def forward(self, tensor_x, tensor_y):
if tensor_x.shape[-1] == 1:
tensor_x = tensor_x.expand_as(tensor_y)
output = tsf.rolling_cov(tensor_x, tensor_y,
window=self._window) / tsf.rolling_var(
tensor_x, window=self._window)
return output
def forward(self, high_ts, low_ts, close_ts, volume_ts):
adv20 = tsf.rolling_mean_(volume_ts, 20)
df = tsf.rolling_cov(adv20, low_ts, 5)
alpha = tsf.rolling_scale(((df + ((high_ts + low_ts) / 2)) - close_ts))
return alpha
def forward(self, high_ts, volume_ts):
ts = tsf.rolling_cov(tsf.ts_rank(volume_ts, 5),
tsf.ts_rank(high_ts, 5), 5)
alpha = -1 * tsf.rolling_max(ts, 3)
return alpha
def forward(self, high_ts, close_ts, volume_ts):
ts = tsf.rolling_cov(high_ts, volume_ts, 5)
alpha = -1 * tsf.diff(ts, 5) * tsf.rank(tsf.rolling_std(close_ts, 20))
return alpha
def forward(self, open_ts, close_ts):
ts = tsf.rolling_cov(close_ts, open_ts, 10)
alpha = -1 * (tsf.rank((tsf.rolling_std(abs((close_ts - open_ts)), 5) +
(close_ts - open_ts)) + ts))
return alpha
def forward(self, high_ts, volume_ts):
alpha = -1 * tsf.rank(
tsf.rolling_cov(tsf.rank(high_ts), tsf.rank(volume_ts), 5))
return alpha
def forward(self, high_ts, volume_ts):
df = tsf.rolling_cov(tsf.rank(high_ts), tsf.rank(volume_ts), 3)
alpha = -1 * tsf.rolling_sum_(tsf.rank(df), 3)
return alpha
def forward(self, open_ts, volume_ts, returns_ts):
df = tsf.rolling_cov(open_ts, volume_ts, 10)
alpha = -1 * tsf.rank(tsf.diff(returns_ts, 3)) * df
return alpha
def forward(self, high_ts, volume_ts):
output_tensor = (-1 * tsf.rank(
tsf.rolling_cov(tsf.rank(high_ts), tsf.rank(volume_ts), 5)))
return output_tensor
