']].to_dict('list')
img_with_class.keys() | code |
104117096/cell_12 | [
"text_plain_output_1.png"
] | import glob
import numpy as np
import pandas as pd
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import os
import glob
glob.glob('../input/skin-cancer-mnist-ham10000/*')
images_path = glob.glob('../input/skin-cancer-mnist-ham10000/HAM10000_images_part_1/*')
img_id = []
for i in images_path:
img_id.append(i.split('/')[4].split('.')[0])
data = pd.read_csv('../input/skin-cancer-mnist-ham10000/HAM10000_metadata.csv')
full_name_cancer_dict = {'akiec': "Bowen's disease", 'bcc': 'basal cell carcinoma', 'bkl': 'benign keratosis-like lesions', 'df': 'dermatofibroma', 'mel': 'melanoma', 'nv': 'melanocytic nevi', 'vasc': 'vascular lesions'}
img_with_class = data.loc[:, ['image_id', 'dx']].to_dict('list')
img_with_class.keys()
img_to_class = {img_id: full_name_cancer_dict[disease] for img_id, disease in zip(img_with_class['image_id'], img_with_class['dx'])}
random = np.random.choice(list(img_to_class.keys()), 10)
{i: img_to_class[i] for i in random} | code |
104117096/cell_5 | [
"text_plain_output_1.png"
] | import pandas as pd
data = pd.read_csv('../input/skin-cancer-mnist-ham10000/HAM10000_metadata.csv')
data.head() | code |
17118100/cell_21 | [
"application_vnd.jupyter.stderr_output_2.png",
"text_plain_output_1.png"
] | from google.cloud import bigquery
import os
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import time
import numpy as np
import pandas as pd
import os
PROJECT_ID = 'YOUR_OWN_PROJECT_ID'
from google.cloud import bigquery
client = bigquery.Client(project=PROJECT_ID, location='US')
path = '../input/'
dataset_id = 'titanic'
dataset_ref = client.dataset(dataset_id)
job_config = bigquery.LoadJobConfig()
job_config.source_format = bigquery.SourceFormat.CSV
job_config.skip_leading_rows = 1
job_config.autodetect = True
existing_datasets = [dataset.dataset_id for dataset in client.list_datasets()]
if not dataset_id in existing_datasets:
client.create_dataset(dataset_id)
dataset_ref = client.dataset(dataset_id)
existing_tables = [table.table_id for table in client.list_tables(dataset_id)]
for filename in os.listdir(path):
table_id = filename.rstrip('.csv')
if not table_id in existing_tables:
table_ref = dataset_ref.table(table_id)
with open(path + filename, 'rb') as source_file:
job = client.load_table_from_file(source_file, table_ref, job_config=job_config)
job.result()
table_ref = dataset_ref.table('train')
train = client.get_table(table_ref)
train.schema
client.list_rows(train, max_results=5).to_dataframe()
train_query = "\nCREATE OR REPLACE MODEL `titanic.logistic_reg_model`\nOPTIONS (\n model_type = 'logistic_reg'\n ) AS (\nSELECT\n Pclass, Name, Sex, Age, SibSp, Parch, Ticket, Fare, Cabin, Embarked, Survived AS label\nFROM\n `titanic.train`)\n "
train_query_job = client.query(train_query)
while 1:
if client.get_job(train_query_job.job_id).state == 'DONE':
break
time.sleep(5)
predict_query = '\nSELECT\n * \nFROM\n ML.PREDICT(MODEL `titanic.logistic_reg_model`, (\nSELECT\n Pclass, Name, Sex, Age, SibSp, Parch, Ticket, Fare, Cabin, Embarked, PassengerId\n\nFROM\n `titanic.test`\n))\n '
predict_query_job = client.query(predict_query)
predicted = predict_query_job.to_dataframe()
sub = pd.read_csv('../input/test.csv', header=0)
sub = pd.merge(sub, predicted, on='PassengerId', how='right')
sub = sub[['PassengerId', 'predicted_label']].rename(index=str, columns={'predicted_label': 'Survived'})
sub.head() | code |
17118100/cell_13 | [
"application_vnd.jupyter.stderr_output_2.png",
"text_plain_output_1.png"
] | from google.cloud import bigquery
import os
import time
import numpy as np
import pandas as pd
import os
PROJECT_ID = 'YOUR_OWN_PROJECT_ID'
from google.cloud import bigquery
client = bigquery.Client(project=PROJECT_ID, location='US')
path = '../input/'
dataset_id = 'titanic'
dataset_ref = client.dataset(dataset_id)
job_config = bigquery.LoadJobConfig()
job_config.source_format = bigquery.SourceFormat.CSV
job_config.skip_leading_rows = 1
job_config.autodetect = True
existing_datasets = [dataset.dataset_id for dataset in client.list_datasets()]
if not dataset_id in existing_datasets:
client.create_dataset(dataset_id)
dataset_ref = client.dataset(dataset_id)
existing_tables = [table.table_id for table in client.list_tables(dataset_id)]
for filename in os.listdir(path):
table_id = filename.rstrip('.csv')
if not table_id in existing_tables:
table_ref = dataset_ref.table(table_id)
with open(path + filename, 'rb') as source_file:
job = client.load_table_from_file(source_file, table_ref, job_config=job_config)
job.result()
table_ref = dataset_ref.table('train')
train = client.get_table(table_ref)
train.schema
client.list_rows(train, max_results=5).to_dataframe()
train_query = "\nCREATE OR REPLACE MODEL `titanic.logistic_reg_model`\nOPTIONS (\n model_type = 'logistic_reg'\n ) AS (\nSELECT\n Pclass, Name, Sex, Age, SibSp, Parch, Ticket, Fare, Cabin, Embarked, Survived AS label\nFROM\n `titanic.train`)\n "
train_query_job = client.query(train_query)
while 1:
if client.get_job(train_query_job.job_id).state == 'DONE':
break
time.sleep(5) | code |
17118100/cell_9 | [
"text_html_output_1.png"
] | from google.cloud import bigquery
import os
import numpy as np
import pandas as pd
import os
PROJECT_ID = 'YOUR_OWN_PROJECT_ID'
from google.cloud import bigquery
client = bigquery.Client(project=PROJECT_ID, location='US')
path = '../input/'
dataset_id = 'titanic'
dataset_ref = client.dataset(dataset_id)
job_config = bigquery.LoadJobConfig()
job_config.source_format = bigquery.SourceFormat.CSV
job_config.skip_leading_rows = 1
job_config.autodetect = True
existing_datasets = [dataset.dataset_id for dataset in client.list_datasets()]
if not dataset_id in existing_datasets:
client.create_dataset(dataset_id)
dataset_ref = client.dataset(dataset_id)
existing_tables = [table.table_id for table in client.list_tables(dataset_id)]
for filename in os.listdir(path):
table_id = filename.rstrip('.csv')
if not table_id in existing_tables:
table_ref = dataset_ref.table(table_id)
with open(path + filename, 'rb') as source_file:
job = client.load_table_from_file(source_file, table_ref, job_config=job_config)
job.result()
table_ref = dataset_ref.table('train')
train = client.get_table(table_ref)
train.schema | code |
17118100/cell_20 | [
"application_vnd.jupyter.stderr_output_1.png"
] | from google.cloud import bigquery
import os
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import time
import numpy as np
import pandas as pd
import os
PROJECT_ID = 'YOUR_OWN_PROJECT_ID'
from google.cloud import bigquery
client = bigquery.Client(project=PROJECT_ID, location='US')
path = '../input/'
dataset_id = 'titanic'
dataset_ref = client.dataset(dataset_id)
job_config = bigquery.LoadJobConfig()
job_config.source_format = bigquery.SourceFormat.CSV
job_config.skip_leading_rows = 1
job_config.autodetect = True
existing_datasets = [dataset.dataset_id for dataset in client.list_datasets()]
if not dataset_id in existing_datasets:
client.create_dataset(dataset_id)
dataset_ref = client.dataset(dataset_id)
existing_tables = [table.table_id for table in client.list_tables(dataset_id)]
for filename in os.listdir(path):
table_id = filename.rstrip('.csv')
if not table_id in existing_tables:
table_ref = dataset_ref.table(table_id)
with open(path + filename, 'rb') as source_file:
job = client.load_table_from_file(source_file, table_ref, job_config=job_config)
job.result()
table_ref = dataset_ref.table('train')
train = client.get_table(table_ref)
train.schema
client.list_rows(train, max_results=5).to_dataframe()
train_query = "\nCREATE OR REPLACE MODEL `titanic.logistic_reg_model`\nOPTIONS (\n model_type = 'logistic_reg'\n ) AS (\nSELECT\n Pclass, Name, Sex, Age, SibSp, Parch, Ticket, Fare, Cabin, Embarked, Survived AS label\nFROM\n `titanic.train`)\n "
train_query_job = client.query(train_query)
while 1:
if client.get_job(train_query_job.job_id).state == 'DONE':
break
time.sleep(5)
predict_query = '\nSELECT\n * \nFROM\n ML.PREDICT(MODEL `titanic.logistic_reg_model`, (\nSELECT\n Pclass, Name, Sex, Age, SibSp, Parch, Ticket, Fare, Cabin, Embarked, PassengerId\n\nFROM\n `titanic.test`\n))\n '
predict_query_job = client.query(predict_query)
predicted = predict_query_job.to_dataframe()
sub = pd.read_csv('../input/test.csv', header=0)
sub = pd.merge(sub, predicted, on='PassengerId', how='right')
sub = sub[['PassengerId', 'predicted_label']].rename(index=str, columns={'predicted_label': 'Survived'}) | code |
17118100/cell_19 | [
"application_vnd.jupyter.stderr_output_2.png",
"text_plain_output_1.png"
] | from google.cloud import bigquery
import os
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
import time
import numpy as np
import pandas as pd
import os
PROJECT_ID = 'YOUR_OWN_PROJECT_ID'
from google.cloud import bigquery
client = bigquery.Client(project=PROJECT_ID, location='US')
path = '../input/'
dataset_id = 'titanic'
dataset_ref = client.dataset(dataset_id)
job_config = bigquery.LoadJobConfig()
job_config.source_format = bigquery.SourceFormat.CSV
job_config.skip_leading_rows = 1
job_config.autodetect = True
existing_datasets = [dataset.dataset_id for dataset in client.list_datasets()]
if not dataset_id in existing_datasets:
client.create_dataset(dataset_id)
dataset_ref = client.dataset(dataset_id)
existing_tables = [table.table_id for table in client.list_tables(dataset_id)]
for filename in os.listdir(path):
table_id = filename.rstrip('.csv')
if not table_id in existing_tables:
table_ref = dataset_ref.table(table_id)
with open(path + filename, 'rb') as source_file:
job = client.load_table_from_file(source_file, table_ref, job_config=job_config)
job.result()
table_ref = dataset_ref.table('train')
train = client.get_table(table_ref)
train.schema
client.list_rows(train, max_results=5).to_dataframe()
train_query = "\nCREATE OR REPLACE MODEL `titanic.logistic_reg_model`\nOPTIONS (\n model_type = 'logistic_reg'\n ) AS (\nSELECT\n Pclass, Name, Sex, Age, SibSp, Parch, Ticket, Fare, Cabin, Embarked, Survived AS label\nFROM\n `titanic.train`)\n "
train_query_job = client.query(train_query)
while 1:
if client.get_job(train_query_job.job_id).state == 'DONE':
break
time.sleep(5)
predict_query = '\nSELECT\n * \nFROM\n ML.PREDICT(MODEL `titanic.logistic_reg_model`, (\nSELECT\n Pclass, Name, Sex, Age, SibSp, Parch, Ticket, Fare, Cabin, Embarked, PassengerId\n\nFROM\n `titanic.test`\n))\n '
predict_query_job = client.query(predict_query)
predicted = predict_query_job.to_dataframe()
sub = pd.read_csv('../input/test.csv', header=0)
sub = pd.merge(sub, predicted, on='PassengerId', how='right') | code |
17118100/cell_1 | [
"text_plain_output_1.png"
] | import os
import numpy as np
import pandas as pd
import os
print(os.listdir('../input')) | code |
17118100/cell_7 | [
"application_vnd.jupyter.stderr_output_1.png"
] | from google.cloud import bigquery
import os
import numpy as np
import pandas as pd
import os
PROJECT_ID = 'YOUR_OWN_PROJECT_ID'
from google.cloud import bigquery
client = bigquery.Client(project=PROJECT_ID, location='US')
path = '../input/'
dataset_id = 'titanic'
dataset_ref = client.dataset(dataset_id)
job_config = bigquery.LoadJobConfig()
job_config.source_format = bigquery.SourceFormat.CSV
job_config.skip_leading_rows = 1
job_config.autodetect = True
existing_datasets = [dataset.dataset_id for dataset in client.list_datasets()]
if not dataset_id in existing_datasets:
client.create_dataset(dataset_id)
dataset_ref = client.dataset(dataset_id)
existing_tables = [table.table_id for table in client.list_tables(dataset_id)]
for filename in os.listdir(path):
print(path + filename)
table_id = filename.rstrip('.csv')
if not table_id in existing_tables:
table_ref = dataset_ref.table(table_id)
with open(path + filename, 'rb') as source_file:
job = client.load_table_from_file(source_file, table_ref, job_config=job_config)
job.result() | code |
17118100/cell_8 | [
"application_vnd.jupyter.stderr_output_1.png"
] | from google.cloud import bigquery
import os
import numpy as np
import pandas as pd
import os
PROJECT_ID = 'YOUR_OWN_PROJECT_ID'
from google.cloud import bigquery
client = bigquery.Client(project=PROJECT_ID, location='US')
path = '../input/'
dataset_id = 'titanic'
dataset_ref = client.dataset(dataset_id)
job_config = bigquery.LoadJobConfig()
job_config.source_format = bigquery.SourceFormat.CSV
job_config.skip_leading_rows = 1
job_config.autodetect = True
existing_datasets = [dataset.dataset_id for dataset in client.list_datasets()]
if not dataset_id in existing_datasets:
client.create_dataset(dataset_id)
dataset_ref = client.dataset(dataset_id)
existing_tables = [table.table_id for table in client.list_tables(dataset_id)]
for filename in os.listdir(path):
table_id = filename.rstrip('.csv')
if not table_id in existing_tables:
table_ref = dataset_ref.table(table_id)
with open(path + filename, 'rb') as source_file:
job = client.load_table_from_file(source_file, table_ref, job_config=job_config)
job.result()
table_ref = dataset_ref.table('train')
train = client.get_table(table_ref) | code |
17118100/cell_15 | [
"application_vnd.jupyter.stderr_output_2.png",
"text_plain_output_1.png"
] | from google.cloud import bigquery
import os
import time
import numpy as np
import pandas as pd
import os
PROJECT_ID = 'YOUR_OWN_PROJECT_ID'
from google.cloud import bigquery
client = bigquery.Client(project=PROJECT_ID, location='US')
path = '../input/'
dataset_id = 'titanic'
dataset_ref = client.dataset(dataset_id)
job_config = bigquery.LoadJobConfig()
job_config.source_format = bigquery.SourceFormat.CSV
job_config.skip_leading_rows = 1
job_config.autodetect = True
existing_datasets = [dataset.dataset_id for dataset in client.list_datasets()]
if not dataset_id in existing_datasets:
client.create_dataset(dataset_id)
dataset_ref = client.dataset(dataset_id)
existing_tables = [table.table_id for table in client.list_tables(dataset_id)]
for filename in os.listdir(path):
table_id = filename.rstrip('.csv')
if not table_id in existing_tables:
table_ref = dataset_ref.table(table_id)
with open(path + filename, 'rb') as source_file:
job = client.load_table_from_file(source_file, table_ref, job_config=job_config)
job.result()
table_ref = dataset_ref.table('train')
train = client.get_table(table_ref)
train.schema
client.list_rows(train, max_results=5).to_dataframe()
train_query = "\nCREATE OR REPLACE MODEL `titanic.logistic_reg_model`\nOPTIONS (\n model_type = 'logistic_reg'\n ) AS (\nSELECT\n Pclass, Name, Sex, Age, SibSp, Parch, Ticket, Fare, Cabin, Embarked, Survived AS label\nFROM\n `titanic.train`)\n "
train_query_job = client.query(train_query)
while 1:
if client.get_job(train_query_job.job_id).state == 'DONE':
break
time.sleep(5)
predict_query = '\nSELECT\n * \nFROM\n ML.PREDICT(MODEL `titanic.logistic_reg_model`, (\nSELECT\n Pclass, Name, Sex, Age, SibSp, Parch, Ticket, Fare, Cabin, Embarked, PassengerId\n\nFROM\n `titanic.test`\n))\n '
predict_query_job = client.query(predict_query) | code |
17118100/cell_16 | [
"application_vnd.jupyter.stderr_output_1.png"
] | from google.cloud import bigquery
import os
import time
import numpy as np
import pandas as pd
import os
PROJECT_ID = 'YOUR_OWN_PROJECT_ID'
from google.cloud import bigquery
client = bigquery.Client(project=PROJECT_ID, location='US')
path = '../input/'
dataset_id = 'titanic'
dataset_ref = client.dataset(dataset_id)
job_config = bigquery.LoadJobConfig()
job_config.source_format = bigquery.SourceFormat.CSV
job_config.skip_leading_rows = 1
job_config.autodetect = True
existing_datasets = [dataset.dataset_id for dataset in client.list_datasets()]
if not dataset_id in existing_datasets:
client.create_dataset(dataset_id)
dataset_ref = client.dataset(dataset_id)
existing_tables = [table.table_id for table in client.list_tables(dataset_id)]
for filename in os.listdir(path):
table_id = filename.rstrip('.csv')
if not table_id in existing_tables:
table_ref = dataset_ref.table(table_id)
with open(path + filename, 'rb') as source_file:
job = client.load_table_from_file(source_file, table_ref, job_config=job_config)
job.result()
table_ref = dataset_ref.table('train')
train = client.get_table(table_ref)
train.schema
client.list_rows(train, max_results=5).to_dataframe()
train_query = "\nCREATE OR REPLACE MODEL `titanic.logistic_reg_model`\nOPTIONS (\n model_type = 'logistic_reg'\n ) AS (\nSELECT\n Pclass, Name, Sex, Age, SibSp, Parch, Ticket, Fare, Cabin, Embarked, Survived AS label\nFROM\n `titanic.train`)\n "
train_query_job = client.query(train_query)
while 1:
if client.get_job(train_query_job.job_id).state == 'DONE':
break
time.sleep(5)
predict_query = '\nSELECT\n * \nFROM\n ML.PREDICT(MODEL `titanic.logistic_reg_model`, (\nSELECT\n Pclass, Name, Sex, Age, SibSp, Parch, Ticket, Fare, Cabin, Embarked, PassengerId\n\nFROM\n `titanic.test`\n))\n '
predict_query_job = client.query(predict_query)
predicted = predict_query_job.to_dataframe() | code |
17118100/cell_3 | [
"application_vnd.jupyter.stderr_output_1.png"
] | from google.cloud import bigquery
PROJECT_ID = 'YOUR_OWN_PROJECT_ID'
from google.cloud import bigquery
client = bigquery.Client(project=PROJECT_ID, location='US') | code |
17118100/cell_17 | [
"application_vnd.jupyter.stderr_output_1.png"
] | from google.cloud import bigquery
import os
import time
import numpy as np
import pandas as pd
import os
PROJECT_ID = 'YOUR_OWN_PROJECT_ID'
from google.cloud import bigquery
client = bigquery.Client(project=PROJECT_ID, location='US')
path = '../input/'
dataset_id = 'titanic'
dataset_ref = client.dataset(dataset_id)
job_config = bigquery.LoadJobConfig()
job_config.source_format = bigquery.SourceFormat.CSV
job_config.skip_leading_rows = 1
job_config.autodetect = True
existing_datasets = [dataset.dataset_id for dataset in client.list_datasets()]
if not dataset_id in existing_datasets:
client.create_dataset(dataset_id)
dataset_ref = client.dataset(dataset_id)
existing_tables = [table.table_id for table in client.list_tables(dataset_id)]
for filename in os.listdir(path):
table_id = filename.rstrip('.csv')
if not table_id in existing_tables:
table_ref = dataset_ref.table(table_id)
with open(path + filename, 'rb') as source_file:
job = client.load_table_from_file(source_file, table_ref, job_config=job_config)
job.result()
table_ref = dataset_ref.table('train')
train = client.get_table(table_ref)
train.schema
client.list_rows(train, max_results=5).to_dataframe()
train_query = "\nCREATE OR REPLACE MODEL `titanic.logistic_reg_model`\nOPTIONS (\n model_type = 'logistic_reg'\n ) AS (\nSELECT\n Pclass, Name, Sex, Age, SibSp, Parch, Ticket, Fare, Cabin, Embarked, Survived AS label\nFROM\n `titanic.train`)\n "
train_query_job = client.query(train_query)
while 1:
if client.get_job(train_query_job.job_id).state == 'DONE':
break
time.sleep(5)
predict_query = '\nSELECT\n * \nFROM\n ML.PREDICT(MODEL `titanic.logistic_reg_model`, (\nSELECT\n Pclass, Name, Sex, Age, SibSp, Parch, Ticket, Fare, Cabin, Embarked, PassengerId\n\nFROM\n `titanic.test`\n))\n '
predict_query_job = client.query(predict_query)
predicted = predict_query_job.to_dataframe()
predicted.head() | code |
17118100/cell_10 | [
"text_plain_output_1.png"
] | from google.cloud import bigquery
import os
import numpy as np
import pandas as pd
import os
PROJECT_ID = 'YOUR_OWN_PROJECT_ID'
from google.cloud import bigquery
client = bigquery.Client(project=PROJECT_ID, location='US')
path = '../input/'
dataset_id = 'titanic'
dataset_ref = client.dataset(dataset_id)
job_config = bigquery.LoadJobConfig()
job_config.source_format = bigquery.SourceFormat.CSV
job_config.skip_leading_rows = 1
job_config.autodetect = True
existing_datasets = [dataset.dataset_id for dataset in client.list_datasets()]
if not dataset_id in existing_datasets:
client.create_dataset(dataset_id)
dataset_ref = client.dataset(dataset_id)
existing_tables = [table.table_id for table in client.list_tables(dataset_id)]
for filename in os.listdir(path):
table_id = filename.rstrip('.csv')
if not table_id in existing_tables:
table_ref = dataset_ref.table(table_id)
with open(path + filename, 'rb') as source_file:
job = client.load_table_from_file(source_file, table_ref, job_config=job_config)
job.result()
table_ref = dataset_ref.table('train')
train = client.get_table(table_ref)
train.schema
client.list_rows(train, max_results=5).to_dataframe() | code |
17118100/cell_12 | [
"application_vnd.jupyter.stderr_output_2.png",
"text_plain_output_1.png"
] | from google.cloud import bigquery
import os
import numpy as np
import pandas as pd
import os
PROJECT_ID = 'YOUR_OWN_PROJECT_ID'
from google.cloud import bigquery
client = bigquery.Client(project=PROJECT_ID, location='US')
path = '../input/'
dataset_id = 'titanic'
dataset_ref = client.dataset(dataset_id)
job_config = bigquery.LoadJobConfig()
job_config.source_format = bigquery.SourceFormat.CSV
job_config.skip_leading_rows = 1
job_config.autodetect = True
existing_datasets = [dataset.dataset_id for dataset in client.list_datasets()]
if not dataset_id in existing_datasets:
client.create_dataset(dataset_id)
dataset_ref = client.dataset(dataset_id)
existing_tables = [table.table_id for table in client.list_tables(dataset_id)]
for filename in os.listdir(path):
table_id = filename.rstrip('.csv')
if not table_id in existing_tables:
table_ref = dataset_ref.table(table_id)
with open(path + filename, 'rb') as source_file:
job = client.load_table_from_file(source_file, table_ref, job_config=job_config)
job.result()
table_ref = dataset_ref.table('train')
train = client.get_table(table_ref)
train.schema
client.list_rows(train, max_results=5).to_dataframe()
train_query = "\nCREATE OR REPLACE MODEL `titanic.logistic_reg_model`\nOPTIONS (\n model_type = 'logistic_reg'\n ) AS (\nSELECT\n Pclass, Name, Sex, Age, SibSp, Parch, Ticket, Fare, Cabin, Embarked, Survived AS label\nFROM\n `titanic.train`)\n "
train_query_job = client.query(train_query) | code |
17118100/cell_5 | [
"application_vnd.jupyter.stderr_output_1.png"
] | from google.cloud import bigquery
PROJECT_ID = 'YOUR_OWN_PROJECT_ID'
from google.cloud import bigquery
client = bigquery.Client(project=PROJECT_ID, location='US')
path = '../input/'
dataset_id = 'titanic'
dataset_ref = client.dataset(dataset_id)
job_config = bigquery.LoadJobConfig()
job_config.source_format = bigquery.SourceFormat.CSV
job_config.skip_leading_rows = 1
job_config.autodetect = True
existing_datasets = [dataset.dataset_id for dataset in client.list_datasets()]
if not dataset_id in existing_datasets:
client.create_dataset(dataset_id) | code |
129037699/cell_13 | [
"text_plain_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
df = pd.read_csv('/kaggle/input/videogamesales/vgsales.csv')
df
dfsorted = df.sort_values('Global_Sales', ascending=False)
top20 = dfsorted.head(20)
top20
NAsorted = df.sort_values('NA_Sales', ascending=False)
NA_median = NAsorted['NA_Sales'].median()
surrounding_games = NAsorted.loc[(NAsorted['NA_Sales'] >= NA_median) & (NAsorted['NA_Sales'] <= NA_median)].head(10)
print(NA_median)
surrounding_games | code |
129037699/cell_9 | [
"text_plain_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
df = pd.read_csv('/kaggle/input/videogamesales/vgsales.csv')
df
genre = df['Genre'].value_counts().idxmax()
genre | code |
129037699/cell_11 | [
"text_html_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
df = pd.read_csv('/kaggle/input/videogamesales/vgsales.csv')
df
dfsorted = df.sort_values('Global_Sales', ascending=False)
top20 = dfsorted.head(20)
top20 | code |
129037699/cell_1 | [
"text_plain_output_1.png"
] | import os
import numpy as np
import pandas as pd
import os
for dirname, _, filenames in os.walk('/kaggle/input'):
for filename in filenames:
print(os.path.join(dirname, filename)) | code |
129037699/cell_7 | [
"text_plain_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
df = pd.read_csv('/kaggle/input/videogamesales/vgsales.csv')
df
platform = df['Platform'].value_counts().idxmax()
platform | code |
129037699/cell_15 | [
"text_plain_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
df = pd.read_csv('/kaggle/input/videogamesales/vgsales.csv')
df
dfsorted = df.sort_values('Global_Sales', ascending=False)
top20 = dfsorted.head(20)
top20
NAsorted= df.sort_values('NA_Sales', ascending=False)
NA_median = NAsorted['NA_Sales'].median()
surrounding_games = NAsorted.loc[(NAsorted['NA_Sales'] >= NA_median) & (NAsorted['NA_Sales'] <= NA_median)].head(10)
print(NA_median)
surrounding_games
df_sorted = df.sort_values('Global_Sales', ascending=False)
na_sales_top_game = df_sorted['NA_Sales'].iloc[0]
mean_na_sales = df_sorted['NA_Sales'].mean()
std_na_sales = df_sorted['NA_Sales'].std()
num_std = (na_sales_top_game - mean_na_sales) / std_na_sales
print(na_sales_top_game)
print(num_std) | code |
129037699/cell_3 | [
"text_html_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
df = pd.read_csv('/kaggle/input/videogamesales/vgsales.csv')
df | code |
129037699/cell_17 | [
"text_plain_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
df = pd.read_csv('/kaggle/input/videogamesales/vgsales.csv')
df
dfsorted = df.sort_values('Global_Sales', ascending=False)
top20 = dfsorted.head(20)
top20
NAsorted= df.sort_values('NA_Sales', ascending=False)
NA_median = NAsorted['NA_Sales'].median()
surrounding_games = NAsorted.loc[(NAsorted['NA_Sales'] >= NA_median) & (NAsorted['NA_Sales'] <= NA_median)].head(10)
print(NA_median)
surrounding_games
df_sorted = df.sort_values('Global_Sales', ascending=False)
na_sales_top_game = df_sorted['NA_Sales'].iloc[0]
mean_na_sales = df_sorted['NA_Sales'].mean()
std_na_sales = df_sorted['NA_Sales'].std()
num_std = (na_sales_top_game - mean_na_sales) / std_na_sales
grouped_by_platform = df.groupby('Platform')['Global_Sales'].mean().reset_index()
wii_data = grouped_by_platform.loc[grouped_by_platform['Platform'] == 'Wii']
mean_except_wii = grouped_by_platform.loc[grouped_by_platform['Platform'] != 'Wii']['Global_Sales'].mean()
if wii_data['Global_Sales'].values > mean_except_wii:
print('The average number of sales for the Nintendo Wii is higher than the other platforms.')
else:
print('The average number of sales for the Nintendo Wii is not higher than the other platforms.') | code |
129037699/cell_5 | [
"text_html_output_1.png",
"text_plain_output_1.png"
] | import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
df = pd.read_csv('/kaggle/input/videogamesales/vgsales.csv')
df
publisher = df['Publisher'].value_counts().idxmax()
publisher | code |
16153263/cell_13 | [
"text_plain_output_1.png"
] | from sklearn import metrics
import pandas as pd
train = pd.read_csv('../input/emnist-balanced-train.csv', header=None)
test = pd.read_csv('../input/emnist-balanced-test.csv', header=None)
train_data = train.values[:, 1:]
train_labels = train.values[:, 0]
test_data = test.values[:, 1:]
test_labels = test.values[:, 0]
model = {}
y_pred = {}
for idx in ['perceptron', 'sgd', 'knn', 'nbayes', 'tree', 'forest', 'boosting']:
model[idx].fit(train_data, train_labels)
y_pred[idx] = model[idx].predict(test_data)
print(idx, metrics.accuracy_score(test_labels, y_pred[idx])) | code |
16153263/cell_2 | [
"text_html_output_1.png"
] | import os
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import os
print(os.listdir('../input')) | code |
16153263/cell_7 | [
"application_vnd.jupyter.stderr_output_2.png",
"text_plain_output_3.png",
"text_plain_output_1.png"
] | import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
train = pd.read_csv('../input/emnist-balanced-train.csv', header=None)
test = pd.read_csv('../input/emnist-balanced-test.csv', header=None)
train_data = train.values[:, 1:]
train_labels = train.values[:, 0]
test_data = test.values[:, 1:]
test_labels = test.values[:, 0]
img_flip = np.transpose(train_data[8].reshape(28, 28), axes=[1, 0])
plt.imshow(img_flip, cmap='Greys_r')
plt.show() | code |
16153263/cell_5 | [
"image_output_1.png"
] | import pandas as pd
train = pd.read_csv('../input/emnist-balanced-train.csv', header=None)
train.head() | code |
48165764/cell_3 | [
"application_vnd.jupyter.stderr_output_1.png"
] | from sklearn.metrics import log_loss
import numpy as np
import pandas as pd
intercept = 0.9668835542286371
coef = [-0.964522806, 1.17172755, 0.179847765, 0.715775158, 0.943909901, 0.771809223, 0.166641048, 3.36556037, -3.04227717, -1.27799877, 3.24426534, 0.0839780267, -2.01609571, -0.204835338, -0.915522223, -2.85557145, 0.193031923, -0.610949126, 2.36655937, -0.688545522, -0.413897388, 2.36123594, -5.32875178, -0.257607263, -1.59041078, 0.56651361, -4.30048221, 0.583728152, -0.755718528, -4.85939641, -4.77435586, 1.20158973, 1.52684705, 1.42038088, -0.470860607, -1.02184815, -0.186602382, 0.350855131, -2.86119402, 0.181263079, -3.98183995, 3.85179998, 0.595682016, 0.976153845, 0.456293927, -1.33132314, 0.363681261, 0.794664912, -3.9104109, -0.120555465, -3.9104109, -2.9114099, 1.36755837, 2.53799497, -4.6606329, -0.630443626, 0.166641048, -7.08818234, -1.03160754, -0.214803976, 4.86850322, 1.36755837, -2.9660072, -1.35837391, -5.08890666, 1.20158973, -3.9104109, -6.2872943, -0.297537878, -2.18535836, -2.59118068, 3.86950222, -0.633065361, -1.89416541, -0.448762478, 0.40582232, 1.36755837, 1.14356994, -1.51346861, -1.77735286, 0.684120905, 1.21749418, -2.01609571, -1.33132314, 0.42970794, -2.50444579, -0.413897388, 1.46527571, 0.475974538, 1.20782761, -1.79853836, 7.02807914, -5.51602225, 0.0839780267, 4.33789914, -1.96700816, -4.87378955, -0.385577422, 1.38106353, -4.9640092, 3.8952826, -2.9189663, 1.36755837, 0.684120905, -4.47665162, -1.27799877, 1.36755837, 3.36556037, -0.204835338, -0.599912328, -0.204314343, -4.45619265, -1.26221209, -1.55571983, -1.93596293, -0.0675983678, -2.55472083, -0.610949126, 2.36655937, -1.85313989, -0.962103086, 4.36456137, 4.86850322, -2.92095622, -1.42047927, 4.36456137, 2.36655937, 0.179847765, 2.61138626, -1.2224172, 2.15956162, 1.36755837, -1.33363593, 2.31341538, -2.49972847, -2.05475149, 0.0309957974, -0.0693110501, 2.36655937, 0.581666839, 4.36456137, -4.30048221, -2.8726154, -3.08550042, -1.08649626, -0.755718528, 1.08199981, -1.42047927, 0.806595523, 4.86850322, 0.409200801, -1.9670062, -0.0693110501, -3.18329286, -2.27241201, -0.00208025864, 0.778295227, -1.16768184, 3.36556037, 2.9219428, 0.121032855, -1.13677382, 4.36456137, -5.05666468, 2.54763868, -2.05653977, -0.836236669, -1.37407599, -1.52765698, -0.55544063, -4.77435586, 0.392717805, 2.54763868, -0.245071712, 0.806595523, 0.42970794, 1.61238526, 5.11333011, 4.11432911, 1.20782761, -1.13677382, -1.39447556, -2.2753443, -1.00108126, 2.61138626, 1.61238526, -2.1169698, -2.04970922, -3.055864, -2.06237301, -5.30942364, -3.060378, 4.56786669, 0.724590988, -1.19599384, -1.69171272, -0.599912328, 0.409200801, 4.36456137, -0.232589722, 2.05197573, -0.304659185, -2.4147726, -0.665305184, 0.794664912, -0.523026461, 1.48758438, -2.1169698, -0.599912328, -1.06918017, -2.92509555, -2.24778342, -2.05998755, -1.91946605, -1.27341101, 3.42621179, -0.945321798, -5.05666468, -1.16768184, -0.385577422, 2.36655937, 1.54600888, -1.96908321, 1.48758438, 1.36755837, -2.93685912, -0.863282842, -1.77796891, 1.36755837, 5.1035469, 3.07753467, 3.36556037, -1.70669288, -3.055864, -5.71550981, 1.35386262, 2.36655937, -0.945456361, -0.945321798, 0.456293927, -0.001963082, -2.52981213, -0.633065361, 4.36456137, -3.22981906, -2.46960559, -2.46960559, 1.26188506, 1.27495232, 1.27947673, -2.71404316, -1.71703518, -1.71742525]
def layers_num(x):
return int(np.round(np.dot(np.array(x), coef) + intercept))
col = pd.read_csv('../input/lish-moa/train_targets_scored.csv').columns[1:]
startword = ['-', 'erase', 'acetylcholine', 'receptor_agonist', 'atp', 'anta', 'abl', 'peptide', 'tgf', 'calcicycl', 'hsp', 'im', 'proteas', 'secret', 'synth']
for subword in startword:
col = [drtitle.replace(subword, '_' + subword + '_') for drtitle in col]
all = [drtitle.split('_') for drtitle in col]
all = sorted([word for item in all for word in item])
all = [word for word in all if len(word) > 0]
allcounted = {word: all.count(word) for word in all}
allcounted = {key: value for key, value in allcounted.items()}
allcounted = list(allcounted.keys()) + ['receptor_antagonist', 'tlr_antagonist', 'tlr_antagonist', 'acetylcholine_receptor_antagonist', 'cannabinoid_receptor_antagonist', 'opioid_receptor_antagonist', 'ppar_receptor_antagonist', 'progesterone_receptor_antagonist', 'serotonin_receptor_antagonist']
eps = 10 ** (-7)
def cube(x):
x /= 10.0
return 10.0 * x * x * x
def tgfunc(x):
return np.tan(np.pi * x / 4.0)
def application_transform(x):
""" x = 2*x-1.0
x1 = weight_tg*tgfunc(x) + (weight_x3 + weight_x5*x**2)*x**3 + weight*x
x1 = x1/2 + 0.5"""
x1 = x
if x1 < eps:
x1 = eps
elif x1 >= 1 - eps:
x1 = 1 - eps
return x1
def transform(x):
x = x[x['cp_type'] == 'trt_cp']
x.pop('cp_type')
x['cp_dose'] = x['cp_dose'].replace({'D1': -1, 'D2': 1})
x['cp_time'] = x['cp_time'] // 24
genes = [feature for feature in x.columns if feature.startswith('g-')]
cells = [feature for feature in x.columns if feature.startswith('c-')]
x['g-mean'] = x[genes].mean(axis=1)
x['g-std'] = x[genes].std(axis=1)
x['g-kur'] = x[genes].kurtosis(axis=1)
x['g-skew'] = x[genes].skew(axis=1)
x['c-mean'] = x[cells].mean(axis=1)
x['c-std'] = x[cells].std(axis=1)
x['c-kur'] = x[cells].kurtosis(axis=1)
x['c-skew'] = x[cells].skew(axis=1)
x['mean'] = x[genes + cells].mean(axis=1)
x['std'] = x[genes + cells].std(axis=1)
x['kur'] = x[genes + cells].kurtosis(axis=1)
x['skew'] = x[genes + cells].skew(axis=1)
x.join(pd.get_dummies(x['cp_time']))
x.pop('cp_time')
return x
def transform2(x):
x.pop('cp_type')
x['cp_dose'] = x['cp_dose'].replace({'D1': -1, 'D2': 1})
x['cp_time'] = x['cp_time'] // 24
genes = [feature for feature in x.columns if feature.startswith('g-')]
cells = [feature for feature in x.columns if feature.startswith('c-')]
x['g-mean'] = x[genes].mean(axis=1)
x['g-std'] = x[genes].std(axis=1)
x['g-kur'] = x[genes].kurtosis(axis=1)
x['g-skew'] = x[genes].skew(axis=1)
x['c-mean'] = x[cells].mean(axis=1)
x['c-std'] = x[cells].std(axis=1)
x['c-kur'] = x[cells].kurtosis(axis=1)
x['c-skew'] = x[cells].skew(axis=1)
x['mean'] = x[genes + cells].mean(axis=1)
x['std'] = x[genes + cells].std(axis=1)
x['kur'] = x[genes + cells].kurtosis(axis=1)
x['skew'] = x[genes + cells].skew(axis=1)
x.join(pd.get_dummies(x['cp_time']))
x.pop('cp_time')
return x
X_all = pd.read_csv('/kaggle/input/lish-moa/train_features.csv')
y_all = pd.read_csv('/kaggle/input/lish-moa/train_targets_scored.csv')
X = transform(X_all[X_all.columns[1:]]).astype(np.float16)
Y = y_all[X_all['cp_type'] == 'trt_cp']
Y = Y[Y.columns[1:]].astype(np.int8)
def metric(y_true, y_pred):
metrics = []
for _target in train_targets.columns:
metrics.append(log_loss(y_true.loc[:, _target], y_pred.loc[:, _target].astype(float), labels=[0, 1]))
return np.mean(metrics) | code |
128047328/cell_42 | [
"image_output_1.png"
] | best = create_model('et')
bagged_ensemble = ensemble_model(best) | code |
128047328/cell_63 | [
"text_html_output_2.png",
"text_plain_output_1.png"
] | from rdkit import Chem
from rdkit.Chem import AllChem
import deepchem as dc
import numpy as np
import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
smiles_list = df_train['SMILES'][:10]
mol_list = []
for smile in smiles_list:
mol = Chem.MolFromSmiles(smile)
mol_list.append(mol)
img = Chem.Draw.MolsToGridImage(mol_list, molsPerRow=5)
img
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
# Define a function to featurize a SMILES string
def featurize_smiles(smiles):
mol = Chem.MolFromSmiles(smiles)
fp = AllChem.GetMorganFingerprintAsBitVect(mol, 3, nBits=1024)
features = np.array(list(fp.ToBitString())).astype(float)
return features
smiles = ['CCC']
featurizer = dc.feat.Mol2VecFingerprint()
features = featurizer.featurize(smiles)
features = featurizer.featurize(df_train['SMILES'])
features.shape
smiles = ['CC(=O)OC1=CC=CC=C1C(=O)O']
featurizer = dc.feat.RDKitDescriptors()
features = featurizer.featurize(smiles)
features.shape | code |
128047328/cell_21 | [
"text_html_output_1.png"
] | from rdkit import Chem
from rdkit.Chem import AllChem
import numpy as np
import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
smiles_list = df_train['SMILES'][:10]
mol_list = []
for smile in smiles_list:
mol = Chem.MolFromSmiles(smile)
mol_list.append(mol)
img = Chem.Draw.MolsToGridImage(mol_list, molsPerRow=5)
img
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
# Define a function to featurize a SMILES string
def featurize_smiles(smiles):
mol = Chem.MolFromSmiles(smiles)
fp = AllChem.GetMorganFingerprintAsBitVect(mol, 3, nBits=1024)
features = np.array(list(fp.ToBitString())).astype(float)
return features
X = np.array([featurize_smiles(smiles) for smiles in df_train['SMILES']]) | code |
128047328/cell_25 | [
"text_plain_output_1.png"
] | from rdkit import Chem
from rdkit.Chem import AllChem
import numpy as np
import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
smiles_list = df_train['SMILES'][:10]
mol_list = []
for smile in smiles_list:
mol = Chem.MolFromSmiles(smile)
mol_list.append(mol)
img = Chem.Draw.MolsToGridImage(mol_list, molsPerRow=5)
img
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
# Define a function to featurize a SMILES string
def featurize_smiles(smiles):
mol = Chem.MolFromSmiles(smiles)
fp = AllChem.GetMorganFingerprintAsBitVect(mol, 3, nBits=1024)
features = np.array(list(fp.ToBitString())).astype(float)
return features
X = np.array([featurize_smiles(smiles) for smiles in df_train['SMILES']])
X.shape
X = pd.DataFrame(X)
X['Solubility'] = df_train['Solubility']
reg = setup(X, target='Solubility', session_id=123, train_size=0.8) | code |
128047328/cell_4 | [
"text_html_output_2.png",
"text_plain_output_1.png"
] | pip install deepchem | code |
128047328/cell_57 | [
"application_vnd.jupyter.stderr_output_16.png",
"application_vnd.jupyter.stderr_output_9.png",
"application_vnd.jupyter.stderr_output_2.png",
"application_vnd.jupyter.stderr_output_7.png",
"application_vnd.jupyter.stderr_output_11.png",
"application_vnd.jupyter.stderr_output_18.png",
"application_vnd.jupyter.stderr_output_4.png",
"application_vnd.jupyter.stderr_output_6.png",
"application_vnd.jupyter.stderr_output_12.png",
"application_vnd.jupyter.stderr_output_8.png",
"application_vnd.jupyter.stderr_output_10.png",
"application_vnd.jupyter.stderr_output_19.png",
"application_vnd.jupyter.stderr_output_13.png",
"application_vnd.jupyter.stderr_output_3.png",
"application_vnd.jupyter.stderr_output_5.png",
"application_vnd.jupyter.stderr_output_15.png",
"application_vnd.jupyter.stderr_output_17.png",
"application_vnd.jupyter.stderr_output_1.png",
"application_vnd.jupyter.stderr_output_14.png"
] | from rdkit import Chem
from rdkit.Chem import AllChem
import deepchem as dc
import numpy as np
import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
smiles_list = df_train['SMILES'][:10]
mol_list = []
for smile in smiles_list:
mol = Chem.MolFromSmiles(smile)
mol_list.append(mol)
img = Chem.Draw.MolsToGridImage(mol_list, molsPerRow=5)
img
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
# Define a function to featurize a SMILES string
def featurize_smiles(smiles):
mol = Chem.MolFromSmiles(smiles)
fp = AllChem.GetMorganFingerprintAsBitVect(mol, 3, nBits=1024)
features = np.array(list(fp.ToBitString())).astype(float)
return features
smiles = ['CCC']
featurizer = dc.feat.Mol2VecFingerprint()
features = featurizer.featurize(smiles)
features = featurizer.featurize(df_train['SMILES'])
features.shape | code |
128047328/cell_23 | [
"text_plain_output_1.png"
] | from rdkit import Chem
from rdkit.Chem import AllChem
import numpy as np
import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
smiles_list = df_train['SMILES'][:10]
mol_list = []
for smile in smiles_list:
mol = Chem.MolFromSmiles(smile)
mol_list.append(mol)
img = Chem.Draw.MolsToGridImage(mol_list, molsPerRow=5)
img
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
# Define a function to featurize a SMILES string
def featurize_smiles(smiles):
mol = Chem.MolFromSmiles(smiles)
fp = AllChem.GetMorganFingerprintAsBitVect(mol, 3, nBits=1024)
features = np.array(list(fp.ToBitString())).astype(float)
return features
X = np.array([featurize_smiles(smiles) for smiles in df_train['SMILES']])
X.shape | code |
128047328/cell_79 | [
"text_html_output_2.png",
"text_plain_output_1.png"
] | import deepchem as dc
import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
smiles = ['CCC']
featurizer = dc.feat.Mol2VecFingerprint()
features = featurizer.featurize(smiles)
features = featurizer.featurize(df_train['SMILES'])
smiles = ['CC(=O)OC1=CC=CC=C1C(=O)O']
featurizer = dc.feat.RDKitDescriptors()
features = featurizer.featurize(smiles)
features = featurizer.featurize(df_train['SMILES'])
smiles = ['CC(=O)OC1=CC=CC=C1C(=O)O']
featurizer = dc.feat.MordredDescriptors()
features = featurizer.featurize(smiles)
features = featurizer.featurize(df_train['SMILES'])
test_features = featurizer.featurize(df_test['SMILES']) | code |
128047328/cell_30 | [
"text_html_output_1.png"
] | from rdkit import Chem
from rdkit.Chem import AllChem
from rdkit.Chem import MolFromSmiles, Descriptors, MolFromSmarts
import numpy as np
import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
smiles_list = df_train['SMILES'][:10]
mol_list = []
for smile in smiles_list:
mol = Chem.MolFromSmiles(smile)
mol_list.append(mol)
img = Chem.Draw.MolsToGridImage(mol_list, molsPerRow=5)
img
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
# Define a function to featurize a SMILES string
def featurize_smiles(smiles):
mol = Chem.MolFromSmiles(smiles)
fp = AllChem.GetMorganFingerprintAsBitVect(mol, 3, nBits=1024)
features = np.array(list(fp.ToBitString())).astype(float)
return features
X = np.array([featurize_smiles(smiles) for smiles in df_train['SMILES']])
X.shape
X = pd.DataFrame(X)
X['Solubility'] = df_train['Solubility']
from rdkit.Chem import MolFromSmiles, Descriptors, MolFromSmarts
def create_descriptors(smiles):
mol = MolFromSmiles(smiles)
mw = Descriptors.MolWt(mol)
logp = Descriptors.MolLogP(mol)
rotb = Descriptors.NumRotatableBonds(mol)
arom_proportion = len(mol.GetSubstructMatches(MolFromSmarts('a'))) / Descriptors.HeavyAtomCount(mol)
hbd = Descriptors.NumHDonors(mol)
hba = Descriptors.NumHAcceptors(mol)
non_carbon_proportion = len(mol.GetSubstructMatches(MolFromSmarts('[!#6]'))) / Descriptors.HeavyAtomCount(mol)
psa = Descriptors.TPSA(mol, includeSandP=True)
fsp3 = Descriptors.FractionCSP3(mol)
return (mw, logp, rotb, arom_proportion, hbd, hba, non_carbon_proportion, psa, fsp3)
descriptors = [create_descriptors(smiles) for smiles in df_train['SMILES']]
col_names = ['mw', 'logp', 'rotb', 'ap', 'hbd', 'hba', 'non_cp', 'psa', 'fsp3']
train = pd.DataFrame(descriptors, columns=col_names)
train = train.join(df_train['Solubility'], how='left')
train | code |
128047328/cell_44 | [
"text_html_output_2.png",
"text_plain_output_1.png"
] | best = create_model('et')
boosted_ensemble = ensemble_model(best, method='Boosting') | code |
128047328/cell_20 | [
"text_plain_output_1.png"
] | import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
df_train['Solubility'].min() | code |
128047328/cell_76 | [
"text_html_output_1.png"
] | best_top3_models = compare_models(n_select=3) | code |
128047328/cell_40 | [
"text_plain_output_1.png"
] | best = create_model('et')
plot_model(best, 'feature') | code |
128047328/cell_29 | [
"text_plain_output_1.png"
] | from rdkit import Chem
from rdkit.Chem import AllChem
from rdkit.Chem import MolFromSmiles, Descriptors, MolFromSmarts
import numpy as np
import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
smiles_list = df_train['SMILES'][:10]
mol_list = []
for smile in smiles_list:
mol = Chem.MolFromSmiles(smile)
mol_list.append(mol)
img = Chem.Draw.MolsToGridImage(mol_list, molsPerRow=5)
img
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
# Define a function to featurize a SMILES string
def featurize_smiles(smiles):
mol = Chem.MolFromSmiles(smiles)
fp = AllChem.GetMorganFingerprintAsBitVect(mol, 3, nBits=1024)
features = np.array(list(fp.ToBitString())).astype(float)
return features
X = np.array([featurize_smiles(smiles) for smiles in df_train['SMILES']])
X.shape
X = pd.DataFrame(X)
X['Solubility'] = df_train['Solubility']
from rdkit.Chem import MolFromSmiles, Descriptors, MolFromSmarts
def create_descriptors(smiles):
mol = MolFromSmiles(smiles)
mw = Descriptors.MolWt(mol)
logp = Descriptors.MolLogP(mol)
rotb = Descriptors.NumRotatableBonds(mol)
arom_proportion = len(mol.GetSubstructMatches(MolFromSmarts('a'))) / Descriptors.HeavyAtomCount(mol)
hbd = Descriptors.NumHDonors(mol)
hba = Descriptors.NumHAcceptors(mol)
non_carbon_proportion = len(mol.GetSubstructMatches(MolFromSmarts('[!#6]'))) / Descriptors.HeavyAtomCount(mol)
psa = Descriptors.TPSA(mol, includeSandP=True)
fsp3 = Descriptors.FractionCSP3(mol)
return (mw, logp, rotb, arom_proportion, hbd, hba, non_carbon_proportion, psa, fsp3)
descriptors = [create_descriptors(smiles) for smiles in df_train['SMILES']]
col_names = ['mw', 'logp', 'rotb', 'ap', 'hbd', 'hba', 'non_cp', 'psa', 'fsp3']
train = pd.DataFrame(descriptors, columns=col_names)
train = train.join(df_train['Solubility'], how='left') | code |
128047328/cell_39 | [
"text_html_output_2.png",
"text_plain_output_1.png"
] | best = create_model('et')
plot_model(best, 'error') | code |
128047328/cell_26 | [
"application_vnd.jupyter.stderr_output_2.png",
"application_vnd.jupyter.stderr_output_1.png"
] | best_top3_models = compare_models(n_select=3) | code |
128047328/cell_48 | [
"text_html_output_2.png",
"text_plain_output_1.png"
] | from rdkit import Chem
from rdkit.Chem import AllChem
from rdkit.Chem import MolFromSmiles, Descriptors, MolFromSmarts
import numpy as np
import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
smiles_list = df_train['SMILES'][:10]
mol_list = []
for smile in smiles_list:
mol = Chem.MolFromSmiles(smile)
mol_list.append(mol)
img = Chem.Draw.MolsToGridImage(mol_list, molsPerRow=5)
img
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
# Define a function to featurize a SMILES string
def featurize_smiles(smiles):
mol = Chem.MolFromSmiles(smiles)
fp = AllChem.GetMorganFingerprintAsBitVect(mol, 3, nBits=1024)
features = np.array(list(fp.ToBitString())).astype(float)
return features
X = np.array([featurize_smiles(smiles) for smiles in df_train['SMILES']])
X.shape
X = pd.DataFrame(X)
X['Solubility'] = df_train['Solubility']
from rdkit.Chem import MolFromSmiles, Descriptors, MolFromSmarts
def create_descriptors(smiles):
mol = MolFromSmiles(smiles)
mw = Descriptors.MolWt(mol)
logp = Descriptors.MolLogP(mol)
rotb = Descriptors.NumRotatableBonds(mol)
arom_proportion = len(mol.GetSubstructMatches(MolFromSmarts('a'))) / Descriptors.HeavyAtomCount(mol)
hbd = Descriptors.NumHDonors(mol)
hba = Descriptors.NumHAcceptors(mol)
non_carbon_proportion = len(mol.GetSubstructMatches(MolFromSmarts('[!#6]'))) / Descriptors.HeavyAtomCount(mol)
psa = Descriptors.TPSA(mol, includeSandP=True)
fsp3 = Descriptors.FractionCSP3(mol)
return (mw, logp, rotb, arom_proportion, hbd, hba, non_carbon_proportion, psa, fsp3)
descriptors = [create_descriptors(smiles) for smiles in df_train['SMILES']]
col_names = ['mw', 'logp', 'rotb', 'ap', 'hbd', 'hba', 'non_cp', 'psa', 'fsp3']
train = pd.DataFrame(descriptors, columns=col_names)
train = train.join(df_train['Solubility'], how='left')
descriptors = [create_descriptors(smiles) for smiles in df_test['SMILES']]
col_names = ['mw', 'logp', 'rotb', 'ap', 'hbd', 'hba', 'non_cp', 'psa', 'fsp3']
test = pd.DataFrame(descriptors, columns=col_names) | code |
128047328/cell_73 | [
"application_vnd.jupyter.stderr_output_27.png",
"application_vnd.jupyter.stderr_output_35.png",
"application_vnd.jupyter.stderr_output_24.png",
"application_vnd.jupyter.stderr_output_16.png",
"application_vnd.jupyter.stderr_output_9.png",
"application_vnd.jupyter.stderr_output_32.png",
"application_vnd.jupyter.stderr_output_2.png",
"application_vnd.jupyter.stderr_output_7.png",
"application_vnd.jupyter.stderr_output_11.png",
"application_vnd.jupyter.stderr_output_18.png",
"application_vnd.jupyter.stderr_output_38.png",
"application_vnd.jupyter.stderr_output_4.png",
"application_vnd.jupyter.stderr_output_26.png",
"application_vnd.jupyter.stderr_output_6.png",
"application_vnd.jupyter.stderr_output_31.png",
"application_vnd.jupyter.stderr_output_33.png",
"application_vnd.jupyter.stderr_output_25.png",
"application_vnd.jupyter.stderr_output_12.png",
"application_vnd.jupyter.stderr_output_8.png",
"application_vnd.jupyter.stderr_output_10.png",
"application_vnd.jupyter.stderr_output_23.png",
"application_vnd.jupyter.stderr_output_34.png",
"application_vnd.jupyter.stderr_output_19.png",
"application_vnd.jupyter.stderr_output_13.png",
"application_vnd.jupyter.stderr_output_3.png",
"application_vnd.jupyter.stderr_output_5.png",
"application_vnd.jupyter.stderr_output_30.png",
"application_vnd.jupyter.stderr_output_15.png",
"application_vnd.jupyter.stderr_output_17.png",
"application_vnd.jupyter.stderr_output_28.png",
"application_vnd.jupyter.stderr_output_20.png",
"application_vnd.jupyter.stderr_output_36.png",
"application_vnd.jupyter.stderr_output_22.png",
"application_vnd.jupyter.stderr_output_29.png",
"application_vnd.jupyter.stderr_output_1.png",
"application_vnd.jupyter.stderr_output_14.png",
"application_vnd.jupyter.stderr_output_21.png",
"application_vnd.jupyter.stderr_output_37.png"
] | from rdkit import Chem
from rdkit.Chem import AllChem
import deepchem as dc
import numpy as np
import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
smiles_list = df_train['SMILES'][:10]
mol_list = []
for smile in smiles_list:
mol = Chem.MolFromSmiles(smile)
mol_list.append(mol)
img = Chem.Draw.MolsToGridImage(mol_list, molsPerRow=5)
img
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
# Define a function to featurize a SMILES string
def featurize_smiles(smiles):
mol = Chem.MolFromSmiles(smiles)
fp = AllChem.GetMorganFingerprintAsBitVect(mol, 3, nBits=1024)
features = np.array(list(fp.ToBitString())).astype(float)
return features
smiles = ['CCC']
featurizer = dc.feat.Mol2VecFingerprint()
features = featurizer.featurize(smiles)
features = featurizer.featurize(df_train['SMILES'])
features.shape
smiles = ['CC(=O)OC1=CC=CC=C1C(=O)O']
featurizer = dc.feat.RDKitDescriptors()
features = featurizer.featurize(smiles)
features.shape
features = featurizer.featurize(df_train['SMILES'])
smiles = ['CC(=O)OC1=CC=CC=C1C(=O)O']
featurizer = dc.feat.MordredDescriptors()
features = featurizer.featurize(smiles)
features = featurizer.featurize(df_train['SMILES'])
features.shape | code |
128047328/cell_41 | [
"image_output_1.png"
] | best = create_model('et')
prediction_holdout = predict_model(best) | code |
128047328/cell_2 | [
"application_vnd.jupyter.stderr_output_1.png",
"image_output_1.png"
] | pip install rdkit | code |
128047328/cell_72 | [
"text_plain_output_1.png"
] | import deepchem as dc
import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
smiles = ['CCC']
featurizer = dc.feat.Mol2VecFingerprint()
features = featurizer.featurize(smiles)
features = featurizer.featurize(df_train['SMILES'])
smiles = ['CC(=O)OC1=CC=CC=C1C(=O)O']
featurizer = dc.feat.RDKitDescriptors()
features = featurizer.featurize(smiles)
features = featurizer.featurize(df_train['SMILES'])
smiles = ['CC(=O)OC1=CC=CC=C1C(=O)O']
featurizer = dc.feat.MordredDescriptors()
features = featurizer.featurize(smiles)
features = featurizer.featurize(df_train['SMILES']) | code |
128047328/cell_67 | [
"text_html_output_1.png"
] | best_top3_models = compare_models(n_select=3) | code |
128047328/cell_60 | [
"text_html_output_1.png"
] | best_top3_models = compare_models(n_select=3) | code |
128047328/cell_19 | [
"text_plain_output_1.png"
] | import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
df_train['Solubility'].max() | code |
128047328/cell_69 | [
"text_html_output_2.png",
"text_plain_output_1.png"
] | !pip install mordred | code |
128047328/cell_50 | [
"application_vnd.jupyter.stderr_output_2.png",
"application_vnd.jupyter.stderr_output_7.png",
"application_vnd.jupyter.stderr_output_4.png",
"application_vnd.jupyter.stderr_output_6.png",
"application_vnd.jupyter.stderr_output_3.png",
"application_vnd.jupyter.stderr_output_5.png",
"application_vnd.jupyter.stderr_output_1.png"
] | from rdkit import Chem
from rdkit.Chem import AllChem
from rdkit.Chem import MolFromSmiles, Descriptors, MolFromSmarts
import numpy as np
import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
smiles_list = df_train['SMILES'][:10]
mol_list = []
for smile in smiles_list:
mol = Chem.MolFromSmiles(smile)
mol_list.append(mol)
img = Chem.Draw.MolsToGridImage(mol_list, molsPerRow=5)
img
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
# Define a function to featurize a SMILES string
def featurize_smiles(smiles):
mol = Chem.MolFromSmiles(smiles)
fp = AllChem.GetMorganFingerprintAsBitVect(mol, 3, nBits=1024)
features = np.array(list(fp.ToBitString())).astype(float)
return features
X = np.array([featurize_smiles(smiles) for smiles in df_train['SMILES']])
X.shape
X = pd.DataFrame(X)
X['Solubility'] = df_train['Solubility']
from rdkit.Chem import MolFromSmiles, Descriptors, MolFromSmarts
def create_descriptors(smiles):
mol = MolFromSmiles(smiles)
mw = Descriptors.MolWt(mol)
logp = Descriptors.MolLogP(mol)
rotb = Descriptors.NumRotatableBonds(mol)
arom_proportion = len(mol.GetSubstructMatches(MolFromSmarts('a'))) / Descriptors.HeavyAtomCount(mol)
hbd = Descriptors.NumHDonors(mol)
hba = Descriptors.NumHAcceptors(mol)
non_carbon_proportion = len(mol.GetSubstructMatches(MolFromSmarts('[!#6]'))) / Descriptors.HeavyAtomCount(mol)
psa = Descriptors.TPSA(mol, includeSandP=True)
fsp3 = Descriptors.FractionCSP3(mol)
return (mw, logp, rotb, arom_proportion, hbd, hba, non_carbon_proportion, psa, fsp3)
descriptors = [create_descriptors(smiles) for smiles in df_train['SMILES']]
col_names = ['mw', 'logp', 'rotb', 'ap', 'hbd', 'hba', 'non_cp', 'psa', 'fsp3']
train = pd.DataFrame(descriptors, columns=col_names)
train = train.join(df_train['Solubility'], how='left')
additional_col = ['MolMR', 'NumValenceElectrons', 'NumAromaticRings', 'NumSaturatedRings', 'NumAliphaticRings', 'RingCount', 'LabuteASA', 'BalabanJ', 'BertzCT']
df = train.join(df_train[additional_col], how='left')
reg = setup(df, target='Solubility', session_id=123, train_size=0.8, numeric_features=list(df.drop(columns=['Solubility']).columns), transformation=True, normalize=True)
best = create_model('et')
final_model = finalize_model(best)
descriptors = [create_descriptors(smiles) for smiles in df_test['SMILES']]
col_names = ['mw', 'logp', 'rotb', 'ap', 'hbd', 'hba', 'non_cp', 'psa', 'fsp3']
test = pd.DataFrame(descriptors, columns=col_names)
df = test.join(df_test[additional_col], how='left')
df = predict_model(final_model, data=df) | code |
128047328/cell_64 | [
"text_plain_output_1.png"
] | import deepchem as dc
import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
smiles = ['CCC']
featurizer = dc.feat.Mol2VecFingerprint()
features = featurizer.featurize(smiles)
features = featurizer.featurize(df_train['SMILES'])
smiles = ['CC(=O)OC1=CC=CC=C1C(=O)O']
featurizer = dc.feat.RDKitDescriptors()
features = featurizer.featurize(smiles)
features = featurizer.featurize(df_train['SMILES']) | code |
128047328/cell_45 | [
"text_html_output_2.png",
"text_plain_output_1.png"
] | best = create_model('et')
boosted_ensemble_50estm = ensemble_model(best, method='Boosting', n_estimators=50) | code |
128047328/cell_32 | [
"application_vnd.jupyter.stderr_output_9.png",
"application_vnd.jupyter.stderr_output_2.png",
"application_vnd.jupyter.stderr_output_7.png",
"application_vnd.jupyter.stderr_output_11.png",
"application_vnd.jupyter.stderr_output_4.png",
"application_vnd.jupyter.stderr_output_6.png",
"application_vnd.jupyter.stderr_output_12.png",
"application_vnd.jupyter.stderr_output_8.png",
"application_vnd.jupyter.stderr_output_10.png",
"application_vnd.jupyter.stderr_output_13.png",
"application_vnd.jupyter.stderr_output_3.png",
"application_vnd.jupyter.stderr_output_5.png",
"application_vnd.jupyter.stderr_output_1.png",
"application_vnd.jupyter.stderr_output_14.png"
] | best_top3_models = compare_models(n_select=3) | code |
128047328/cell_59 | [
"text_plain_output_1.png"
] | from rdkit import Chem
from rdkit.Chem import AllChem
from rdkit.Chem import MolFromSmiles, Descriptors, MolFromSmarts
import deepchem as dc
import numpy as np
import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
smiles_list = df_train['SMILES'][:10]
mol_list = []
for smile in smiles_list:
mol = Chem.MolFromSmiles(smile)
mol_list.append(mol)
img = Chem.Draw.MolsToGridImage(mol_list, molsPerRow=5)
img
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
# Define a function to featurize a SMILES string
def featurize_smiles(smiles):
mol = Chem.MolFromSmiles(smiles)
fp = AllChem.GetMorganFingerprintAsBitVect(mol, 3, nBits=1024)
features = np.array(list(fp.ToBitString())).astype(float)
return features
X = np.array([featurize_smiles(smiles) for smiles in df_train['SMILES']])
X.shape
X = pd.DataFrame(X)
X['Solubility'] = df_train['Solubility']
from rdkit.Chem import MolFromSmiles, Descriptors, MolFromSmarts
def create_descriptors(smiles):
mol = MolFromSmiles(smiles)
mw = Descriptors.MolWt(mol)
logp = Descriptors.MolLogP(mol)
rotb = Descriptors.NumRotatableBonds(mol)
arom_proportion = len(mol.GetSubstructMatches(MolFromSmarts('a'))) / Descriptors.HeavyAtomCount(mol)
hbd = Descriptors.NumHDonors(mol)
hba = Descriptors.NumHAcceptors(mol)
non_carbon_proportion = len(mol.GetSubstructMatches(MolFromSmarts('[!#6]'))) / Descriptors.HeavyAtomCount(mol)
psa = Descriptors.TPSA(mol, includeSandP=True)
fsp3 = Descriptors.FractionCSP3(mol)
return (mw, logp, rotb, arom_proportion, hbd, hba, non_carbon_proportion, psa, fsp3)
descriptors = [create_descriptors(smiles) for smiles in df_train['SMILES']]
col_names = ['mw', 'logp', 'rotb', 'ap', 'hbd', 'hba', 'non_cp', 'psa', 'fsp3']
train = pd.DataFrame(descriptors, columns=col_names)
train = train.join(df_train['Solubility'], how='left')
additional_col = ['MolMR', 'NumValenceElectrons', 'NumAromaticRings', 'NumSaturatedRings', 'NumAliphaticRings', 'RingCount', 'LabuteASA', 'BalabanJ', 'BertzCT']
df = train.join(df_train[additional_col], how='left')
reg = setup(df, target='Solubility', session_id=123, train_size=0.8, numeric_features=list(df.drop(columns=['Solubility']).columns), transformation=True, normalize=True)
best = create_model('et')
final_model = finalize_model(best)
descriptors = [create_descriptors(smiles) for smiles in df_test['SMILES']]
col_names = ['mw', 'logp', 'rotb', 'ap', 'hbd', 'hba', 'non_cp', 'psa', 'fsp3']
test = pd.DataFrame(descriptors, columns=col_names)
df = test.join(df_test[additional_col], how='left')
df = predict_model(final_model, data=df)
df.rename(columns={'prediction_label': 'Solubility'}, inplace=True)
smiles = ['CCC']
featurizer = dc.feat.Mol2VecFingerprint()
features = featurizer.featurize(smiles)
features = featurizer.featurize(df_train['SMILES'])
features.shape
df = pd.DataFrame(features)
df['Solubility'] = df_train['Solubility']
reg = setup(df, target='Solubility', session_id=123, train_size=0.8, transformation=True) | code |
128047328/cell_16 | [
"application_vnd.jupyter.stderr_output_2.png",
"application_vnd.jupyter.stderr_output_1.png"
] | import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
reg = setup(X, target='Solubility', session_id=123, train_size=0.8) | code |
128047328/cell_38 | [
"text_html_output_1.png"
] | best = create_model('et')
print(best) | code |
128047328/cell_75 | [
"text_plain_output_1.png"
] | from rdkit import Chem
from rdkit.Chem import AllChem
from rdkit.Chem import MolFromSmiles, Descriptors, MolFromSmarts
import deepchem as dc
import numpy as np
import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
smiles_list = df_train['SMILES'][:10]
mol_list = []
for smile in smiles_list:
mol = Chem.MolFromSmiles(smile)
mol_list.append(mol)
img = Chem.Draw.MolsToGridImage(mol_list, molsPerRow=5)
img
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
# Define a function to featurize a SMILES string
def featurize_smiles(smiles):
mol = Chem.MolFromSmiles(smiles)
fp = AllChem.GetMorganFingerprintAsBitVect(mol, 3, nBits=1024)
features = np.array(list(fp.ToBitString())).astype(float)
return features
X = np.array([featurize_smiles(smiles) for smiles in df_train['SMILES']])
X.shape
X = pd.DataFrame(X)
X['Solubility'] = df_train['Solubility']
from rdkit.Chem import MolFromSmiles, Descriptors, MolFromSmarts
def create_descriptors(smiles):
mol = MolFromSmiles(smiles)
mw = Descriptors.MolWt(mol)
logp = Descriptors.MolLogP(mol)
rotb = Descriptors.NumRotatableBonds(mol)
arom_proportion = len(mol.GetSubstructMatches(MolFromSmarts('a'))) / Descriptors.HeavyAtomCount(mol)
hbd = Descriptors.NumHDonors(mol)
hba = Descriptors.NumHAcceptors(mol)
non_carbon_proportion = len(mol.GetSubstructMatches(MolFromSmarts('[!#6]'))) / Descriptors.HeavyAtomCount(mol)
psa = Descriptors.TPSA(mol, includeSandP=True)
fsp3 = Descriptors.FractionCSP3(mol)
return (mw, logp, rotb, arom_proportion, hbd, hba, non_carbon_proportion, psa, fsp3)
descriptors = [create_descriptors(smiles) for smiles in df_train['SMILES']]
col_names = ['mw', 'logp', 'rotb', 'ap', 'hbd', 'hba', 'non_cp', 'psa', 'fsp3']
train = pd.DataFrame(descriptors, columns=col_names)
train = train.join(df_train['Solubility'], how='left')
additional_col = ['MolMR', 'NumValenceElectrons', 'NumAromaticRings', 'NumSaturatedRings', 'NumAliphaticRings', 'RingCount', 'LabuteASA', 'BalabanJ', 'BertzCT']
df = train.join(df_train[additional_col], how='left')
reg = setup(df, target='Solubility', session_id=123, train_size=0.8, numeric_features=list(df.drop(columns=['Solubility']).columns), transformation=True, normalize=True)
best = create_model('et')
final_model = finalize_model(best)
descriptors = [create_descriptors(smiles) for smiles in df_test['SMILES']]
col_names = ['mw', 'logp', 'rotb', 'ap', 'hbd', 'hba', 'non_cp', 'psa', 'fsp3']
test = pd.DataFrame(descriptors, columns=col_names)
df = test.join(df_test[additional_col], how='left')
df = predict_model(final_model, data=df)
df.rename(columns={'prediction_label': 'Solubility'}, inplace=True)
smiles = ['CCC']
featurizer = dc.feat.Mol2VecFingerprint()
features = featurizer.featurize(smiles)
features = featurizer.featurize(df_train['SMILES'])
features.shape
df = pd.DataFrame(features)
df['Solubility'] = df_train['Solubility']
smiles = ['CC(=O)OC1=CC=CC=C1C(=O)O']
featurizer = dc.feat.RDKitDescriptors()
features = featurizer.featurize(smiles)
features.shape
features = featurizer.featurize(df_train['SMILES'])
df = pd.DataFrame(features)
df['Solubility'] = df_train['Solubility']
smiles = ['CC(=O)OC1=CC=CC=C1C(=O)O']
featurizer = dc.feat.MordredDescriptors()
features = featurizer.featurize(smiles)
features = featurizer.featurize(df_train['SMILES'])
features.shape
df = pd.DataFrame(features)
df['Solubility'] = df_train['Solubility']
reg = setup(df, target='Solubility', session_id=123, train_size=0.8) | code |
128047328/cell_3 | [
"text_plain_output_1.png"
] | pip install pycaret | code |
128047328/cell_66 | [
"application_vnd.jupyter.stderr_output_27.png",
"application_vnd.jupyter.stderr_output_35.png",
"application_vnd.jupyter.stderr_output_24.png",
"application_vnd.jupyter.stderr_output_16.png",
"application_vnd.jupyter.stderr_output_9.png",
"application_vnd.jupyter.stderr_output_32.png",
"application_vnd.jupyter.stderr_output_2.png",
"application_vnd.jupyter.stderr_output_7.png",
"application_vnd.jupyter.stderr_output_11.png",
"application_vnd.jupyter.stderr_output_18.png",
"application_vnd.jupyter.stderr_output_4.png",
"application_vnd.jupyter.stderr_output_26.png",
"application_vnd.jupyter.stderr_output_6.png",
"application_vnd.jupyter.stderr_output_31.png",
"application_vnd.jupyter.stderr_output_33.png",
"application_vnd.jupyter.stderr_output_25.png",
"application_vnd.jupyter.stderr_output_12.png",
"application_vnd.jupyter.stderr_output_8.png",
"application_vnd.jupyter.stderr_output_10.png",
"application_vnd.jupyter.stderr_output_23.png",
"application_vnd.jupyter.stderr_output_34.png",
"application_vnd.jupyter.stderr_output_19.png",
"application_vnd.jupyter.stderr_output_13.png",
"application_vnd.jupyter.stderr_output_3.png",
"application_vnd.jupyter.stderr_output_5.png",
"application_vnd.jupyter.stderr_output_30.png",
"application_vnd.jupyter.stderr_output_15.png",
"application_vnd.jupyter.stderr_output_17.png",
"application_vnd.jupyter.stderr_output_28.png",
"application_vnd.jupyter.stderr_output_20.png",
"application_vnd.jupyter.stderr_output_22.png",
"application_vnd.jupyter.stderr_output_29.png",
"application_vnd.jupyter.stderr_output_1.png",
"application_vnd.jupyter.stderr_output_14.png",
"application_vnd.jupyter.stderr_output_21.png"
] | from rdkit import Chem
from rdkit.Chem import AllChem
from rdkit.Chem import MolFromSmiles, Descriptors, MolFromSmarts
import deepchem as dc
import numpy as np
import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
smiles_list = df_train['SMILES'][:10]
mol_list = []
for smile in smiles_list:
mol = Chem.MolFromSmiles(smile)
mol_list.append(mol)
img = Chem.Draw.MolsToGridImage(mol_list, molsPerRow=5)
img
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
# Define a function to featurize a SMILES string
def featurize_smiles(smiles):
mol = Chem.MolFromSmiles(smiles)
fp = AllChem.GetMorganFingerprintAsBitVect(mol, 3, nBits=1024)
features = np.array(list(fp.ToBitString())).astype(float)
return features
X = np.array([featurize_smiles(smiles) for smiles in df_train['SMILES']])
X.shape
X = pd.DataFrame(X)
X['Solubility'] = df_train['Solubility']
from rdkit.Chem import MolFromSmiles, Descriptors, MolFromSmarts
def create_descriptors(smiles):
mol = MolFromSmiles(smiles)
mw = Descriptors.MolWt(mol)
logp = Descriptors.MolLogP(mol)
rotb = Descriptors.NumRotatableBonds(mol)
arom_proportion = len(mol.GetSubstructMatches(MolFromSmarts('a'))) / Descriptors.HeavyAtomCount(mol)
hbd = Descriptors.NumHDonors(mol)
hba = Descriptors.NumHAcceptors(mol)
non_carbon_proportion = len(mol.GetSubstructMatches(MolFromSmarts('[!#6]'))) / Descriptors.HeavyAtomCount(mol)
psa = Descriptors.TPSA(mol, includeSandP=True)
fsp3 = Descriptors.FractionCSP3(mol)
return (mw, logp, rotb, arom_proportion, hbd, hba, non_carbon_proportion, psa, fsp3)
descriptors = [create_descriptors(smiles) for smiles in df_train['SMILES']]
col_names = ['mw', 'logp', 'rotb', 'ap', 'hbd', 'hba', 'non_cp', 'psa', 'fsp3']
train = pd.DataFrame(descriptors, columns=col_names)
train = train.join(df_train['Solubility'], how='left')
additional_col = ['MolMR', 'NumValenceElectrons', 'NumAromaticRings', 'NumSaturatedRings', 'NumAliphaticRings', 'RingCount', 'LabuteASA', 'BalabanJ', 'BertzCT']
df = train.join(df_train[additional_col], how='left')
reg = setup(df, target='Solubility', session_id=123, train_size=0.8, numeric_features=list(df.drop(columns=['Solubility']).columns), transformation=True, normalize=True)
best = create_model('et')
final_model = finalize_model(best)
descriptors = [create_descriptors(smiles) for smiles in df_test['SMILES']]
col_names = ['mw', 'logp', 'rotb', 'ap', 'hbd', 'hba', 'non_cp', 'psa', 'fsp3']
test = pd.DataFrame(descriptors, columns=col_names)
df = test.join(df_test[additional_col], how='left')
df = predict_model(final_model, data=df)
df.rename(columns={'prediction_label': 'Solubility'}, inplace=True)
smiles = ['CCC']
featurizer = dc.feat.Mol2VecFingerprint()
features = featurizer.featurize(smiles)
features = featurizer.featurize(df_train['SMILES'])
features.shape
df = pd.DataFrame(features)
df['Solubility'] = df_train['Solubility']
smiles = ['CC(=O)OC1=CC=CC=C1C(=O)O']
featurizer = dc.feat.RDKitDescriptors()
features = featurizer.featurize(smiles)
features.shape
features = featurizer.featurize(df_train['SMILES'])
df = pd.DataFrame(features)
df['Solubility'] = df_train['Solubility']
reg = setup(df, target='Solubility', session_id=123, train_size=0.8) | code |
128047328/cell_17 | [
"text_plain_output_1.png"
] | best_top3_models = compare_models(n_select=3) | code |
128047328/cell_35 | [
"text_html_output_1.png"
] | from rdkit import Chem
from rdkit.Chem import AllChem
from rdkit.Chem import MolFromSmiles, Descriptors, MolFromSmarts
import numpy as np
import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
smiles_list = df_train['SMILES'][:10]
mol_list = []
for smile in smiles_list:
mol = Chem.MolFromSmiles(smile)
mol_list.append(mol)
img = Chem.Draw.MolsToGridImage(mol_list, molsPerRow=5)
img
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
# Define a function to featurize a SMILES string
def featurize_smiles(smiles):
mol = Chem.MolFromSmiles(smiles)
fp = AllChem.GetMorganFingerprintAsBitVect(mol, 3, nBits=1024)
features = np.array(list(fp.ToBitString())).astype(float)
return features
X = np.array([featurize_smiles(smiles) for smiles in df_train['SMILES']])
X.shape
X = pd.DataFrame(X)
X['Solubility'] = df_train['Solubility']
from rdkit.Chem import MolFromSmiles, Descriptors, MolFromSmarts
def create_descriptors(smiles):
mol = MolFromSmiles(smiles)
mw = Descriptors.MolWt(mol)
logp = Descriptors.MolLogP(mol)
rotb = Descriptors.NumRotatableBonds(mol)
arom_proportion = len(mol.GetSubstructMatches(MolFromSmarts('a'))) / Descriptors.HeavyAtomCount(mol)
hbd = Descriptors.NumHDonors(mol)
hba = Descriptors.NumHAcceptors(mol)
non_carbon_proportion = len(mol.GetSubstructMatches(MolFromSmarts('[!#6]'))) / Descriptors.HeavyAtomCount(mol)
psa = Descriptors.TPSA(mol, includeSandP=True)
fsp3 = Descriptors.FractionCSP3(mol)
return (mw, logp, rotb, arom_proportion, hbd, hba, non_carbon_proportion, psa, fsp3)
descriptors = [create_descriptors(smiles) for smiles in df_train['SMILES']]
col_names = ['mw', 'logp', 'rotb', 'ap', 'hbd', 'hba', 'non_cp', 'psa', 'fsp3']
train = pd.DataFrame(descriptors, columns=col_names)
train = train.join(df_train['Solubility'], how='left')
additional_col = ['MolMR', 'NumValenceElectrons', 'NumAromaticRings', 'NumSaturatedRings', 'NumAliphaticRings', 'RingCount', 'LabuteASA', 'BalabanJ', 'BertzCT']
df = train.join(df_train[additional_col], how='left')
reg = setup(df, target='Solubility', session_id=123, train_size=0.8, numeric_features=list(df.drop(columns=['Solubility']).columns), transformation=True, normalize=True) | code |
128047328/cell_77 | [
"text_html_output_2.png",
"text_plain_output_1.png"
] | et = create_model('et') | code |
128047328/cell_43 | [
"text_html_output_1.png"
] | best = create_model('et')
bagged_ensemble_50estm = ensemble_model(best, n_estimators=50) | code |
128047328/cell_31 | [
"text_html_output_2.png",
"text_plain_output_1.png"
] | from rdkit import Chem
from rdkit.Chem import AllChem
from rdkit.Chem import MolFromSmiles, Descriptors, MolFromSmarts
import numpy as np
import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
smiles_list = df_train['SMILES'][:10]
mol_list = []
for smile in smiles_list:
mol = Chem.MolFromSmiles(smile)
mol_list.append(mol)
img = Chem.Draw.MolsToGridImage(mol_list, molsPerRow=5)
img
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
# Define a function to featurize a SMILES string
def featurize_smiles(smiles):
mol = Chem.MolFromSmiles(smiles)
fp = AllChem.GetMorganFingerprintAsBitVect(mol, 3, nBits=1024)
features = np.array(list(fp.ToBitString())).astype(float)
return features
X = np.array([featurize_smiles(smiles) for smiles in df_train['SMILES']])
X.shape
X = pd.DataFrame(X)
X['Solubility'] = df_train['Solubility']
from rdkit.Chem import MolFromSmiles, Descriptors, MolFromSmarts
def create_descriptors(smiles):
mol = MolFromSmiles(smiles)
mw = Descriptors.MolWt(mol)
logp = Descriptors.MolLogP(mol)
rotb = Descriptors.NumRotatableBonds(mol)
arom_proportion = len(mol.GetSubstructMatches(MolFromSmarts('a'))) / Descriptors.HeavyAtomCount(mol)
hbd = Descriptors.NumHDonors(mol)
hba = Descriptors.NumHAcceptors(mol)
non_carbon_proportion = len(mol.GetSubstructMatches(MolFromSmarts('[!#6]'))) / Descriptors.HeavyAtomCount(mol)
psa = Descriptors.TPSA(mol, includeSandP=True)
fsp3 = Descriptors.FractionCSP3(mol)
return (mw, logp, rotb, arom_proportion, hbd, hba, non_carbon_proportion, psa, fsp3)
descriptors = [create_descriptors(smiles) for smiles in df_train['SMILES']]
col_names = ['mw', 'logp', 'rotb', 'ap', 'hbd', 'hba', 'non_cp', 'psa', 'fsp3']
train = pd.DataFrame(descriptors, columns=col_names)
train = train.join(df_train['Solubility'], how='left')
reg = setup(train, target='Solubility', session_id=123, train_size=0.8, transformation=True, normalize=True) | code |
128047328/cell_46 | [
"text_html_output_2.png",
"text_plain_output_1.png"
] | best_top3_models = compare_models(n_select=3)
best_top3_models = compare_models(n_select=3)
best_top3_models = compare_models(n_select=3)
best_top3_models = compare_models(n_select=3)
stacker = stack_models(best_top3_models) | code |
128047328/cell_14 | [
"text_plain_output_5.png",
"text_plain_output_4.png",
"text_plain_output_6.png",
"text_plain_output_3.png",
"text_plain_output_7.png",
"text_plain_output_2.png",
"text_plain_output_1.png"
] | import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
df_train.columns | code |
128047328/cell_22 | [
"text_html_output_2.png",
"text_plain_output_1.png"
] | from rdkit import Chem
from rdkit.Chem import AllChem
import numpy as np
import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
smiles_list = df_train['SMILES'][:10]
mol_list = []
for smile in smiles_list:
mol = Chem.MolFromSmiles(smile)
mol_list.append(mol)
img = Chem.Draw.MolsToGridImage(mol_list, molsPerRow=5)
img
df_train.columns
X = df_train.drop(columns=['ID', 'Name', 'InChI', 'InChIKey', 'SMILES', 'SD', 'Ocurrences', 'Group', 'comp_id'])
# Define a function to featurize a SMILES string
def featurize_smiles(smiles):
mol = Chem.MolFromSmiles(smiles)
fp = AllChem.GetMorganFingerprintAsBitVect(mol, 3, nBits=1024)
features = np.array(list(fp.ToBitString())).astype(float)
return features
X = np.array([featurize_smiles(smiles) for smiles in df_train['SMILES']])
X | code |
128047328/cell_10 | [
"text_plain_output_5.png",
"text_plain_output_9.png",
"text_plain_output_4.png",
"text_plain_output_13.png",
"text_plain_output_14.png",
"text_plain_output_10.png",
"text_plain_output_6.png",
"text_plain_output_3.png",
"text_plain_output_7.png",
"text_plain_output_8.png",
"text_plain_output_2.png",
"text_plain_output_1.png",
"text_plain_output_11.png",
"text_plain_output_12.png"
] | import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
df_train.info() | code |
128047328/cell_37 | [
"text_html_output_2.png",
"text_plain_output_1.png"
] | best = create_model('et') | code |
128047328/cell_12 | [
"text_plain_output_100.png",
"text_plain_output_84.png",
"text_plain_output_56.png",
"text_plain_output_35.png",
"text_plain_output_98.png",
"text_plain_output_43.png",
"text_plain_output_78.png",
"text_plain_output_106.png",
"text_plain_output_37.png",
"text_plain_output_90.png",
"text_plain_output_79.png",
"text_plain_output_5.png",
"text_plain_output_75.png",
"text_plain_output_48.png",
"text_plain_output_30.png",
"text_plain_output_73.png",
"text_plain_output_15.png",
"text_plain_output_70.png",
"text_plain_output_9.png",
"text_plain_output_44.png",
"text_plain_output_86.png",
"text_plain_output_40.png",
"text_plain_output_74.png",
"text_plain_output_31.png",
"text_plain_output_20.png",
"text_plain_output_102.png",
"text_plain_output_101.png",
"text_plain_output_60.png",
"text_plain_output_68.png",
"text_plain_output_4.png",
"text_plain_output_65.png",
"text_plain_output_64.png",
"text_plain_output_13.png",
"text_plain_output_107.png",
"text_plain_output_52.png",
"text_plain_output_66.png",
"text_plain_output_45.png",
"text_plain_output_14.png",
"text_plain_output_32.png",
"text_plain_output_88.png",
"text_plain_output_29.png",
"text_plain_output_58.png",
"text_plain_output_49.png",
"text_plain_output_63.png",
"text_plain_output_27.png",
"text_plain_output_76.png",
"text_plain_output_108.png",
"text_plain_output_54.png",
"text_plain_output_10.png",
"text_plain_output_6.png",
"text_plain_output_92.png",
"text_plain_output_57.png",
"text_plain_output_24.png",
"text_plain_output_21.png",
"text_plain_output_104.png",
"text_plain_output_47.png",
"text_plain_output_25.png",
"text_plain_output_77.png",
"text_plain_output_18.png",
"text_plain_output_50.png",
"text_plain_output_36.png",
"text_plain_output_96.png",
"text_plain_output_87.png",
"text_plain_output_3.png",
"text_plain_output_22.png",
"text_plain_output_81.png",
"text_plain_output_69.png",
"text_plain_output_38.png",
"text_plain_output_7.png",
"text_plain_output_91.png",
"text_plain_output_16.png",
"text_plain_output_59.png",
"text_plain_output_103.png",
"text_plain_output_71.png",
"text_plain_output_8.png",
"text_plain_output_26.png",
"text_plain_output_109.png",
"text_plain_output_41.png",
"text_plain_output_34.png",
"text_plain_output_85.png",
"text_plain_output_42.png",
"text_plain_output_67.png",
"text_plain_output_53.png",
"text_plain_output_23.png",
"text_plain_output_89.png",
"text_plain_output_51.png",
"text_plain_output_28.png",
"text_plain_output_72.png",
"text_plain_output_99.png",
"text_plain_output_2.png",
"text_plain_output_97.png",
"text_plain_output_1.png",
"text_plain_output_33.png",
"text_plain_output_39.png",
"text_plain_output_55.png",
"text_plain_output_82.png",
"text_plain_output_93.png",
"text_plain_output_19.png",
"text_plain_output_105.png",
"text_plain_output_80.png",
"text_plain_output_94.png",
"text_plain_output_17.png",
"text_plain_output_11.png",
"text_plain_output_12.png",
"text_plain_output_62.png",
"text_plain_output_95.png",
"text_plain_output_61.png",
"text_plain_output_83.png",
"text_plain_output_46.png"
] | from rdkit import Chem
import pandas as pd
df_train = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/train.csv')
df_test = pd.read_csv('/kaggle/input/aqueous-solubility-predictioin/test.csv')
df_train.shape
smiles_list = df_train['SMILES'][:10]
mol_list = []
for smile in smiles_list:
mol = Chem.MolFromSmiles(smile)
mol_list.append(mol)
img = Chem.Draw.MolsToGridImage(mol_list, molsPerRow=5)
img | code |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.