# configure logger
Log.init_instance(config().logfile(),
config().loglevel(),
config().logformat())
assets = {}
correlations = {}
if __name__ == "__main__":
for path in config().input_path():
Log.info("Searching in %s", path)
for filename in os.listdir(path):
if fnmatch.fnmatch(filename, "*.csv"):
Log.info("File %s", filename)
asset = Asset()
asset.read_csv(os.path.join(path, filename))
if len(asset.data["Volume"][
asset.data["Volume"] > 10000]) < args.min_data_points:
Log.info(
"Ignoring data set since it has too few entries ({} instead of {})"
.format(
len(asset.data["Volume"][
asset.data["Volume"] > 10000]),
args.min_data_points))
continue
if asset.exchange is None or asset.symbol is None:
Log.info(
"Ignoring data set since exchange or symbol is missing"
)
continue
assets[asset.exchange + "/" + asset.symbol] = asset
help="Log level: DEBUG, INFO, WARN, ERROR")
parser.add_argument("--logfile",
action="store",
metavar="PATH",
help="Path to log file.")
parser.add_argument("-o",
"--output",
metavar="FILE",
required=True,
action="store",
help="Output file")
parser.add_argument("filename", metavar="FILE", help="files to process")
args = parser.parse_args()
# configuration files
Configuration.init_instance(args.config, args)
# configure logger
Log.init_instance(config().logfile(),
config().loglevel(),
config().logformat())
if __name__ == "__main__":
Log.info("Generating indicators for %s", args.filename)
asset = Asset()
asset.read_csv(args.filename)
asset = gen_indicators(asset)
asset.write_csv(args.output)
Log.info("Output written to %s", args.output)
"svm=SVM regressor, "
"knn=KNN regressor, "
"nn=neural network regressor")
parser.add_argument("filename",
metavar="FILE",
nargs='*',
help="Stock data input files to process")
args = parser.parse_args()
# configuration files
Configuration.init_instance(args.config, args)
# configure logger
Log.init_instance(config().logfile(),
config().loglevel(),
config().logformat())
if __name__ == "__main__":
for filename in config().input_filenames():
Log.info("Running models for %s", filename)
asset = Asset()
asset.read_csv(filename)
for model in args.models.split(","):
if model == "lstm":
lstm_model.fit_LSTM_models(asset)
elif model == "clf":
models.fit_classifiers(asset, classifiers=args.clf.split(","))
elif model == "reg":
models.fit_regressors(asset, regressors=args.reg.split(","))
