6181253459

Committed 08 Sep 2023 06:04AM UTC coverage: 94.614% (+0.06%) from 94.556%

Build # 6181253459

Build Type

push

github-actions

Committed by

web-flow

Commit Message

Merge pull request #9159 from stash86/fix-adjust

remove old codes when we only can do partial entries

Run Details

2 of 2 new or added lines in 1 file covered. (100.0%)

19114 of 20202 relevant lines covered (94.61%)

0.95 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

96.55

/freqtrade/freqai/prediction_models/PyTorchTransformerRegressor.py

from typing import Any, Dict, Tuple

import numpy as np
import numpy.typing as npt
import pandas as pd
import torch

from freqtrade.freqai.base_models.BasePyTorchRegressor import BasePyTorchRegressor
from freqtrade.freqai.data_kitchen import FreqaiDataKitchen
from freqtrade.freqai.torch.PyTorchDataConvertor import (DefaultPyTorchDataConvertor,
                                                         PyTorchDataConvertor)
from freqtrade.freqai.torch.PyTorchModelTrainer import PyTorchTransformerTrainer
from freqtrade.freqai.torch.PyTorchTransformerModel import PyTorchTransformerModel


class PyTorchTransformerRegressor(BasePyTorchRegressor):
    """
    This class implements the fit method of IFreqaiModel.
    in the fit method we initialize the model and trainer objects.
    the only requirement from the model is to be aligned to PyTorchRegressor
    predict method that expects the model to predict tensor of type float.
    the trainer defines the training loop.

    parameters are passed via `model_training_parameters` under the freqai
    section in the config file. e.g:
    {
        ...
        "freqai": {
            ...
            "model_training_parameters" : {
                "learning_rate": 3e-4,
                "trainer_kwargs": {
                    "n_steps": 5000,
                    "batch_size": 64,
                    "n_epochs": null
                },
                "model_kwargs": {
                    "hidden_dim": 512,
                    "dropout_percent": 0.2,
                    "n_layer": 1,
                },
            }
        }
    }
    """

    @property
    def data_convertor(self) -> PyTorchDataConvertor:
        return DefaultPyTorchDataConvertor(target_tensor_type=torch.float)

    def __init__(self, **kwargs) -> None:
        super().__init__(**kwargs)
        config = self.freqai_info.get("model_training_parameters", {})
        self.learning_rate: float = config.get("learning_rate",  3e-4)
        self.model_kwargs: Dict[str, Any] = config.get("model_kwargs",  {})
        self.trainer_kwargs: Dict[str, Any] = config.get("trainer_kwargs",  {})

    def fit(self, data_dictionary: Dict, dk: FreqaiDataKitchen, **kwargs) -> Any:
        """
        User sets up the training and test data to fit their desired model here
        :param data_dictionary: the dictionary holding all data for train, test,
            labels, weights
        :param dk: The datakitchen object for the current coin/model
        """

        n_features = data_dictionary["train_features"].shape[-1]
        n_labels = data_dictionary["train_labels"].shape[-1]
        model = PyTorchTransformerModel(
            input_dim=n_features,
            output_dim=n_labels,
            time_window=self.window_size,
            **self.model_kwargs
        )
        model.to(self.device)
        optimizer = torch.optim.AdamW(model.parameters(), lr=self.learning_rate)
        criterion = torch.nn.MSELoss()
        # check if continual_learning is activated, and retreive the model to continue training
        trainer = self.get_init_model(dk.pair)
        if trainer is None:
            trainer = PyTorchTransformerTrainer(
                model=model,
                optimizer=optimizer,
                criterion=criterion,
                device=self.device,
                data_convertor=self.data_convertor,
                window_size=self.window_size,
                tb_logger=self.tb_logger,
                **self.trainer_kwargs,
            )
        trainer.fit(data_dictionary, self.splits)
        return trainer

    def predict(
        self, unfiltered_df: pd.DataFrame, dk: FreqaiDataKitchen, **kwargs
    ) -> Tuple[pd.DataFrame, npt.NDArray[np.int_]]:
        """
        Filter the prediction features data and predict with it.
        :param unfiltered_df: Full dataframe for the current backtest period.
        :return:
        :pred_df: dataframe containing the predictions
        :do_predict: np.array of 1s and 0s to indicate places where freqai needed to remove
        data (NaNs) or felt uncertain about data (PCA and DI index)
        """

        dk.find_features(unfiltered_df)
        dk.data_dictionary["prediction_features"], _ = dk.filter_features(
            unfiltered_df, dk.training_features_list, training_filter=False
        )

        dk.data_dictionary["prediction_features"], outliers, _ = dk.feature_pipeline.transform(
            dk.data_dictionary["prediction_features"], outlier_check=True)

        x = self.data_convertor.convert_x(
            dk.data_dictionary["prediction_features"],
            device=self.device
        )
        # if user is asking for multiple predictions, slide the window
        # along the tensor
        x = x.unsqueeze(0)
        # create empty torch tensor
        self.model.model.eval()
        yb = torch.empty(0).to(self.device)
        if x.shape[1] > 1:
            ws = self.window_size
            for i in range(0, x.shape[1] - ws):
                xb = x[:, i:i + ws, :].to(self.device)
                y = self.model.model(xb)
                yb = torch.cat((yb, y), dim=0)
        else:
            yb = self.model.model(x)

        yb = yb.cpu().squeeze()
        pred_df = pd.DataFrame(yb.detach().numpy(), columns=dk.label_list)
        pred_df, _, _ = dk.label_pipeline.inverse_transform(pred_df)

        if self.freqai_info.get("DI_threshold", 0) > 0:
            dk.DI_values = dk.feature_pipeline["di"].di_values
        else:
            dk.DI_values = np.zeros(outliers.shape[0])
        dk.do_predict = outliers

        if x.shape[1] > 1:
            zeros_df = pd.DataFrame(np.zeros((x.shape[1] - len(pred_df), len(pred_df.columns))),
                                    columns=pred_df.columns)
            pred_df = pd.concat([zeros_df, pred_df], axis=0, ignore_index=True)
        return (pred_df, dk.do_predict)

1	from typing import Any, Dict, Tuple	1✔
2
3	import numpy as np	1✔
4	import numpy.typing as npt	1✔
5	import pandas as pd	1✔
6	import torch	1✔
7
8	from freqtrade.freqai.base_models.BasePyTorchRegressor import BasePyTorchRegressor	1✔
9	from freqtrade.freqai.data_kitchen import FreqaiDataKitchen	1✔
10	from freqtrade.freqai.torch.PyTorchDataConvertor import (DefaultPyTorchDataConvertor,	1✔
11	PyTorchDataConvertor)
12	from freqtrade.freqai.torch.PyTorchModelTrainer import PyTorchTransformerTrainer	1✔
13	from freqtrade.freqai.torch.PyTorchTransformerModel import PyTorchTransformerModel	1✔
14
15
16	class PyTorchTransformerRegressor(BasePyTorchRegressor):	1✔
17	"""
18	This class implements the fit method of IFreqaiModel.
19	in the fit method we initialize the model and trainer objects.
20	the only requirement from the model is to be aligned to PyTorchRegressor
21	predict method that expects the model to predict tensor of type float.
22	the trainer defines the training loop.
23
24	parameters are passed via `model_training_parameters` under the freqai
25	section in the config file. e.g:
26	{
27	...
28	"freqai": {
29	...
30	"model_training_parameters" : {
31	"learning_rate": 3e-4,
32	"trainer_kwargs": {
33	"n_steps": 5000,
34	"batch_size": 64,
35	"n_epochs": null
36	},
37	"model_kwargs": {
38	"hidden_dim": 512,
39	"dropout_percent": 0.2,
40	"n_layer": 1,
41	},
42	}
43	}
44	}
45	"""
46
47	@property	1✔
48	def data_convertor(self) -> PyTorchDataConvertor:	1✔
49	return DefaultPyTorchDataConvertor(target_tensor_type=torch.float)	1✔
50
51	def __init__(self, **kwargs) -> None:	1✔
52	super().__init__(**kwargs)	1✔
53	config = self.freqai_info.get("model_training_parameters", {})	1✔
54	self.learning_rate: float = config.get("learning_rate", 3e-4)	1✔
55	self.model_kwargs: Dict[str, Any] = config.get("model_kwargs", {})	1✔
56	self.trainer_kwargs: Dict[str, Any] = config.get("trainer_kwargs", {})	1✔
57
58	def fit(self, data_dictionary: Dict, dk: FreqaiDataKitchen, **kwargs) -> Any:	1✔
59	"""
60	User sets up the training and test data to fit their desired model here
61	:param data_dictionary: the dictionary holding all data for train, test,
62	labels, weights
63	:param dk: The datakitchen object for the current coin/model
64	"""
65
66	n_features = data_dictionary["train_features"].shape[-1]	1✔
67	n_labels = data_dictionary["train_labels"].shape[-1]	1✔
68	model = PyTorchTransformerModel(	1✔
69	input_dim=n_features,
70	output_dim=n_labels,
71	time_window=self.window_size,
72	**self.model_kwargs
73	)
74	model.to(self.device)	1✔
75	optimizer = torch.optim.AdamW(model.parameters(), lr=self.learning_rate)	1✔
76	criterion = torch.nn.MSELoss()	1✔
77	# check if continual_learning is activated, and retreive the model to continue training
78	trainer = self.get_init_model(dk.pair)	1✔
79	if trainer is None:	1✔
80	trainer = PyTorchTransformerTrainer(	1✔
81	model=model,
82	optimizer=optimizer,
83	criterion=criterion,
84	device=self.device,
85	data_convertor=self.data_convertor,
86	window_size=self.window_size,
87	tb_logger=self.tb_logger,
88	**self.trainer_kwargs,
89	)
90	trainer.fit(data_dictionary, self.splits)	1✔
91	return trainer	1✔
92
93	def predict(	1✔
94	self, unfiltered_df: pd.DataFrame, dk: FreqaiDataKitchen, **kwargs
95	) -> Tuple[pd.DataFrame, npt.NDArray[np.int_]]:
96	"""
97	Filter the prediction features data and predict with it.
98	:param unfiltered_df: Full dataframe for the current backtest period.
99	:return:
100	:pred_df: dataframe containing the predictions
101	:do_predict: np.array of 1s and 0s to indicate places where freqai needed to remove
102	data (NaNs) or felt uncertain about data (PCA and DI index)
103	"""
104
105	dk.find_features(unfiltered_df)	1✔
106	dk.data_dictionary["prediction_features"], _ = dk.filter_features(	1✔
107	unfiltered_df, dk.training_features_list, training_filter=False
108	)
109
110	dk.data_dictionary["prediction_features"], outliers, _ = dk.feature_pipeline.transform(	1✔
111	dk.data_dictionary["prediction_features"], outlier_check=True)
112
113	x = self.data_convertor.convert_x(	1✔
114	dk.data_dictionary["prediction_features"],
115	device=self.device
116	)
117	# if user is asking for multiple predictions, slide the window
118	# along the tensor
119	x = x.unsqueeze(0)	1✔
120	# create empty torch tensor
121	self.model.model.eval()	1✔
122	yb = torch.empty(0).to(self.device)	1✔
123	if x.shape[1] > 1:	1✔
124	ws = self.window_size	1✔
125	for i in range(0, x.shape[1] - ws):	1✔
126	xb = x[:, i:i + ws, :].to(self.device)	1✔
127	y = self.model.model(xb)	1✔
128	yb = torch.cat((yb, y), dim=0)	1✔
129	else:
130	yb = self.model.model(x)	×
131
132	yb = yb.cpu().squeeze()	1✔
133	pred_df = pd.DataFrame(yb.detach().numpy(), columns=dk.label_list)	1✔
134	pred_df, _, _ = dk.label_pipeline.inverse_transform(pred_df)	1✔
135
136	if self.freqai_info.get("DI_threshold", 0) > 0:	1✔
137	dk.DI_values = dk.feature_pipeline["di"].di_values	×
138	else:
139	dk.DI_values = np.zeros(outliers.shape[0])	1✔
140	dk.do_predict = outliers	1✔
141
142	if x.shape[1] > 1:	1✔
143	zeros_df = pd.DataFrame(np.zeros((x.shape[1] - len(pred_df), len(pred_df.columns))),	1✔
144	columns=pred_df.columns)
145	pred_df = pd.concat([zeros_df, pred_df], axis=0, ignore_index=True)	1✔
146	return (pred_df, dk.do_predict)	1✔

freqtrade / freqtrade / 6181253459

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous