def gold_miner_mold():
gm = GoldMiner()
gm.data = pickle_load("cleaned_data")[:20]
print(len(gm.data))
print(len(gm.data[0]))
start = time.time()
gm.mold()
print(len(gm.data[0]))
print("elapse {0} seconds".format(time.time() - start))
def cast_hound():
h = Hound()
data = smart_decode(pickle_load("train_data"), cast=True)[:10]
h.x_predict, h.y_predict = split_x_y(data)
h.load_model()
h.predict()
h.print_info()
data = list(map(lambda x, y: x + y, [line[:4] for line in data], h.prediction.astype(unicode).tolist()))
for line in data:
print(len("\t".join(line)))
def gold_miner_hive_test():
gm = GoldMiner()
gm.data = smart_decode(pickle_load("raw_data"), cast=True)
gm.pan()
print(len(gm.data))
gm.smelt()
def train_hound():
h = Hound()
h.x_train, h.y_train = split_x_y(smart_decode(pickle_load("train_data"), cast=True))
h.train()
h.save_model()
h.print_info()
def transform():
pickle_dump("id_content",
{line[0]: line[1]
for line in pickle_load("raw")})