SYMBOL INDEX (196 symbols across 22 files)

FILE: examples/prediction_batch_example.py
  function plot_prediction (line 8) | def plot_prediction(kline_df, pred_df):

FILE: examples/prediction_cn_markets_day.py
  function load_data (line 48) | def load_data(symbol: str) -> pd.DataFrame:
  function prepare_inputs (line 112) | def prepare_inputs(df):
  function apply_price_limits (line 118) | def apply_price_limits(pred_df, last_close, limit_rate=0.1):
  function plot_result (line 143) | def plot_result(df_hist, df_pred, symbol):
  function predict_future (line 159) | def predict_future(symbol):

FILE: examples/prediction_example.py
  function plot_prediction (line 8) | def plot_prediction(kline_df, pred_df):

FILE: examples/prediction_wo_vol_example.py
  function plot_prediction (line 8) | def plot_prediction(kline_df, pred_df):

FILE: finetune/config.py
  class Config (line 3) | class Config:
    method __init__ (line 8) | def __init__(self):
    method _set_benchmark (line 122) | def _set_benchmark(self, instrument):

FILE: finetune/dataset.py
  class QlibDataset (line 9) | class QlibDataset(Dataset):
    method __init__ (line 23) | def __init__(self, data_type: str = 'train'):
    method set_epoch_seed (line 77) | def set_epoch_seed(self, epoch: int):
    method __len__ (line 88) | def __len__(self) -> int:
    method __getitem__ (line 92) | def __getitem__(self, idx: int) -> tuple[torch.Tensor, torch.Tensor]:

FILE: finetune/qlib_data_preprocess.py
  class QlibDataPreprocessor (line 14) | class QlibDataPreprocessor:
    method __init__ (line 19) | def __init__(self):
    method initialize_qlib (line 25) | def initialize_qlib(self):
    method load_qlib_data (line 30) | def load_qlib_data(self):
    method prepare_dataset (line 85) | def prepare_dataset(self):

FILE: finetune/qlib_test.py
  class QlibTestDataset (line 32) | class QlibTestDataset(Dataset):
    method __init__ (line 41) | def __init__(self, data: dict, config: Config):
    method __len__ (line 67) | def __len__(self) -> int:
    method __getitem__ (line 70) | def __getitem__(self, idx: int):
  class QlibBacktest (line 96) | class QlibBacktest:
    method __init__ (line 101) | def __init__(self, config: Config):
    method initialize_qlib (line 105) | def initialize_qlib(self):
    method run_single_backtest (line 110) | def run_single_backtest(self, signal_series: pd.Series) -> pd.DataFrame:
    method run_and_plot_results (line 164) | def run_and_plot_results(self, signals: dict[str, pd.DataFrame]):
  function load_models (line 207) | def load_models(config: dict) -> tuple[KronosTokenizer, Kronos]:
  function collate_fn_for_inference (line 216) | def collate_fn_for_inference(batch):
  function generate_predictions (line 239) | def generate_predictions(config: dict, test_data: dict) -> dict[str, pd....
  function main (line 302) | def main():

FILE: finetune/train_predictor.py
  function create_dataloaders (line 29) | def create_dataloaders(config: dict, rank: int, world_size: int):
  function train_model (line 60) | def train_model(model, tokenizer, device, config, save_dir, logger, rank...
  function main (line 182) | def main(config: dict):

FILE: finetune/train_tokenizer.py
  function create_dataloaders (line 32) | def create_dataloaders(config: dict, rank: int, world_size: int):
  function train_model (line 74) | def train_model(model, device, config, save_dir, logger, rank, world_size):
  function main (line 218) | def main(config: dict):

FILE: finetune/utils/training_utils.py
  function setup_ddp (line 9) | def setup_ddp():
  function cleanup_ddp (line 35) | def cleanup_ddp():
  function set_seed (line 41) | def set_seed(seed: int, rank: int = 0):
  function get_model_size (line 62) | def get_model_size(model: torch.nn.Module) -> str:
  function reduce_tensor (line 83) | def reduce_tensor(tensor: torch.Tensor, world_size: int, op=dist.ReduceO...
  function format_time (line 105) | def format_time(seconds: float) -> str:

FILE: finetune_csv/config_loader.py
  class ConfigLoader (line 6) | class ConfigLoader:
    method __init__ (line 8) | def __init__(self, config_path: str):
    method _load_config (line 13) | def _load_config(self) -> Dict[str, Any]:
    method _resolve_dynamic_paths (line 25) | def _resolve_dynamic_paths(self, config: Dict[str, Any]) -> Dict[str, ...
    method get (line 51) | def get(self, key: str, default=None):
    method get_data_config (line 63) | def get_data_config(self) -> Dict[str, Any]:
    method get_training_config (line 66) | def get_training_config(self) -> Dict[str, Any]:
    method get_model_paths (line 69) | def get_model_paths(self) -> Dict[str, str]:
    method get_experiment_config (line 72) | def get_experiment_config(self) -> Dict[str, Any]:
    method get_device_config (line 75) | def get_device_config(self) -> Dict[str, Any]:
    method get_distributed_config (line 78) | def get_distributed_config(self) -> Dict[str, Any]:
    method update_config (line 81) | def update_config(self, updates: Dict[str, Any]):
    method save_config (line 93) | def save_config(self, save_path: str = None):
    method print_config (line 101) | def print_config(self):
  class CustomFinetuneConfig (line 109) | class CustomFinetuneConfig:
    method __init__ (line 111) | def __init__(self, config_path: str = None):
    method _load_all_configs (line 119) | def _load_all_configs(self):
    method _compute_full_paths (line 184) | def _compute_full_paths(self):
    method get_tokenizer_config (line 192) | def get_tokenizer_config(self):
    method get_basemodel_config (line 218) | def get_basemodel_config(self):
    method print_config_summary (line 245) | def print_config_summary(self):

FILE: finetune_csv/finetune_base_model.py
  class CustomKlineDataset (line 25) | class CustomKlineDataset(Dataset):
    method __init__ (line 27) | def __init__(self, data_path, data_type='train', lookback_window=90, p...
    method _load_and_preprocess_data (line 52) | def _load_and_preprocess_data(self):
    method _split_data_by_time (line 75) | def _split_data_by_time(self):
    method set_epoch_seed (line 99) | def set_epoch_seed(self, epoch):
    method __len__ (line 104) | def __len__(self):
    method __getitem__ (line 107) | def __getitem__(self, idx):
  function setup_logging (line 137) | def setup_logging(exp_name: str, log_dir: str, rank: int = 0) -> logging...
  function create_dataloaders (line 181) | def create_dataloaders(config):
  function train_model (line 239) | def train_model(model, tokenizer, device, config, save_dir, logger):
  function main (line 367) | def main():

FILE: finetune_csv/finetune_tokenizer.py
  function set_seed (line 24) | def set_seed(seed: int, rank: int = 0):
  function get_model_size (line 35) | def get_model_size(model: torch.nn.Module) -> str:
  function format_time (line 45) | def format_time(seconds: float) -> str:
  function setup_logging (line 49) | def setup_logging(exp_name: str, log_dir: str, rank: int = 0) -> logging...
  function create_dataloaders (line 93) | def create_dataloaders(config):
  function train_tokenizer (line 151) | def train_tokenizer(model, device, config, save_dir, logger):
  function main (line 281) | def main():

FILE: finetune_csv/train_sequential.py
  class SequentialTrainer (line 18) | class SequentialTrainer:
    method __init__ (line 20) | def __init__(self, config_path: str = None):
    method _setup_device (line 29) | def _setup_device(self):
    method _setup_distributed (line 40) | def _setup_distributed(self):
    method _check_existing_models (line 51) | def _check_existing_models(self):
    method _create_directories (line 60) | def _create_directories(self):
    method train_tokenizer_phase (line 66) | def train_tokenizer_phase(self):
    method train_basemodel_phase (line 148) | def train_basemodel_phase(self):
    method run_training (line 264) | def run_training(self):
  function main (line 319) | def main():

FILE: model/__init__.py
  function get_model_class (line 10) | def get_model_class(model_name):

FILE: model/kronos.py
  class KronosTokenizer (line 13) | class KronosTokenizer(nn.Module, PyTorchModelHubMixin):
    method __init__ (line 40) | def __init__(self, d_in, d_model, n_heads, ff_dim, n_enc_layers, n_dec...
    method forward (line 74) | def forward(self, x):
    method indices_to_bits (line 115) | def indices_to_bits(self, x, half=False):
    method encode (line 142) | def encode(self, x, half=False):
    method decode (line 161) | def decode(self, x, half=False):
  class Kronos (line 180) | class Kronos(nn.Module, PyTorchModelHubMixin):
    method __init__ (line 198) | def __init__(self, s1_bits, s2_bits, n_layers, d_model, n_heads, ff_di...
    method _init_weights (line 225) | def _init_weights(self, module):
    method forward (line 239) | def forward(self, s1_ids, s2_ids, stamp=None, padding_mask=None, use_t...
    method decode_s1 (line 278) | def decode_s1(self, s1_ids, s2_ids, stamp=None, padding_mask=None):
    method decode_s2 (line 310) | def decode_s2(self, context, s1_ids, padding_mask=None):
  function top_k_top_p_filtering (line 331) | def top_k_top_p_filtering(
  function sample_from_logits (line 373) | def sample_from_logits(logits, temperature=1.0, top_k=None, top_p=None, ...
  function auto_regressive_inference (line 389) | def auto_regressive_inference(tokenizer, model, x, x_stamp, y_stamp, max...
  function calc_time_stamps (line 472) | def calc_time_stamps(x_timestamp):
  class KronosPredictor (line 482) | class KronosPredictor:
    method __init__ (line 484) | def __init__(self, model, tokenizer, device=None, max_context=512, cli...
    method generate (line 508) | def generate(self, x, x_stamp, y_stamp, pred_len, T, top_k, top_p, sam...
    method predict (line 519) | def predict(self, df, x_timestamp, y_timestamp, pred_len, T=1.0, top_k...
    method predict_batch (line 562) | def predict_batch(self, df_list, x_timestamp_list, y_timestamp_list, p...

FILE: model/module.py
  class DifferentiableEntropyFunction (line 10) | class DifferentiableEntropyFunction(Function):
    method forward (line 12) | def forward(ctx, zq, basis, K, eps):
    method backward (line 27) | def backward(ctx, grad_output):
  function codebook_entropy (line 35) | def codebook_entropy(zq, basis, K, eps=1e-4):
  class BinarySphericalQuantizer (line 39) | class BinarySphericalQuantizer(nn.Module):
    method __init__ (line 40) | def __init__(self, embed_dim, beta, gamma0, gamma, zeta,
    method quantize (line 82) | def quantize(self, z):
    method forward (line 90) | def forward(self, z, collect_metrics=True):
    method soft_entropy_loss (line 131) | def soft_entropy_loss(self, z):
    method get_hard_per_sample_entropy (line 157) | def get_hard_per_sample_entropy(self, zb_by_sample):
    method codes_to_indexes (line 163) | def codes_to_indexes(self, zhat):
    method codes_to_group_indexes (line 171) | def codes_to_group_indexes(self, zhat):
    method indexes_to_codes (line 179) | def indexes_to_codes(self, indices):
    method group_indexes_to_codes (line 187) | def group_indexes_to_codes(self, group_indices):
    method get_entropy (line 196) | def get_entropy(self, count, dim=-1, eps=1e-4, normalize=True):
    method get_group_codebook_entry (line 204) | def get_group_codebook_entry(self, group_indices):
    method get_codebook_entry (line 214) | def get_codebook_entry(self, indices):
  class BSQuantizer (line 225) | class BSQuantizer(nn.Module):
    method __init__ (line 227) | def __init__(self, s1_bits, s2_bits, beta, gamma0, gamma, zeta, group_...
    method bits_to_indices (line 234) | def bits_to_indices(self, bits):
    method forward (line 245) | def forward(self, z, half=False, collect_metrics=True):
  class RMSNorm (line 257) | class RMSNorm(torch.nn.Module):
    method __init__ (line 258) | def __init__(self, dim: int, eps: float = 1e-5):
    method _norm (line 263) | def _norm(self, x):
    method forward (line 266) | def forward(self, x):
  class FeedForward (line 271) | class FeedForward(nn.Module):
    method __init__ (line 272) | def __init__(self, d_model, ff_dim, ffn_dropout_p=0.0):
    method forward (line 280) | def forward(self, x):
  class RotaryPositionalEmbedding (line 284) | class RotaryPositionalEmbedding(nn.Module):
    method __init__ (line 285) | def __init__(self, dim):
    method _update_cos_sin_cache (line 293) | def _update_cos_sin_cache(self, x, seq_len):
    method forward (line 303) | def forward(self, q, k):
    method _rotate_half (line 310) | def _rotate_half(self, x):
  class MultiHeadAttentionWithRoPE (line 315) | class MultiHeadAttentionWithRoPE(nn.Module):
    method __init__ (line 316) | def __init__(self, d_model, n_heads, attn_dropout_p=0.0, resid_dropout...
    method forward (line 330) | def forward(self, x, key_padding_mask=None):
  class MultiHeadCrossAttentionWithRoPE (line 356) | class MultiHeadCrossAttentionWithRoPE(nn.Module):
    method __init__ (line 357) | def __init__(self, d_model, n_heads, attn_dropout_p=0.0, resid_dropout...
    method forward (line 371) | def forward(self, query, key, value, key_padding_mask=None):
  class HierarchicalEmbedding (line 400) | class HierarchicalEmbedding(nn.Module):
    method __init__ (line 401) | def __init__(self, s1_bits, s2_bits, d_model=256):
    method split_token (line 417) | def split_token(self, token_ids: torch.Tensor, s2_bits: int):
    method forward (line 430) | def forward(self, token_ids):
  class DependencyAwareLayer (line 446) | class DependencyAwareLayer(nn.Module):
    method __init__ (line 447) | def __init__(self, d_model, n_heads=4, attn_dropout_p=0.0, resid_dropo...
    method forward (line 452) | def forward(self, hidden_states, sibling_embed, key_padding_mask=None):
  class TransformerBlock (line 465) | class TransformerBlock(nn.Module):
    method __init__ (line 466) | def __init__(self, d_model, n_heads, ff_dim=1024, ffn_dropout_p=0.0, a...
    method forward (line 473) | def forward(self, x, key_padding_mask=None):
  class DualHead (line 486) | class DualHead(nn.Module):
    method __init__ (line 487) | def __init__(self, s1_bits, s2_bits, d_model):
    method compute_loss (line 494) | def compute_loss(self, s1_logits, s2_logits, s1_targets, s2_targets, p...
    method forward (line 509) | def forward(self, x):
    method cond_forward (line 512) | def cond_forward(self, x2):
  class FixedEmbedding (line 516) | class FixedEmbedding(nn.Module):
    method __init__ (line 517) | def __init__(self, c_in, d_model):
    method forward (line 532) | def forward(self, x):
  class TemporalEmbedding (line 536) | class TemporalEmbedding(nn.Module):
    method __init__ (line 537) | def __init__(self, d_model, learn_pe):
    method forward (line 553) | def forward(self, x):

FILE: tests/data/generate_regression_output.py
  function set_seed (line 26) | def set_seed(seed: int) -> None:
  function generate_output (line 35) | def generate_output(ctx_len: int) -> None:

FILE: tests/test_kronos_regression.py
  function set_seed (line 36) | def set_seed(seed: int) -> None:
  function test_kronos_predictor_regression (line 46) | def test_kronos_predictor_regression(context_len):
  function test_kronos_predictor_mse (line 91) | def test_kronos_predictor_mse(context_len, expected_mse):

FILE: webui/app.py
  function load_data_files (line 60) | def load_data_files():
  function load_data_file (line 78) | def load_data_file(file_path):
  function save_prediction_results (line 125) | def save_prediction_results(file_path, prediction_type, prediction_resul...
  function create_prediction_chart (line 209) | def create_prediction_chart(df, pred_df, lookback, pred_len, actual_df=N...
  function index (line 331) | def index():
  function get_data_files (line 336) | def get_data_files():
  function load_data (line 342) | def load_data():
  function predict (line 405) | def predict():
  function load_model (line 627) | def load_model():
  function get_available_models (line 666) | def get_available_models():
  function get_model_status (line 674) | def get_model_status():

FILE: webui/run.py
  function check_dependencies (line 12) | def check_dependencies():
  function install_dependencies (line 27) | def install_dependencies():
  function main (line 38) | def main():