holmesal

plt.plot(x, y, 'go', label = 'from data', alpha = .5)
plt.legend()
plt.show()
plt.clf()

holmesal

# stick our data in a dataframe:
df = pd.DataFrame({'x': x, 'y': y});

# use 20% of our data for validation:
valid_idx = random.sample(list(df.index), int(0.2 * len(df)));

# what we're predicting:
dep_var = 'y'

# we don't have any categorical variables:

holmesal

model = LinearRegressionModel();
print(model);

holmesal

learn = Learner(data, model);

holmesal

model = TabularModel(emb_szs={}, n_cont=1, out_sz=1, layers=[10]);
print(model)

holmesal

preds, targs = learn.get_preds()
x_val = x[valid_idx];
y_val = np.array(targs);
y_hat = np.array(preds.flatten())

plt.plot(x_val, y_val, 'go', label = 'tagets', alpha = .5)
plt.plot(x_val, y_hat, 'ro', label = 'preds', alpha = .5)
plt.legend()
plt.show()

holmesal

import {
    GraphQLBoolean,
    GraphQLError,
    GraphQLList,
    GraphQLNonNull,
    GraphQLString
} from 'graphql';

import {
    fromGlobalId,

holmesal

holmesal

extend type Query {
  __schema: __Schema!
  __type(name: String!): __Type
}

type __Schema {
  description: String
  types: [__Type!]!
  queryType: __Type!
  mutationType: __Type

holmesal