Fixed catalyst tests except example tests

.travis.yml

@@ -41,7 +41,7 @@ install:
   - if [[ "$TRAVIS_SECURE_ENV_VARS" = "true" && "$TRAVIS_BRANCH" = "master" && "$TRAVIS_PULL_REQUEST" = "false" ]]; then DO_UPLOAD="true"; else DO_UPLOAD="false"; fi
   - |
     for recipe in $(ls -d conda/*/ | xargs -I {} basename {}); do
-      if [[ "$recipe" = "zipline" ]]; then continue; fi
+      if [[ "$recipe" = "catalyst" ]]; then continue; fi
 
       conda build conda/$recipe --python=$CONDA_PY --numpy=$CONDA_NPY --skip-existing -c quantopian -c quantopian/label/ci
       RECIPE_OUTPUT=$(conda build conda/$recipe --python=$CONDA_PY --numpy=$CONDA_NPY --output)
@@ -61,14 +61,14 @@ install:
 before_script:
   - pip freeze | sort
 script:
-  - flake8 zipline tests
+  - flake8 catalyst tests
   - nosetests --with-coverage
   # deactive env to get access to anaconda command
   - source deactivate
 
   # unshallow the clone so the conda build can clone it.
   - git fetch --unshallow
-  - exec 3>&1; ZP_OUT=$(conda build conda/zipline --python=$CONDA_PY --numpy=$CONDA_NPY -c quantopian -c quantopian/label/ci | tee >(cat - >&3))
+  - exec 3>&1; ZP_OUT=$(conda build conda/catalyst --python=$CONDA_PY --numpy=$CONDA_NPY -c quantopian -c quantopian/label/ci | tee >(cat - >&3))
   - ZP_OUTPUT=$(echo "$ZP_OUT" | grep "anaconda upload" | awk '{print $NF}')
   - if [[ "$DO_UPLOAD" = "true" ]]; then anaconda -t $ANACONDA_TOKEN upload $ZP_OUTPUT -u quantopian --label ci; fi
   # reactivate env (necessary for coveralls)

catalyst/algorithm.py

@@ -1115,7 +1115,7 @@ def schedule_function(self,
         if calendar is None:
             cal = self.trading_calendar
         elif calendar is calendars.CRYPTO_ASSETS:
-            cal = get_environment('OPEN')
+            cal = get_calendar('OPEN')
         elif calendar is calendars.US_EQUITIES:
             cal = get_calendar('NYSE')
         elif calendar is calendars.US_FUTURES:
@@ -1124,7 +1124,9 @@ def schedule_function(self,
             raise ScheduleFunctionInvalidCalendar(
                 given_calendar=calendar,
                 allowed_calendars=(
-                    '[calendars.CRYPTO_ASSETS, calendars.US_EQUITIES, calendars.US_FUTURES]'
+                    '[calendars.CRYPTO_ASSETS, '
+                    'calendars.US_EQUITIES, '
+                    'calendars.US_FUTURES]'
                 ),
             )
 

catalyst/data/benchmarks.py

@@ -43,6 +43,9 @@ def get_benchmark_returns(symbol, first_date, last_date):
     first_date is **not** included because we need the close from day N - 1 to
     compute the returns for day N.
     """
+    if symbol == '^GSPC':
+        symbol = 'spy'
+
     data = pd_reader.DataReader(
         symbol,
         'google',

catalyst/data/bundles/quandl.py

@@ -412,6 +412,7 @@ def quantopian_quandl_bundle(environ,
     'https://s3.amazonaws.com/quantopian-public-zipline-data/quandl'
 )
 
+
 @bundles.register(
     'catalyst',
     calendar_name='NYSE',

catalyst/data/us_equity_pricing.py

@@ -67,7 +67,10 @@
     preprocess,
     verify_indices_all_unique,
 )
-from catalyst.utils.sqlite_utils import group_into_chunks, coerce_string_to_conn
+from catalyst.utils.sqlite_utils import (
+    group_into_chunks,
+    coerce_string_to_conn,
+)
 from catalyst.utils.memoize import lazyval
 from catalyst.utils.cli import maybe_show_progress
 from ._equities import _compute_row_slices, _read_bcolz_data

catalyst/examples/momentum_pipeline.py

@@ -10,26 +10,23 @@
     pipeline_output,
     record,
     schedule_function,
+    symbol,
 )
 from catalyst.pipeline import Pipeline
-from catalyst.pipeline.factors.crypto import RSI as cRSI
-from catalyst.pipeline.factors.equity import RSI as eRSI
+from catalyst.pipeline.factors.equity import RSI
 
 
 def make_pipeline():
-    crsi = cRSI()
-    ersi = eRSI()
+    rsi = RSI()
     return Pipeline(
         columns={
-            'longs': crsi.top(3),
-            'shorts': crsi.bottom(3),
-            'equity': ersi.top(3),
+            'longs': rsi.top(3),
+            'shorts': rsi.bottom(3),
         },
     )
 
 
 def rebalance(context, data):
-
     # Pipeline data will be a dataframe with boolean columns named 'longs' and
     # 'shorts'.
     pipeline_data = context.pipeline_data

catalyst/pipeline/api_utils.py

@@ -10,8 +10,9 @@ def restrict_to_dtype(dtype, message_template):
     Terms with a specific dtype.
 
     This is conceptually similar to
-    catalyst.utils.input_validation.expect_dtypes, but provides more flexibility
-    for providing error messages that are specifically targeting Term methods.
+    catalyst.utils.input_validation.expect_dtypes, but provides more
+    flexibility for providing error messages that are specifically targeting
+    Term methods.
 
     Parameters
     ----------

catalyst/pipeline/classifiers/classifier.py

@@ -330,7 +330,8 @@ def to_workspace_value(self, result, assets):
         Called with the result of a pipeline. This needs to return an object
         which can be put into the workspace to continue doing computations.
 
-        This is the inverse of :func:`~catalyst.pipeline.term.Term.postprocess`.
+        This is the inverse of
+        :func:`~catalyst.pipeline.term.Term.postprocess`.
         """
         if self.dtype == int64_dtype:
             return super(Classifier, self).to_workspace_value(result, assets)

catalyst/pipeline/data/crypto_pricing.py

@@ -8,8 +8,8 @@ class CryptoPricing(DataSet):
     """
     Dataset representing daily trading prices and volumes of crypto-assets.
     """
-    open   = Column(float64_dtype)
-    high   = Column(float64_dtype)
-    low    = Column(float64_dtype)
-    close  = Column(float64_dtype)
+    open = Column(float64_dtype)
+    high = Column(float64_dtype)
+    low = Column(float64_dtype)
+    close = Column(float64_dtype)
     volume = Column(float64_dtype)

catalyst/pipeline/data/dataset.py

@@ -7,7 +7,10 @@
     with_metaclass,
 )
 
-from catalyst.pipeline.classifiers import Classifier, Latest as LatestClassifier
+from catalyst.pipeline.classifiers import (
+    Classifier,
+    Latest as LatestClassifier,
+)
 from catalyst.pipeline.factors import Factor, Latest as LatestFactor
 from catalyst.pipeline.filters import Filter, Latest as LatestFilter
 from catalyst.pipeline.sentinels import NotSpecified

catalyst/pipeline/data/equity_pricing.py

@@ -8,8 +8,8 @@ class USEquityPricing(DataSet):
     """
     Dataset representing daily trading prices and volumes.
     """
-    open   = Column(float64_dtype)
-    high   = Column(float64_dtype)
-    low    = Column(float64_dtype)
-    close  = Column(float64_dtype)
+    open = Column(float64_dtype)
+    high = Column(float64_dtype)
+    low = Column(float64_dtype)
+    close = Column(float64_dtype)
     volume = Column(float64_dtype)

catalyst/pipeline/factors/equity/__init__.py

@@ -1,6 +1,9 @@
 from .statistical import (
+    RollingPearson,
+    RollingLinearRegression,
     RollingLinearRegressionOfReturns,
     RollingPearsonOfReturns,
+    RollingSpearman,
     RollingSpearmanOfReturns,
 )
 from .technical import (

catalyst/pipeline/factors/factor.py

@@ -338,8 +338,9 @@ class Factor(RestrictedDTypeMixin, ComputableTerm):
         >>> f2 = SomeOtherFactor(...)  # doctest: +SKIP
         >>> average = (f1 + f2) / 2.0  # doctest: +SKIP
 
-    Factors can also be converted into :class:`catalyst.pipeline.Filter` objects
-    via comparison operators: (``<``, ``<=``, ``!=``, ``eq``, ``>``, ``>=``).
+    Factors can also be converted into :class:`catalyst.pipeline.Filter`
+    objects via comparison operators:
+    (``<``, ``<=``, ``!=``, ``eq``, ``>``, ``>=``).
 
     There are many natural operators defined on Factors besides the basic
     numerical operators. These include methods identifying missing or
@@ -699,7 +700,7 @@ def pearsonr(self, target, correlation_length, mask=NotSpecified):
         :class:`catalyst.pipeline.factors.RollingPearsonOfReturns`
         :meth:`Factor.spearmanr`
         """
-        from .statistical import RollingPearson
+        from .equity import RollingPearson
         return RollingPearson(
             base_factor=self,
             target=target,
@@ -764,7 +765,7 @@ def spearmanr(self, target, correlation_length, mask=NotSpecified):
         :class:`catalyst.pipeline.factors.RollingSpearmanOfReturns`
         :meth:`Factor.pearsonr`
         """
-        from .statistical import RollingSpearman
+        from .equity import RollingSpearman
         return RollingSpearman(
             base_factor=self,
             target=target,
@@ -826,7 +827,7 @@ def linear_regression(self, target, regression_length, mask=NotSpecified):
         :func:`scipy.stats.linregress`
         :class:`catalyst.pipeline.factors.RollingLinearRegressionOfReturns`
         """
-        from .statistical import RollingLinearRegression
+        from .equity import RollingLinearRegression
         return RollingLinearRegression(
             dependent=self,
             independent=target,

catalyst/pipeline/filters/filter.py

@@ -141,7 +141,7 @@ class Filter(RestrictedDTypeMixin, ComputableTerm):
     construct a Factor computing 10-day VWAP and compare it to the scalar value
     20.0::
 
-        >>> from catalyst.pipeline.factors import VWAP
+        >>> from catalyst.pipeline.factors.equity import VWAP
         >>> vwap_10 = VWAP(window_length=10)
         >>> vwaps_under_20 = (vwap_10 <= 20)
 

catalyst/pipeline/loaders/blaze/estimates.py

@@ -130,6 +130,10 @@ def load_adjusted_array(self, columns, dates, assets, mask):
             mask,
         )
 
+    @property
+    def columns(self):
+        return self._columns
+
 
 class BlazeNextEstimatesLoader(BlazeEstimatesLoader):
     loader = NextEarningsEstimatesLoader

catalyst/pipeline/loaders/blaze/events.py

@@ -91,6 +91,8 @@ def __init__(self,
         self._data_query_time = data_query_time
         self._data_query_tz = data_query_tz
 
+        self._columns = next_value_columns
+
     def load_adjusted_array(self, columns, dates, assets, mask):
         raw = load_raw_data(assets,
                             dates,
@@ -109,3 +111,7 @@ def load_adjusted_array(self, columns, dates, assets, mask):
             assets,
             mask,
         )
+
+    @property
+    def columns(self):
+        return self._columns

catalyst/pipeline/loaders/crypto_pricing_loader.py

@@ -16,10 +16,7 @@
     uint32,
 )
 
-from catalyst.data.us_equity_pricing import (
-    BcolzDailyBarReader,
-    SQLiteAdjustmentReader,
-)
+from catalyst.data.us_equity_pricing import BcolzDailyBarReader
 from catalyst.lib.adjusted_array import AdjustedArray
 from catalyst.errors import NoFurtherDataError
 from catalyst.utils.calendars import get_calendar

catalyst/pipeline/loaders/earnings_estimates.py

@@ -162,6 +162,7 @@ def __init__(self,
         }
 
         self.name_map = name_map
+        self._columns = set(name_map.keys())
 
     @abstractmethod
     def get_zeroth_quarter_idx(self, stacked_last_per_qtr):
@@ -664,6 +665,10 @@ def load_adjusted_array(self, columns, dates, assets, mask):
                 )
         return out
 
+    @property
+    def columns(self):
+        return self._columns
+
     def get_last_data_per_qtr(self, assets_with_data, columns, dates):
         """
         Determine the last piece of information we know for each column on each

catalyst/pipeline/loaders/equity_pricing_loader.py

@@ -95,7 +95,7 @@ def load_adjusted_array(self, columns, dates, assets, mask):
                 c.missing_value,
             )
         return out
-    
+
     @property
     def columns(self):
         return self._columns

catalyst/pipeline/loaders/events.py

@@ -115,6 +115,8 @@ def __init__(self,
         # Columns to load with self.load_previous_events.
         self.previous_value_columns = previous_value_columns
 
+        self._columns = events.columns
+
     def split_next_and_previous_event_columns(self, requested_columns):
         """
         Split requested columns into columns that should load the next known
@@ -229,3 +231,7 @@ def load_adjusted_array(self, columns, dates, sids, mask):
             self.load_next_events(n, dates, sids, mask),
             self.load_previous_events(p, dates, sids, mask),
         )
+
+    @property
+    def columns(self):
+        return self._columns

catalyst/pipeline/loaders/frame.py

@@ -64,6 +64,8 @@ def __init__(self, column, baseline, adjustments=None):
         self.dates = baseline.index
         self.assets = baseline.columns
 
+        self._columns = self.assets
+
         if adjustments is None:
             adjustments = DataFrame(
                 index=DatetimeIndex([]),
@@ -174,3 +176,8 @@ def load_adjusted_array(self, columns, dates, assets, mask):
                 missing_value=column.missing_value,
             ),
         }
+
+    @property
+    def columns(self):
+        return self._columns
+

catalyst/pipeline/loaders/synthetic.py

@@ -61,7 +61,9 @@ class PrecomputedLoader(PipelineLoader):
     """
     def __init__(self, constants, dates, sids):
         loaders = {}
+        columns = []
         for column, const in iteritems(constants):
+            columns.append(column)
             frame = DataFrame(
                 const,
                 index=dates,
@@ -75,6 +77,7 @@ def __init__(self, constants, dates, sids):
             )
 
         self._loaders = loaders
+        self._columns = set(columns)
 
     def load_adjusted_array(self, columns, dates, assets, mask):
         """
@@ -91,6 +94,9 @@ def load_adjusted_array(self, columns, dates, assets, mask):
             )
         return out
 
+    def columns(self):
+        return self._columns
+
 
 class EyeLoader(PrecomputedLoader):
     """
@@ -108,12 +114,17 @@ class EyeLoader(PrecomputedLoader):
     """
     def __init__(self, columns, dates, sids):
         shape = (len(dates), len(sids))
+        self._columns = columns
         super(EyeLoader, self).__init__(
             {column: eye(shape, dtype=column.dtype) for column in columns},
             dates,
             sids,
         )
 
+    @property
+    def columns(self):
+        return self._columns
+
 
 class SeededRandomLoader(PrecomputedLoader):
     """
@@ -133,6 +144,7 @@ class SeededRandomLoader(PrecomputedLoader):
 
     def __init__(self, seed, columns, dates, sids):
         self._seed = seed
+        self._columns = columns
         super(SeededRandomLoader, self).__init__(
             {c: self.values(c.dtype, dates, sids) for c in columns},
             dates,
@@ -151,6 +163,10 @@ def values(self, dtype, dates, sids):
             bool_dtype: self._bool_values,
             object_dtype: self._object_values,
         }[dtype](shape)
+
+    @property
+    def columns(self):
+        return self._columns
 
     @property
     def state(self):