def test_calculation_positions():
"""Checks algorithms calculate positions and returns."""
list_of_algos = Api.available_algorithms(filter_by_category=PandasEnum.ALLOCATION.value)
for ii_df in test_dfs:
for jj_algo_name in list_of_algos:
df_with_allocations = Api.calculate_allocations(ii_df, jj_algo_name, "close")
isinstance(df_with_allocations, pd.DataFrame)
df_with_returns = Api.calculate_returns(df_with_allocations)
isinstance(df_with_returns, pd.DataFrame)
for ii_value in df_with_returns[PandasEnum.VALUATION.value]:
if not isinstance(ii_value, float):
assert ii_value is "NaN"
def test_specific_rule():
"""Checks using SMA."""
from infertrade.data.simulate_data import simulated_market_data_4_years_gen
data = simulated_market_data_4_years_gen()
sma_signal = Api.calculate_signal(data, "SMA")
assert isinstance(sma_signal, pd.DataFrame)
def test_signals_creation():
"""Checks signal algorithms can create a signal in a Pandas dataframe."""
list_of_algos = Api.available_algorithms(
filter_by_category=PandasEnum.SIGNAL.value)
for ii_df in test_dfs:
for jj_algo_name in list_of_algos:
original_columns = ii_df.columns
df_with_signal = Api.calculate_signal(ii_df, jj_algo_name)
assert isinstance(df_with_signal, pd.DataFrame)
# Signal algorithms should be adding new columns with float, int or NaN data.
new_columns = False
for ii_column_name in df_with_signal:
if ii_column_name not in original_columns:
new_columns = True
for ii_value in df_with_signal[ii_column_name]:
if not isinstance(ii_value, (float, int)):
assert ii_value is "NaN"
# At least one new column should have been added.
assert new_columns
def test_signals_creation():
"""Checks signal algorithms can create a signal in a Pandas dataframe."""
list_of_algos = Api.available_algorithms(filter_by_category=PandasEnum.SIGNAL.value)
for ii_df in test_dfs:
for jj_algo_name in list_of_algos:
original_columns = ii_df.columns
# We check if the test series has the columns needed for the rule to calculate.
required_columns = Api.required_inputs_for_algorithm(jj_algo_name)
all_present = True
for ii_requirement in required_columns:
if ii_requirement not in ii_df.columns:
all_present = False
# If columns are missing, we anticipate a KeyError will trigger.
if not all_present:
with pytest.raises(KeyError):
Api.calculate_signal(ii_df, jj_algo_name)
return True
# Otherwise we expect to parse successfully.
df_with_signal = Api.calculate_signal(ii_df, jj_algo_name)
if not isinstance(df_with_signal, pd.DataFrame):
print(df_with_signal)
print("Type was: ", type(df_with_signal))
raise TypeError("Bad output format.")
# Signal algorithms should be adding new columns with float, int or NaN data.
new_columns = False
for ii_column_name in df_with_signal:
if ii_column_name not in original_columns:
new_columns = True
for ii_value in df_with_signal[ii_column_name]:
if not isinstance(ii_value, (float, int)):
assert ii_value is "NaN"
# At least one new column should have been added.
assert new_columns
def test_get_available_algorithms():
"""Checks can get algorithm list and that returned algorithms can supply all expected properties."""
list_of_algos = Api.available_algorithms()
assert isinstance(list_of_algos, list)
for ii_algo_name in list_of_algos:
assert isinstance(ii_algo_name, str)
assert Api.return_algorithm_category(ii_algo_name) in Api.algorithm_categories()
assert Api.determine_package_of_algorithm(ii_algo_name) in Api.available_packages()
inputs = Api.required_inputs_for_algorithm(ii_algo_name)
assert isinstance(inputs, list)
for ii_required_input in inputs:
assert isinstance(ii_required_input, str)
params = Api.required_parameters_for_algorithm(ii_algo_name)
assert isinstance(params, dict)
for ii_param_name in params:
assert isinstance(ii_param_name, str)
assert isinstance(params[ii_param_name], (int, float))
def test_get_ta_rules():
"""Checks we can get rules from ta."""
available_ta_algos = Api().available_algorithms(
filter_by_package=name_of_ta_package)
# Check there are some algorithms.
assert available_ta_algos
def test_ta_in_package_list():
"""Checks ta is in the package list"""
packages = Api().available_packages()
assert name_of_ta_package in packages
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
Created by: Thomas Oliver
Created date: 25th March 2021
"""
import pandas as pd
import pytest
from infertrade.PandasEnum import PandasEnum
from infertrade.api import Api
from infertrade.data.simulate_data import simulated_market_data_4_years_gen
api_instance = Api()
test_dfs = [simulated_market_data_4_years_gen(), simulated_market_data_4_years_gen()]
def test_get_available_algorithms():
"""Checks can get algorithm list and that returned algorithms can supply all expected properties."""
list_of_algos = Api.available_algorithms()
assert isinstance(list_of_algos, list)
for ii_algo_name in list_of_algos:
assert isinstance(ii_algo_name, str)
assert Api.return_algorithm_category(ii_algo_name) in Api.algorithm_categories()
assert Api.determine_package_of_algorithm(ii_algo_name) in Api.available_packages()
inputs = Api.required_inputs_for_algorithm(ii_algo_name)
assert isinstance(inputs, list)
for ii_required_input in inputs: