diff --git a/xformers/xformers.ipynb b/xformers/xformers.ipynb new file mode 100644 index 0000000..de5a38b --- /dev/null +++ b/xformers/xformers.ipynb @@ -0,0 +1,1692 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 4, + "id": "855dbf2e", + "metadata": {}, + "outputs": [], + "source": [ + "%matplotlib inline" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "8dc0844c", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "WARNING:root:WARNING: \n", + "Need to compile C++ extensions to get sparse attention suport. Please run python setup.py build develop\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "WARNING:root:Blocksparse is not available: the current GPU does not expose Tensor cores\n" + ] + } + ], + "source": [ + "from typing import List, Optional, Iterable, Dict, Any\n", + "from itertools import islice\n", + "import queue\n", + "\n", + "import numpy as np\n", + "from matplotlib import pyplot as plt\n", + "import matplotlib.dates as mdates\n", + "import tqdm.auto as tqdm\n", + "\n", + "import torch\n", + "import torch.nn as nn\n", + "from torch.utils.data import DataLoader\n", + "from torch.optim.lr_scheduler import _LRScheduler, MultiplicativeLR\n", + "from torch.optim import Optimizer\n", + "\n", + "import pytorch_lightning as pl\n", + "from pytorch_lightning.loggers import CSVLogger\n", + "\n", + "from xformers.factory.model_factory import xFormer, xFormerConfig\n", + "\n", + "from gluonts.core.component import validated\n", + "from gluonts.dataset.common import Dataset\n", + "from gluonts.dataset.field_names import FieldName\n", + "from gluonts.itertools import Cyclic, PseudoShuffled, IterableSlice\n", + "from gluonts.time_feature import (\n", + " TimeFeature,\n", + " time_features_from_frequency_str,\n", + ")\n", + "from gluonts.torch.modules.loss import DistributionLoss, NegativeLogLikelihood\n", + "from gluonts.transform import (\n", + " Transformation,\n", + " Chain,\n", + " RemoveFields,\n", + " SetField,\n", + " AsNumpyArray,\n", + " AddObservedValuesIndicator,\n", + " AddTimeFeatures,\n", + " AddAgeFeature,\n", + " VstackFeatures,\n", + " InstanceSplitter,\n", + " ValidationSplitSampler,\n", + " TestSplitSampler,\n", + " ExpectedNumInstanceSampler,\n", + " SelectFields,\n", + ")\n", + "from gluonts.torch.util import (\n", + " IterableDataset,\n", + ")\n", + "from gluonts.evaluation import make_evaluation_predictions, Evaluator\n", + "from gluonts.torch.model.estimator import PyTorchLightningEstimator\n", + "from gluonts.torch.model.predictor import PyTorchPredictor\n", + "from gluonts.torch.modules.distribution_output import (\n", + " DistributionOutput,\n", + " StudentTOutput,\n", + ")\n", + "from gluonts.torch.util import weighted_average\n", + "from gluonts.torch.modules.scaler import MeanScaler, NOPScaler\n", + "from gluonts.torch.modules.feature import FeatureEmbedder\n", + "from gluonts.time_feature import get_lags_for_frequency\n", + "from gluonts.dataset.repository.datasets import get_dataset" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "b0aa10bd", + "metadata": {}, + "outputs": [], + "source": [ + "class ABEL(_LRScheduler):\n", + " def __init__(self, optimizer: Optimizer, gamma: float=0.15, last_epoch: int = -1, debug: bool=False) -> None:\n", + " self.optimizer = optimizer\n", + " self.gamma = gamma\n", + " self.debug = debug\n", + " self.norms = []\n", + " self.reached_minimum = False\n", + "\n", + " super().__init__(optimizer, last_epoch=last_epoch)\n", + "\n", + " def get_lr(self):\n", + " if not self._get_lr_called_within_step:\n", + " warnings.warn(\n", + " \"To get the last learning rate computed by the scheduler, \" \"please use `get_last_lr()`.\",\n", + " UserWarning,\n", + " )\n", + "\n", + " norm = sum([sum([p.view(-1).norm(2) for p in group[\"params\"]]) for group in self.optimizer.param_groups])\n", + " self.norms.append(norm.cpu().item())\n", + " gamma = 1\n", + " if len(self.norms) > 3 and (self.norms[-1] - self.norms[-2]) * (self.norms[-2] - self.norms[-3]) < 0:\n", + " if self.reached_minimum:\n", + " self.reached_minimum = False\n", + " gamma = self.gamma\n", + " \n", + " if self.debug:\n", + " print(f\"ABEL decreases LR at {self.last_epoch}th step\")\n", + " else:\n", + " self.reached_minimum = True\n", + " return [group[\"lr\"] * gamma for group in self.optimizer.param_groups]" + ] + }, + { + "cell_type": "code", + "execution_count": 59, + "id": "b8af6d90", + "metadata": {}, + "outputs": [], + "source": [ + "class TransformerModel(nn.Module):\n", + " @validated()\n", + " def __init__(\n", + " self,\n", + " freq: str,\n", + " context_length: int,\n", + " prediction_length: int,\n", + " num_feat_dynamic_real: int,\n", + " num_feat_static_real: int,\n", + " num_feat_static_cat: int,\n", + " cardinality: List[int],\n", + " \n", + " # transformer arguments\n", + " nhead: int,\n", + " num_encoder_layers: int,\n", + " num_decoder_layers: int,\n", + " dim_feedforward: int,\n", + " activation: str = \"gelu\",\n", + " dropout: float = 0.1,\n", + "\n", + " # univariate input\n", + " input_size: int = 1,\n", + " embedding_dimension: Optional[List[int]] = None,\n", + " distr_output: DistributionOutput = StudentTOutput(),\n", + " lags_seq: Optional[List[int]] = None,\n", + " scaling: bool = True,\n", + " num_parallel_samples: int = 100,\n", + " ) -> None:\n", + " super().__init__()\n", + " \n", + " self.input_size = input_size\n", + " \n", + " self.target_shape = distr_output.event_shape\n", + " self.num_feat_dynamic_real = num_feat_dynamic_real\n", + " self.num_feat_static_cat = num_feat_static_cat\n", + " self.num_feat_static_real = num_feat_static_real\n", + " self.embedding_dimension = (\n", + " embedding_dimension\n", + " if embedding_dimension is not None or cardinality is None\n", + " else [min(50, (cat + 1) // 2) for cat in cardinality]\n", + " )\n", + " self.lags_seq = lags_seq or get_lags_for_frequency(freq_str=freq)\n", + " self.num_parallel_samples = num_parallel_samples\n", + " self.history_length = context_length + max(self.lags_seq)\n", + " self.embedder = FeatureEmbedder(\n", + " cardinalities=cardinality,\n", + " embedding_dims=self.embedding_dimension,\n", + " )\n", + " if scaling:\n", + " self.scaler = MeanScaler(dim=1, keepdim=True)\n", + " else:\n", + " self.scaler = NOPScaler(dim=1, keepdim=True)\n", + " \n", + " # total feature size\n", + " d_model = self.input_size * len(self.lags_seq) + self._number_of_features\n", + " \n", + " xformer_config = [\n", + " # A list of the encoder or decoder blocks which constitute the Transformer.\n", + " # Note that a sequence of different encoder blocks can be used, same for decoders\n", + " {\n", + " \"reversible\": False, # Optionally make these layers reversible, to save memory\n", + " \"block_config\": {\n", + " \"block_type\": \"encoder\",\n", + " \"num_layers\": num_encoder_layers, # Optional, this means that this config will repeat N times\n", + " \"dim_model\": d_model,\n", + " \"layer_norm_style\": \"pre\", # Optional, pre/post\n", + "# \"position_encoding_config\": {\n", + "# \"name\": \"sine\", # whatever position encoding makes sense\n", + "# \"dim_model\": d_model,\n", + "# },\n", + " \"multi_head_config\": {\n", + " \"num_heads\": nhead,\n", + " \"residual_dropout\": dropout,\n", + " \"attention\": {\n", + " \"name\": \"linformer\", # whatever attention mechanism\n", + " \"dropout\": dropout,\n", + " \"causal\": False,\n", + " \"seq_len\": self.history_length,\n", + " },\n", + " },\n", + " \"feedforward_config\": {\n", + " \"name\": \"MLP\",\n", + " \"dropout\": dropout,\n", + " \"activation\": activation,\n", + " \"hidden_layer_multiplier\": 2,\n", + " },\n", + " },\n", + " },\n", + " {\n", + " \"reversible\": False, # Optionally make these layers reversible, to save memory\n", + " \"block_config\": {\n", + " \"block_type\": \"decoder\",\n", + " \"num_layers\": num_decoder_layers, # Optional, this means that this config will repeat N times\n", + " \"dim_model\": d_model,\n", + " \"layer_norm_style\": \"pre\", # Optional, pre/post\n", + "# \"position_encoding_config\": {\n", + "# \"name\": \"sine\", # whatever position encoding makes sense\n", + "# \"dim_model\": d_model,\n", + "# },\n", + " \"multi_head_config_masked\": {\n", + " \"num_heads\": nhead,\n", + " \"residual_dropout\": dropout,\n", + " \"attention\": {\n", + " \"name\": \"nystrom\", # whatever attention mechanism\n", + " \"dropout\": dropout,\n", + " \"causal\": True,\n", + " \"seq_len\": prediction_length,\n", + " },\n", + " },\n", + " \"multi_head_config_cross\": {\n", + " \"num_heads\": nhead,\n", + " \"residual_dropout\": dropout,\n", + " \"attention\": {\n", + " \"name\": \"favor\", # whatever attention mechanism\n", + " \"dropout\": dropout,\n", + " \"causal\": True,\n", + " \"seq_len\": prediction_length,\n", + " },\n", + " },\n", + " \"feedforward_config\": {\n", + " \"name\": \"MLP\",\n", + " \"dropout\": dropout,\n", + " \"activation\": activation,\n", + " \"hidden_layer_multiplier\": 2,\n", + " },\n", + " },\n", + " },\n", + " ]\n", + " config = xFormerConfig(xformer_config)\n", + " \n", + " self.context_length = context_length\n", + " self.prediction_length = prediction_length\n", + " self.distr_output = distr_output\n", + " self.param_proj = distr_output.get_args_proj(d_model)\n", + " \n", + " # xformer\n", + " self.transformer = xFormer.from_config(config)\n", + " \n", + "# # causal decoder tgt mask\n", + "# self.register_buffer(\n", + "# \"tgt_mask\",\n", + "# self.transformer.generate_square_subsequent_mask(prediction_length),\n", + "# )\n", + " \n", + " @property\n", + " def _number_of_features(self) -> int:\n", + " return (\n", + " sum(self.embedding_dimension)\n", + " + self.num_feat_dynamic_real\n", + " + self.num_feat_static_real\n", + " + 1 # the log(scale)\n", + " )\n", + "\n", + " @property\n", + " def _past_length(self) -> int:\n", + " return self.context_length + max(self.lags_seq)\n", + " \n", + " def get_lagged_subsequences(\n", + " self,\n", + " sequence: torch.Tensor,\n", + " subsequences_length: int,\n", + " shift: int = 0\n", + " ) -> torch.Tensor:\n", + " \"\"\"\n", + " Returns lagged subsequences of a given sequence.\n", + " Parameters\n", + " ----------\n", + " sequence : Tensor\n", + " the sequence from which lagged subsequences should be extracted.\n", + " Shape: (N, T, C).\n", + " subsequences_length : int\n", + " length of the subsequences to be extracted.\n", + " shift: int\n", + " shift the lags by this amount back.\n", + " Returns\n", + " --------\n", + " lagged : Tensor\n", + " a tensor of shape (N, S, C, I), where S = subsequences_length and\n", + " I = len(indices), containing lagged subsequences. Specifically,\n", + " lagged[i, j, :, k] = sequence[i, -indices[k]-S+j, :].\n", + " \"\"\"\n", + " sequence_length = sequence.shape[1]\n", + " indices = [l - shift for l in self.lags_seq]\n", + "\n", + " assert max(indices) + subsequences_length <= sequence_length, (\n", + " f\"lags cannot go further than history length, found lag {max(indices)} \"\n", + " f\"while history length is only {sequence_length}\"\n", + " )\n", + "\n", + " lagged_values = []\n", + " for lag_index in indices:\n", + " begin_index = -lag_index - subsequences_length\n", + " end_index = -lag_index if lag_index > 0 else None\n", + " lagged_values.append(sequence[:, begin_index:end_index, ...])\n", + " return torch.stack(lagged_values, dim=-1)\n", + "\n", + " def _check_shapes(\n", + " self,\n", + " prior_input: torch.Tensor,\n", + " inputs: torch.Tensor,\n", + " features: Optional[torch.Tensor],\n", + " ) -> None:\n", + " assert len(prior_input.shape) == len(inputs.shape)\n", + " assert (\n", + " len(prior_input.shape) == 2 and self.input_size == 1\n", + " ) or prior_input.shape[2] == self.input_size\n", + " assert (len(inputs.shape) == 2 and self.input_size == 1) or inputs.shape[\n", + " -1\n", + " ] == self.input_size\n", + " assert (\n", + " features is None or features.shape[2] == self._number_of_features\n", + " ), f\"{features.shape[2]}, expected {self._number_of_features}\"\n", + " \n", + " \n", + " def create_network_inputs(\n", + " self, \n", + " feat_static_cat: torch.Tensor, \n", + " feat_static_real: torch.Tensor,\n", + " past_time_feat: torch.Tensor,\n", + " past_target: torch.Tensor,\n", + " past_observed_values: torch.Tensor,\n", + " future_time_feat: Optional[torch.Tensor] = None,\n", + " future_target: Optional[torch.Tensor] = None,\n", + " ): \n", + " # time feature\n", + " time_feat = (\n", + " past_time_feat[:, self._past_length - self.context_length :, ...]\n", + " if future_time_feat is None or future_target is None\n", + " else torch.cat(\n", + " (\n", + " past_time_feat[:, self._past_length - self.context_length :, ...],\n", + " future_time_feat,\n", + " ),\n", + " dim=1,\n", + " )\n", + " )\n", + "\n", + " # target\n", + " context = past_target[:, -self.context_length :]\n", + " observed_context = past_observed_values[:, -self.context_length :]\n", + " _, scale = self.scaler(context, observed_context)\n", + "\n", + " inputs = (\n", + " torch.cat((past_target, future_target), dim=1) / scale\n", + " if future_target is not None\n", + " else past_target / scale\n", + " )\n", + "\n", + " inputs_length = (\n", + " self._past_length + self.prediction_length\n", + " if future_target is not None\n", + " else self._past_length\n", + " )\n", + " assert inputs.shape[1] == inputs_length\n", + " \n", + " subsequences_length = (\n", + " self.context_length\n", + " if future_time_feat is None or future_target is None\n", + " else self.context_length + self.prediction_length\n", + " )\n", + " \n", + " # embeddings\n", + " embedded_cat = self.embedder(feat_static_cat)\n", + " static_feat = torch.cat(\n", + " (embedded_cat, feat_static_real, scale.log()),\n", + " dim=1,\n", + " )\n", + " expanded_static_feat = static_feat.unsqueeze(1).expand(\n", + " -1, time_feat.shape[1], -1\n", + " )\n", + " \n", + " \n", + " features = torch.cat((expanded_static_feat, time_feat), dim=-1)\n", + " \n", + " \n", + " #self._check_shapes(prior_input, inputs, features)\n", + "\n", + " #sequence = torch.cat((prior_input, inputs), dim=1)\n", + " lagged_sequence = self.get_lagged_subsequences(\n", + " sequence=inputs,\n", + " subsequences_length=subsequences_length,\n", + " )\n", + "\n", + " lags_shape = lagged_sequence.shape\n", + " reshaped_lagged_sequence = lagged_sequence.reshape(\n", + " lags_shape[0], lags_shape[1], -1\n", + " )\n", + "\n", + " if features is None:\n", + " transformer_inputs = reshaped_lagged_sequence\n", + " else:\n", + " transformer_inputs = torch.cat((reshaped_lagged_sequence, features), dim=-1)\n", + " \n", + " return transformer_inputs, scale, static_feat\n", + " \n", + " def output_params(self, transformer_inputs):\n", + " enc_input = transformer_inputs[:, :self.context_length, ...]\n", + " dec_input = transformer_inputs[:, self.context_length:, ...]\n", + " \n", + " dec_output = self.transformer(src=enc_input, tgt=dec_input)\n", + "# enc_out = self.transformer.encoders(\n", + "# enc_input\n", + "# )\n", + "# dec_output = self.transformer.decoders(\n", + "# dec_input,\n", + "# enc_out,\n", + "# #tgt_mask=self.tgt_mask #TODO\n", + "# )\n", + " \n", + " return self.param_proj(dec_output)\n", + "\n", + " @torch.jit.ignore\n", + " def output_distribution(\n", + " self, params, scale=None, trailing_n=None\n", + " ) -> torch.distributions.Distribution:\n", + " sliced_params = params\n", + " if trailing_n is not None:\n", + " sliced_params = [p[:, -trailing_n:] for p in params]\n", + " return self.distr_output.distribution(sliced_params, scale=scale)\n", + " \n", + " # for prediction\n", + " def forward(\n", + " self,\n", + " feat_static_cat: torch.Tensor,\n", + " feat_static_real: torch.Tensor,\n", + " past_time_feat: torch.Tensor,\n", + " past_target: torch.Tensor,\n", + " past_observed_values: torch.Tensor,\n", + " future_time_feat: torch.Tensor,\n", + " num_parallel_samples: Optional[int] = None,\n", + " ) -> torch.Tensor:\n", + " \n", + " \n", + " if num_parallel_samples is None:\n", + " num_parallel_samples = self.num_parallel_samples\n", + " \n", + " encoder_inputs, scale, static_feat = self.create_network_inputs(\n", + " feat_static_cat,\n", + " feat_static_real,\n", + " past_time_feat,\n", + " past_target,\n", + " past_observed_values,\n", + " future_time_feat,\n", + " )\n", + " \n", + " enc_out = self.transformer.encoders(encoder_inputs)\n", + " params = self.param_proj(enc_out)\n", + " distr = self.output_distribution(params, trailing_n=1)\n", + " \n", + " repeated_scale = scale.repeat_interleave(\n", + " repeats=self.num_parallel_samples, dim=0\n", + " )\n", + " repeated_static_feat = static_feat.repeat_interleave(\n", + " repeats=self.num_parallel_samples, dim=0\n", + " ).unsqueeze(dim=1)\n", + " repeated_past_target = (\n", + " past_target.repeat_interleave(\n", + " repeats=self.num_parallel_samples, dim=0\n", + " )\n", + " / repeated_scale\n", + " )\n", + " repeated_time_feat = future_time_feat.repeat_interleave(\n", + " repeats=self.num_parallel_samples, dim=0\n", + " )\n", + " repeated_enc_out = enc_out.repeat_interleave(\n", + " repeats=self.num_parallel_samples, dim=0\n", + " )\n", + "\n", + " future_samples = []\n", + " \n", + " for k in range(self.prediction_length):\n", + " next_features = torch.cat(\n", + " (repeated_static_feat, repeated_time_feat[:, k : k + 1]),\n", + " dim=-1,\n", + " )\n", + " \n", + " #self._check_shapes(repeated_past_target, next_sample, next_features)\n", + "\n", + " #sequence = torch.cat((repeated_past_target, next_sample), dim=1)\n", + " \n", + " lagged_sequence = self.get_lagged_subsequences(\n", + " sequence=repeated_past_target,\n", + " subsequences_length=1,\n", + " shift=1, \n", + " )\n", + "\n", + " lags_shape = lagged_sequence.shape\n", + " reshaped_lagged_sequence = lagged_sequence.reshape(\n", + " lags_shape[0], lags_shape[1], -1\n", + " )\n", + " \n", + " decoder_input = torch.cat((reshaped_lagged_sequence, next_features), dim=-1)\n", + "\n", + " output = self.transformer.decoders(decoder_input, repeated_enc_out)\n", + " \n", + " params = self.param_proj(output)\n", + " distr = self.output_distribution(params)\n", + " next_sample = distr.sample()\n", + " \n", + " repeated_past_target = torch.cat(\n", + " (repeated_past_target, next_sample), dim=1\n", + " )\n", + " future_samples.append(next_sample)\n", + "\n", + " unscaled_future_samples = (\n", + " torch.cat(future_samples, dim=1) * repeated_scale\n", + " )\n", + " return unscaled_future_samples.reshape(\n", + " (-1, self.num_parallel_samples, self.prediction_length)\n", + " + self.target_shape,\n", + " )" + ] + }, + { + "cell_type": "code", + "execution_count": 60, + "id": "9b1529a8", + "metadata": {}, + "outputs": [], + "source": [ + "class TransformerLightningModule(pl.LightningModule):\n", + " def __init__(\n", + " self,\n", + " model: TransformerModel,\n", + " loss: DistributionLoss = NegativeLogLikelihood(),\n", + " lr: float = 1e-3,\n", + " weight_decay: float = 1e-8,\n", + " ) -> None:\n", + " super().__init__()\n", + " self.save_hyperparameters()\n", + " self.model = model\n", + " self.loss = loss\n", + " self.lr = lr\n", + " self.weight_decay = weight_decay\n", + " \n", + " def training_step(self, batch, batch_idx: int):\n", + " \"\"\"Execute training step\"\"\"\n", + " train_loss = self(batch)\n", + " self.log(\n", + " \"train_loss\",\n", + " train_loss,\n", + " on_epoch=True,\n", + " on_step=False,\n", + " prog_bar=True,\n", + " )\n", + " return train_loss\n", + "\n", + " def validation_step(self, batch, batch_idx: int):\n", + " \"\"\"Execute validation step\"\"\"\n", + " with torch.inference_mode():\n", + " val_loss = self(batch)\n", + " self.log(\n", + " \"val_loss\", val_loss, on_epoch=True, on_step=False, prog_bar=True\n", + " )\n", + " return val_loss\n", + "\n", + " def configure_optimizers(self):\n", + " \"\"\"Returns the optimizer and schedular to use\"\"\"\n", + " \n", + " optim = torch.optim.Adam(\n", + " self.model.parameters(),\n", + " lr=self.lr,\n", + " weight_decay=self.weight_decay,\n", + " )\n", + " lr_scheduler = {\n", + " 'scheduler': ABEL(optim, gamma=0.2, debug=True),\n", + " 'name': 'ABEL'\n", + " }\n", + " \n", + " return [optim], [lr_scheduler]\n", + "\n", + " def forward(self, batch):\n", + " feat_static_cat = batch[\"feat_static_cat\"]\n", + " feat_static_real = batch[\"feat_static_real\"]\n", + " past_time_feat = batch[\"past_time_feat\"]\n", + " past_target = batch[\"past_target\"]\n", + " future_time_feat = batch[\"future_time_feat\"]\n", + " future_target = batch[\"future_target\"]\n", + " past_observed_values = batch[\"past_observed_values\"]\n", + " future_observed_values = batch[\"future_observed_values\"]\n", + " \n", + " transformer_inputs, scale, _ = self.model.create_network_inputs(\n", + " feat_static_cat,\n", + " feat_static_real,\n", + " past_time_feat,\n", + " past_target,\n", + " past_observed_values,\n", + " future_time_feat,\n", + " future_target,\n", + " )\n", + " params = self.model.output_params(transformer_inputs)\n", + " distr = self.model.output_distribution(params, scale)\n", + "\n", + " loss_values = self.loss(distr, future_target)\n", + " \n", + " if len(self.model.target_shape) == 0:\n", + " loss_weights = future_observed_values\n", + " else:\n", + " loss_weights = future_observed_values.min(dim=-1, keepdim=False)\n", + "\n", + " return weighted_average(loss_values, weights=loss_weights)" + ] + }, + { + "cell_type": "code", + "execution_count": 61, + "id": "f7b4c72f", + "metadata": {}, + "outputs": [], + "source": [ + "PREDICTION_INPUT_NAMES = [\n", + " \"feat_static_cat\",\n", + " \"feat_static_real\",\n", + " \"past_time_feat\",\n", + " \"past_target\",\n", + " \"past_observed_values\",\n", + " \"future_time_feat\",\n", + "]\n", + "\n", + "TRAINING_INPUT_NAMES = PREDICTION_INPUT_NAMES + [\n", + " \"future_target\",\n", + " \"future_observed_values\",\n", + "]" + ] + }, + { + "cell_type": "code", + "execution_count": 62, + "id": "1937891a", + "metadata": {}, + "outputs": [], + "source": [ + "class TransformerEstimator(PyTorchLightningEstimator):\n", + " @validated()\n", + " def __init__(\n", + " self,\n", + " freq: str,\n", + " prediction_length: int,\n", + " \n", + " # Transformer arguments\n", + " nhead: int,\n", + " num_encoder_layers: int,\n", + " num_decoder_layers: int,\n", + " dim_feedforward: int,\n", + " input_size: int = 1,\n", + " activation: str = \"gelu\",\n", + " dropout: float = 0.1,\n", + "\n", + " context_length: Optional[int] = None,\n", + "\n", + " num_feat_dynamic_real: int = 0,\n", + " num_feat_static_cat: int = 0,\n", + " num_feat_static_real: int = 0,\n", + " cardinality: Optional[List[int]] = None,\n", + " embedding_dimension: Optional[List[int]] = None,\n", + " distr_output: DistributionOutput = StudentTOutput(),\n", + " loss: DistributionLoss = NegativeLogLikelihood(),\n", + " scaling: bool = True,\n", + " lags_seq: Optional[List[int]] = None,\n", + " time_features: Optional[List[TimeFeature]] = None,\n", + " num_parallel_samples: int = 100,\n", + " batch_size: int = 32,\n", + " num_batches_per_epoch: int = 50,\n", + " trainer_kwargs: Optional[Dict[str, Any]] = dict(),\n", + " ) -> None:\n", + " trainer_kwargs = {\n", + " \"max_epochs\": 100,\n", + " **trainer_kwargs,\n", + " }\n", + " super().__init__(trainer_kwargs=trainer_kwargs)\n", + " \n", + " self.freq = freq\n", + " self.context_length = (\n", + " context_length if context_length is not None else prediction_length\n", + " )\n", + " self.prediction_length = prediction_length\n", + " self.distr_output = distr_output\n", + " self.loss = loss\n", + " \n", + " self.input_size = input_size\n", + " self.nhead = nhead\n", + " self.num_encoder_layers = num_encoder_layers\n", + " self.num_decoder_layers = num_decoder_layers\n", + " self.activation = activation\n", + " self.dim_feedforward = dim_feedforward\n", + " self.dropout = dropout\n", + " \n", + " self.num_feat_dynamic_real = num_feat_dynamic_real\n", + " self.num_feat_static_cat = num_feat_static_cat\n", + " self.num_feat_static_real = num_feat_static_real\n", + " self.cardinality = (\n", + " cardinality if cardinality and num_feat_static_cat > 0 else [1]\n", + " )\n", + " self.embedding_dimension = embedding_dimension\n", + " self.scaling = scaling\n", + " self.lags_seq = lags_seq\n", + " self.time_features = (\n", + " time_features\n", + " if time_features is not None\n", + " else time_features_from_frequency_str(self.freq)\n", + " )\n", + "\n", + " self.num_parallel_samples = num_parallel_samples\n", + " self.batch_size = batch_size\n", + " self.num_batches_per_epoch = num_batches_per_epoch\n", + "\n", + " self.train_sampler = ExpectedNumInstanceSampler(\n", + " num_instances=1.0, min_future=prediction_length\n", + " )\n", + " self.validation_sampler = ValidationSplitSampler(\n", + " min_future=prediction_length\n", + " )\n", + " \n", + " def create_transformation(self) -> Transformation:\n", + " remove_field_names = []\n", + " if self.num_feat_static_real == 0:\n", + " remove_field_names.append(FieldName.FEAT_STATIC_REAL)\n", + " if self.num_feat_dynamic_real == 0:\n", + " remove_field_names.append(FieldName.FEAT_DYNAMIC_REAL)\n", + "\n", + " return Chain(\n", + " [RemoveFields(field_names=remove_field_names)]\n", + " + (\n", + " [SetField(output_field=FieldName.FEAT_STATIC_CAT, value=[0])]\n", + " if not self.num_feat_static_cat > 0\n", + " else []\n", + " )\n", + " + (\n", + " [\n", + " SetField(\n", + " output_field=FieldName.FEAT_STATIC_REAL, value=[0.0]\n", + " )\n", + " ]\n", + " if not self.num_feat_static_real > 0\n", + " else []\n", + " )\n", + " + [\n", + " AsNumpyArray(\n", + " field=FieldName.FEAT_STATIC_CAT,\n", + " expected_ndim=1,\n", + " dtype=np.long,\n", + " ),\n", + " AsNumpyArray(\n", + " field=FieldName.FEAT_STATIC_REAL,\n", + " expected_ndim=1,\n", + " ),\n", + " AsNumpyArray(\n", + " field=FieldName.TARGET,\n", + " # in the following line, we add 1 for the time dimension\n", + " expected_ndim=1 + len(self.distr_output.event_shape),\n", + " ),\n", + " AddObservedValuesIndicator(\n", + " target_field=FieldName.TARGET,\n", + " output_field=FieldName.OBSERVED_VALUES,\n", + " ),\n", + " AddTimeFeatures(\n", + " start_field=FieldName.START,\n", + " target_field=FieldName.TARGET,\n", + " output_field=FieldName.FEAT_TIME,\n", + " time_features=self.time_features,\n", + " pred_length=self.prediction_length,\n", + " ),\n", + " AddAgeFeature(\n", + " target_field=FieldName.TARGET,\n", + " output_field=FieldName.FEAT_AGE,\n", + " pred_length=self.prediction_length,\n", + " log_scale=True,\n", + " ),\n", + " VstackFeatures(\n", + " output_field=FieldName.FEAT_TIME,\n", + " input_fields=[FieldName.FEAT_TIME, FieldName.FEAT_AGE]\n", + " + (\n", + " [FieldName.FEAT_DYNAMIC_REAL]\n", + " if self.num_feat_dynamic_real > 0\n", + " else []\n", + " ),\n", + " ),\n", + " ]\n", + " )\n", + "\n", + " def _create_instance_splitter(\n", + " self, module: TransformerLightningModule, mode: str\n", + " ):\n", + " assert mode in [\"training\", \"validation\", \"test\"]\n", + "\n", + " instance_sampler = {\n", + " \"training\": self.train_sampler,\n", + " \"validation\": self.validation_sampler,\n", + " \"test\": TestSplitSampler(),\n", + " }[mode]\n", + "\n", + " return InstanceSplitter(\n", + " target_field=FieldName.TARGET,\n", + " is_pad_field=FieldName.IS_PAD,\n", + " start_field=FieldName.START,\n", + " forecast_start_field=FieldName.FORECAST_START,\n", + " instance_sampler=instance_sampler,\n", + " past_length=module.model._past_length,\n", + " future_length=self.prediction_length,\n", + " time_series_fields=[\n", + " FieldName.FEAT_TIME,\n", + " FieldName.OBSERVED_VALUES,\n", + " ],\n", + " dummy_value=self.distr_output.value_in_support,\n", + " )\n", + "\n", + " def create_training_data_loader(\n", + " self,\n", + " data: Dataset,\n", + " module: TransformerLightningModule,\n", + " shuffle_buffer_length: Optional[int] = None,\n", + " **kwargs,\n", + " ) -> Iterable:\n", + " transformation = self._create_instance_splitter(\n", + " module, \"training\"\n", + " ) + SelectFields(TRAINING_INPUT_NAMES)\n", + "\n", + " training_instances = transformation.apply(\n", + " Cyclic(data)\n", + " if shuffle_buffer_length is None\n", + " else PseudoShuffled(\n", + " Cyclic(data), shuffle_buffer_length=shuffle_buffer_length\n", + " )\n", + " )\n", + "\n", + " return IterableSlice(\n", + " iter(\n", + " DataLoader(\n", + " IterableDataset(training_instances),\n", + " batch_size=self.batch_size,\n", + " **kwargs,\n", + " )\n", + " ),\n", + " self.num_batches_per_epoch,\n", + " )\n", + "\n", + " def create_validation_data_loader(\n", + " self,\n", + " data: Dataset,\n", + " module: TransformerLightningModule,\n", + " **kwargs,\n", + " ) -> Iterable:\n", + " transformation = self._create_instance_splitter(\n", + " module, \"validation\"\n", + " ) + SelectFields(TRAINING_INPUT_NAMES)\n", + "\n", + " validation_instances = transformation.apply(data)\n", + "\n", + " return DataLoader(\n", + " IterableDataset(validation_instances),\n", + " batch_size=self.batch_size,\n", + " **kwargs,\n", + " )\n", + " \n", + " def create_predictor(\n", + " self,\n", + " transformation: Transformation,\n", + " module: TransformerLightningModule,\n", + " ) -> PyTorchPredictor:\n", + " prediction_splitter = self._create_instance_splitter(module, \"test\")\n", + " \n", + " return PyTorchPredictor(\n", + " input_transform=transformation + prediction_splitter,\n", + " input_names=PREDICTION_INPUT_NAMES,\n", + " prediction_net=module.model,\n", + " batch_size=self.batch_size,\n", + " freq=self.freq,\n", + " prediction_length=self.prediction_length,\n", + " device=\"cuda\",\n", + " )\n", + "\n", + " def create_lightning_module(self) -> TransformerLightningModule:\n", + " model = TransformerModel(\n", + " freq=self.freq,\n", + " context_length=self.context_length,\n", + " prediction_length=self.prediction_length,\n", + " num_feat_dynamic_real=1 + self.num_feat_dynamic_real + len(self.time_features),\n", + " num_feat_static_real=max(1, self.num_feat_static_real),\n", + " num_feat_static_cat=max(1, self.num_feat_static_cat),\n", + " cardinality=self.cardinality,\n", + " embedding_dimension=self.embedding_dimension,\n", + "\n", + " # transformer arguments\n", + " nhead=self.nhead,\n", + " num_encoder_layers=self.num_encoder_layers,\n", + " num_decoder_layers=self.num_decoder_layers,\n", + " activation=self.activation,\n", + " dropout=self.dropout,\n", + " dim_feedforward=self.dim_feedforward,\n", + "\n", + " # univariate input\n", + " input_size=self.input_size,\n", + " distr_output=self.distr_output,\n", + " lags_seq=self.lags_seq,\n", + " scaling=self.scaling,\n", + " num_parallel_samples=self.num_parallel_samples,\n", + " )\n", + " \n", + " return TransformerLightningModule(model=model, loss=self.loss)" + ] + }, + { + "cell_type": "code", + "execution_count": 63, + "id": "17df7928", + "metadata": {}, + "outputs": [], + "source": [ + "dataset = get_dataset(\"traffic\")" + ] + }, + { + "cell_type": "code", + "execution_count": 64, + "id": "2ed78c2e", + "metadata": {}, + "outputs": [], + "source": [ + "estimator = TransformerEstimator(\n", + " freq=dataset.metadata.freq,\n", + " prediction_length=dataset.metadata.prediction_length,\n", + " context_length=2*dataset.metadata.prediction_length,\n", + " \n", + " nhead=1,\n", + " num_encoder_layers=2,\n", + " num_decoder_layers=2,\n", + " dim_feedforward=32,\n", + " activation=\"gelu\",\n", + " dropout=0.2,\n", + "\n", + " batch_size=32,\n", + " num_batches_per_epoch=100,\n", + " trainer_kwargs=dict(max_epochs=100, gpus='1', logger=CSVLogger(\"logs\", name=\"transformer\")),\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": 65, + "id": "635ee7af", + "metadata": { + "scrolled": true + }, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_60513/2550442513.py:109: DeprecationWarning: `np.long` is a deprecated alias for `np.compat.long`. To silence this warning, use `np.compat.long` by itself. In the likely event your code does not need to work on Python 2 you can use the builtin `int` for which `np.compat.long` is itself an alias. Doing this will not modify any behaviour and is safe. When replacing `np.long`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.\n", + "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n", + " dtype=np.long,\n", + "GPU available: True, used: True\n", + "TPU available: False, using: 0 TPU cores\n", + "IPU available: False, using: 0 IPUs\n", + "/home/kashif/.env/pytorch/lib/python3.8/site-packages/pytorch_lightning/utilities/parsing.py:104: UserWarning: attribute 'model' removed from hparams because it cannot be pickled\n", + " rank_zero_warn(f\"attribute '{k}' removed from hparams because it cannot be pickled\")\n", + "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n", + "/home/kashif/gluon-ts-PR/src/gluonts/transform/split.py:36: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " return _shift_timestamp_helper(ts, ts.freq, offset)\n", + "/home/kashif/gluon-ts-PR/src/gluonts/transform/feature.py:352: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " self._min_time_point, self._max_time_point, freq=start.freq\n", + "/home/kashif/gluon-ts-PR/src/gluonts/transform/split.py:36: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " return _shift_timestamp_helper(ts, ts.freq, offset)\n", + "/home/kashif/gluon-ts-PR/src/gluonts/transform/split.py:36: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " return _shift_timestamp_helper(ts, ts.freq, offset)\n", + "/home/kashif/gluon-ts-PR/src/gluonts/transform/feature.py:352: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " self._min_time_point, self._max_time_point, freq=start.freq\n", + "/home/kashif/gluon-ts-PR/src/gluonts/transform/feature.py:352: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " self._min_time_point, self._max_time_point, freq=start.freq\n", + "/home/kashif/gluon-ts-PR/src/gluonts/transform/split.py:36: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " return _shift_timestamp_helper(ts, ts.freq, offset)\n", + "/home/kashif/gluon-ts-PR/src/gluonts/transform/feature.py:352: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " self._min_time_point, self._max_time_point, freq=start.freq\n", + "/home/kashif/gluon-ts-PR/src/gluonts/transform/split.py:36: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " return _shift_timestamp_helper(ts, ts.freq, offset)\n", + "/home/kashif/gluon-ts-PR/src/gluonts/transform/feature.py:352: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " self._min_time_point, self._max_time_point, freq=start.freq\n", + "/home/kashif/gluon-ts-PR/src/gluonts/transform/split.py:36: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " return _shift_timestamp_helper(ts, ts.freq, offset)\n", + "/home/kashif/gluon-ts-PR/src/gluonts/transform/feature.py:352: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " self._min_time_point, self._max_time_point, freq=start.freq\n", + "/home/kashif/gluon-ts-PR/src/gluonts/transform/split.py:36: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " return _shift_timestamp_helper(ts, ts.freq, offset)\n", + "/home/kashif/gluon-ts-PR/src/gluonts/transform/feature.py:352: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " self._min_time_point, self._max_time_point, freq=start.freq\n", + "/home/kashif/gluon-ts-PR/src/gluonts/transform/split.py:36: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " return _shift_timestamp_helper(ts, ts.freq, offset)\n", + "/home/kashif/gluon-ts-PR/src/gluonts/transform/feature.py:352: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " self._min_time_point, self._max_time_point, freq=start.freq\n", + "\n", + " | Name | Type | Params\n", + "------------------------------------------------\n", + "0 | model | TransformerModel | 685 K \n", + "1 | loss | NegativeLogLikelihood | 0 \n", + "------------------------------------------------\n", + "685 K Trainable params\n", + "0 Non-trainable params\n", + "685 K Total params\n", + "2.742 Total estimated model params size (MB)\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "2699d52cbfea4c39a43c83f223b3a71b", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "Validation sanity check: 0it [00:00, ?it/s]" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/kashif/.env/pytorch/lib/python3.8/site-packages/pytorch_lightning/trainer/data_loading.py:453: UserWarning: Your `val_dataloader` has `shuffle=True`,it is strongly recommended that you turn this off for val/test/predict dataloaders.\n", + " rank_zero_warn(\n", + "/home/kashif/.env/pytorch/lib/python3.8/site-packages/pytorch_lightning/trainer/data_loading.py:111: UserWarning: The dataloader, val_dataloader 0, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 16 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n", + " rank_zero_warn(\n", + "/home/kashif/gluon-ts-PR/src/gluonts/transform/split.py:36: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " return _shift_timestamp_helper(ts, ts.freq, offset)\n", + "/home/kashif/gluon-ts-PR/src/gluonts/transform/feature.py:352: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " self._min_time_point, self._max_time_point, freq=start.freq\n" + ] + }, + { + "ename": "RuntimeError", + "evalue": "einsum(): operands do not broadcast with remapped shapes [original->remapped]: [32, 48, 48, 48]->[32, 48, 48, 48] [32, 24, 48]->[32, 24, 1, 48]", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mRuntimeError\u001b[0m Traceback (most recent call last)", + "\u001b[0;32m/tmp/ipykernel_60513/2175919108.py\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m predictor = estimator.train(\n\u001b[0m\u001b[1;32m 2\u001b[0m \u001b[0mtraining_data\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 3\u001b[0m \u001b[0mvalidation_data\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdataset\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0mnum_workers\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m8\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5\u001b[0m \u001b[0mshuffle_buffer_length\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1024\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/gluon-ts-PR/src/gluonts/torch/model/estimator.py\u001b[0m in \u001b[0;36mtrain\u001b[0;34m(self, training_data, validation_data, num_workers, shuffle_buffer_length, cache_data, **kwargs)\u001b[0m\n\u001b[1;32m 184\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 185\u001b[0m ) -> PyTorchPredictor:\n\u001b[0;32m--> 186\u001b[0;31m return self.train_model(\n\u001b[0m\u001b[1;32m 187\u001b[0m \u001b[0mtraining_data\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 188\u001b[0m \u001b[0mvalidation_data\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/gluon-ts-PR/src/gluonts/torch/model/estimator.py\u001b[0m in \u001b[0;36mtrain_model\u001b[0;34m(self, training_data, validation_data, num_workers, shuffle_buffer_length, cache_data, **kwargs)\u001b[0m\n\u001b[1;32m 153\u001b[0m \u001b[0mtrainer\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpl\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTrainer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mtrainer_kwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 154\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 155\u001b[0;31m trainer.fit(\n\u001b[0m\u001b[1;32m 156\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtraining_network\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 157\u001b[0m \u001b[0mtrain_dataloaders\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtraining_data_loader\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, model, train_dataloaders, val_dataloaders, datamodule, train_dataloader, ckpt_path)\u001b[0m\n\u001b[1;32m 735\u001b[0m )\n\u001b[1;32m 736\u001b[0m \u001b[0mtrain_dataloaders\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtrain_dataloader\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 737\u001b[0;31m self._call_and_handle_interrupt(\n\u001b[0m\u001b[1;32m 738\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_fit_impl\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtrain_dataloaders\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mval_dataloaders\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdatamodule\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mckpt_path\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 739\u001b[0m )\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py\u001b[0m in \u001b[0;36m_call_and_handle_interrupt\u001b[0;34m(self, trainer_fn, *args, **kwargs)\u001b[0m\n\u001b[1;32m 680\u001b[0m \"\"\"\n\u001b[1;32m 681\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 682\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mtrainer_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 683\u001b[0m \u001b[0;31m# TODO: treat KeyboardInterrupt as BaseException (delete the code below) in v1.7\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 684\u001b[0m \u001b[0;32mexcept\u001b[0m \u001b[0mKeyboardInterrupt\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mexception\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py\u001b[0m in \u001b[0;36m_fit_impl\u001b[0;34m(self, model, train_dataloaders, val_dataloaders, datamodule, ckpt_path)\u001b[0m\n\u001b[1;32m 770\u001b[0m \u001b[0;31m# TODO: ckpt_path only in v1.7\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 771\u001b[0m \u001b[0mckpt_path\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mckpt_path\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mresume_from_checkpoint\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 772\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_run\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mckpt_path\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mckpt_path\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 773\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 774\u001b[0m \u001b[0;32massert\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstate\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstopped\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py\u001b[0m in \u001b[0;36m_run\u001b[0;34m(self, model, ckpt_path)\u001b[0m\n\u001b[1;32m 1193\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1194\u001b[0m \u001b[0;31m# dispatch `start_training` or `start_evaluating` or `start_predicting`\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1195\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_dispatch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1196\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1197\u001b[0m \u001b[0;31m# plugin will finalized fitting (e.g. ddp_spawn will load trained model)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py\u001b[0m in \u001b[0;36m_dispatch\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 1272\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtraining_type_plugin\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstart_predicting\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1273\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1274\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtraining_type_plugin\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstart_training\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1275\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1276\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mrun_stage\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/pytorch_lightning/plugins/training_type/training_type_plugin.py\u001b[0m in \u001b[0;36mstart_training\u001b[0;34m(self, trainer)\u001b[0m\n\u001b[1;32m 200\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mstart_training\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtrainer\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;34m\"pl.Trainer\"\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 201\u001b[0m \u001b[0;31m# double dispatch to initiate the training loop\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 202\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_results\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtrainer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun_stage\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 203\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 204\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mstart_evaluating\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtrainer\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;34m\"pl.Trainer\"\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py\u001b[0m in \u001b[0;36mrun_stage\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 1282\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredicting\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1283\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_run_predict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1284\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_run_train\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1285\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1286\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m_pre_training_routine\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py\u001b[0m in \u001b[0;36m_run_train\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 1304\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprogress_bar_callback\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdisable\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1305\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1306\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_run_sanity_check\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlightning_module\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1307\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1308\u001b[0m \u001b[0;31m# enable train mode\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py\u001b[0m in \u001b[0;36m_run_sanity_check\u001b[0;34m(self, ref_model)\u001b[0m\n\u001b[1;32m 1368\u001b[0m \u001b[0;31m# run eval step\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1369\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mno_grad\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1370\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_evaluation_loop\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1371\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1372\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcall_hook\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"on_sanity_check_end\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/pytorch_lightning/loops/base.py\u001b[0m in \u001b[0;36mrun\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m 143\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 144\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mon_advance_start\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 145\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0madvance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 146\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mon_advance_end\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 147\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrestarting\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/pytorch_lightning/loops/dataloader/evaluation_loop.py\u001b[0m in \u001b[0;36madvance\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m 107\u001b[0m \u001b[0mdl_max_batches\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_max_batches\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mdataloader_idx\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 108\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 109\u001b[0;31m \u001b[0mdl_outputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mepoch_loop\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdataloader\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdataloader_idx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdl_max_batches\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnum_dataloaders\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 110\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 111\u001b[0m \u001b[0;31m# store batch level output per dataloader\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/pytorch_lightning/loops/base.py\u001b[0m in \u001b[0;36mrun\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m 143\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 144\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mon_advance_start\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 145\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0madvance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 146\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mon_advance_end\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 147\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrestarting\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/pytorch_lightning/loops/epoch/evaluation_epoch_loop.py\u001b[0m in \u001b[0;36madvance\u001b[0;34m(self, data_fetcher, dataloader_idx, dl_max_batches, num_dataloaders)\u001b[0m\n\u001b[1;32m 120\u001b[0m \u001b[0;31m# lightning module methods\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 121\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrainer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprofiler\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprofile\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"evaluation_step_and_end\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 122\u001b[0;31m \u001b[0moutput\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_evaluation_step\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbatch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch_idx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdataloader_idx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 123\u001b[0m \u001b[0moutput\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_evaluation_step_end\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 124\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/pytorch_lightning/loops/epoch/evaluation_epoch_loop.py\u001b[0m in \u001b[0;36m_evaluation_step\u001b[0;34m(self, batch, batch_idx, dataloader_idx)\u001b[0m\n\u001b[1;32m 215\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrainer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlightning_module\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_current_fx_name\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m\"validation_step\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 216\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrainer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprofiler\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprofile\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"validation_step\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 217\u001b[0;31m \u001b[0moutput\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrainer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0maccelerator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalidation_step\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mstep_kwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 218\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 219\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0moutput\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/pytorch_lightning/accelerators/accelerator.py\u001b[0m in \u001b[0;36mvalidation_step\u001b[0;34m(self, step_kwargs)\u001b[0m\n\u001b[1;32m 234\u001b[0m \"\"\"\n\u001b[1;32m 235\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprecision_plugin\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mval_step_context\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 236\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtraining_type_plugin\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalidation_step\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mstep_kwargs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 237\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 238\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mtest_step\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstep_kwargs\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mDict\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mstr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mUnion\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mAny\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mint\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0mOptional\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mSTEP_OUTPUT\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/pytorch_lightning/plugins/training_type/training_type_plugin.py\u001b[0m in \u001b[0;36mvalidation_step\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m 217\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 218\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mvalidation_step\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 219\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalidation_step\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 220\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 221\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mtest_step\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/tmp/ipykernel_60513/1381652195.py\u001b[0m in \u001b[0;36mvalidation_step\u001b[0;34m(self, batch, batch_idx)\u001b[0m\n\u001b[1;32m 29\u001b[0m \u001b[0;34m\"\"\"Execute validation step\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 30\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0minference_mode\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 31\u001b[0;31m \u001b[0mval_loss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbatch\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 32\u001b[0m self.log(\n\u001b[1;32m 33\u001b[0m \u001b[0;34m\"val_loss\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mval_loss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mon_epoch\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mon_step\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mprog_bar\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/mnt/scratch/kashif/pytorch/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m 1108\u001b[0m if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[1;32m 1109\u001b[0m or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1110\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1111\u001b[0m \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1112\u001b[0m \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/tmp/ipykernel_60513/1381652195.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, batch)\u001b[0m\n\u001b[1;32m 69\u001b[0m \u001b[0mfuture_target\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 70\u001b[0m )\n\u001b[0;32m---> 71\u001b[0;31m \u001b[0mparams\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutput_params\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtransformer_inputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 72\u001b[0m \u001b[0mdistr\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutput_distribution\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mparams\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mscale\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 73\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/tmp/ipykernel_60513/482295423.py\u001b[0m in \u001b[0;36moutput_params\u001b[0;34m(self, transformer_inputs)\u001b[0m\n\u001b[1;32m 298\u001b[0m \u001b[0mdec_input\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtransformer_inputs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcontext_length\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m...\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 299\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 300\u001b[0;31m \u001b[0mdec_output\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtransformer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msrc\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0menc_input\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtgt\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdec_input\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 301\u001b[0m \u001b[0;31m# enc_out = self.transformer.encoders(\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 302\u001b[0m \u001b[0;31m# enc_input\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/mnt/scratch/kashif/pytorch/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m 1108\u001b[0m if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[1;32m 1109\u001b[0m or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1110\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1111\u001b[0m \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1112\u001b[0m \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/xformers/factory/model_factory.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, src, tgt, encoder_input_mask, decoder_input_mask)\u001b[0m\n\u001b[1;32m 198\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 199\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mdecoder\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdecoders\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 200\u001b[0;31m tgt = decoder(\n\u001b[0m\u001b[1;32m 201\u001b[0m \u001b[0mtarget\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtgt\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 202\u001b[0m \u001b[0;31m# pyre-fixme[61]: `memory` is not always initialized here.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/mnt/scratch/kashif/pytorch/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m 1108\u001b[0m if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[1;32m 1109\u001b[0m or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1110\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1111\u001b[0m \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1112\u001b[0m \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/xformers/factory/block_factory.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, target, memory, encoder_att_mask, decoder_att_mask, input_mask)\u001b[0m\n\u001b[1;32m 366\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 367\u001b[0m \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwrap_att\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mtarget_q\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtarget_k\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtarget_v\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0matt_mask\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdecoder_att_mask\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 368\u001b[0;31m \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwrap_cross\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmemory\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmemory\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0matt_mask\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mencoder_att_mask\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 369\u001b[0m \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwrap_ff\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 370\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/mnt/scratch/kashif/pytorch/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m 1108\u001b[0m if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[1;32m 1109\u001b[0m or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1110\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1111\u001b[0m \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1112\u001b[0m \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/xformers/components/residual.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, inputs, *args, **kwargs)\u001b[0m\n\u001b[1;32m 55\u001b[0m \u001b[0minputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_to_tensor_list\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 56\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 57\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlayer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 58\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 59\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/mnt/scratch/kashif/pytorch/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m 1108\u001b[0m if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[1;32m 1109\u001b[0m or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1110\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1111\u001b[0m \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1112\u001b[0m \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/xformers/components/residual.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, inputs, *args, **kwargs)\u001b[0m\n\u001b[1;32m 76\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 77\u001b[0m \u001b[0mx_norm\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnorm\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx_\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mx_\u001b[0m \u001b[0;32min\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 78\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msublayer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mx_norm\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 79\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 80\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/mnt/scratch/kashif/pytorch/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m 1108\u001b[0m if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[1;32m 1109\u001b[0m or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1110\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1111\u001b[0m \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1112\u001b[0m \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/xformers/components/multi_head_dispatch.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, query, key, value, att_mask)\u001b[0m\n\u001b[1;32m 176\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 177\u001b[0m \u001b[0;31m# Self-attend\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 178\u001b[0;31m \u001b[0my\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mattention\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mq\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mq\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mk\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mk\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mv\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mv\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0matt_mask\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0matt_mask\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 179\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 180\u001b[0m \u001b[0;31m# Re-assemble all head outputs side by side\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/mnt/scratch/kashif/pytorch/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m 1108\u001b[0m if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[1;32m 1109\u001b[0m or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1110\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1111\u001b[0m \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1112\u001b[0m \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/xformers/components/attention/favor.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, q, k, v, *_, **__)\u001b[0m\n\u001b[1;32m 154\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 155\u001b[0m \u001b[0;31m# Actually compute attention\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 156\u001b[0;31m \u001b[0matt_raw\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0matt_normalization\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_causal_attention\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mk_prime\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mq_prime\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mv\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 157\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 158\u001b[0m \u001b[0;31m# Normalize\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m~/.env/pytorch/lib/python3.8/site-packages/xformers/components/attention/favor.py\u001b[0m in \u001b[0;36m_causal_attention\u001b[0;34m(k_prime, q_prime, v)\u001b[0m\n\u001b[1;32m 117\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 118\u001b[0m \u001b[0;31m# Consolidate against the feature dimension\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 119\u001b[0;31m \u001b[0matt_raw\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0meinsum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"bcfe,bcf->bce\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mGps\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mq_prime\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 120\u001b[0m \u001b[0matt_norm\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0meinsum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"bcfe,bcf->bce\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mGrenorm\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mq_prime\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 121\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;32m/mnt/scratch/kashif/pytorch/torch/functional.py\u001b[0m in \u001b[0;36meinsum\u001b[0;34m(*args)\u001b[0m\n\u001b[1;32m 328\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0meinsum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mequation\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0m_operands\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 329\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 330\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0m_VF\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0meinsum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mequation\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moperands\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;31m# type: ignore[attr-defined]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 331\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 332\u001b[0m \u001b[0;31m# Wrapper around _histogramdd and _histogramdd_bin_edges needed due to (Tensor, Tensor[]) return type.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0;31mRuntimeError\u001b[0m: einsum(): operands do not broadcast with remapped shapes [original->remapped]: [32, 48, 48, 48]->[32, 48, 48, 48] [32, 24, 48]->[32, 24, 1, 48]" + ] + } + ], + "source": [ + "predictor = estimator.train(\n", + " training_data=dataset.train,\n", + " validation_data=dataset.train,\n", + " num_workers=8,\n", + " shuffle_buffer_length=1024\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": 51, + "id": "ec16ae1e", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "> \u001b[0;32m/mnt/scratch/kashif/pytorch/torch/functional.py\u001b[0m(330)\u001b[0;36meinsum\u001b[0;34m()\u001b[0m\n", + "\u001b[0;32m 328 \u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0meinsum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mequation\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0m_operands\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 329 \u001b[0;31m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m--> 330 \u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0m_VF\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0meinsum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mequation\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moperands\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;31m# type: ignore[attr-defined]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 331 \u001b[0;31m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 332 \u001b[0;31m\u001b[0;31m# Wrapper around _histogramdd and _histogramdd_bin_edges needed due to (Tensor, Tensor[]) return type.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\n", + "ipdb> up\n", + "> \u001b[0;32m/home/kashif/.env/pytorch/lib/python3.8/site-packages/xformers/components/attention/favor.py\u001b[0m(119)\u001b[0;36m_causal_attention\u001b[0;34m()\u001b[0m\n", + "\u001b[0;32m 117 \u001b[0;31m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 118 \u001b[0;31m \u001b[0;31m# Consolidate against the feature dimension\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m--> 119 \u001b[0;31m \u001b[0matt_raw\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0meinsum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"bcfe,bcf->bce\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mGps\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mq_prime\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 120 \u001b[0;31m \u001b[0matt_norm\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0meinsum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"bcfe,bcf->bce\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mGrenorm\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mq_prime\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 121 \u001b[0;31m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\n", + "ipdb> up\n", + "> \u001b[0;32m/home/kashif/.env/pytorch/lib/python3.8/site-packages/xformers/components/attention/favor.py\u001b[0m(156)\u001b[0;36mforward\u001b[0;34m()\u001b[0m\n", + "\u001b[0;32m 154 \u001b[0;31m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 155 \u001b[0;31m \u001b[0;31m# Actually compute attention\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m--> 156 \u001b[0;31m \u001b[0matt_raw\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0matt_normalization\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_causal_attention\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mk_prime\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mq_prime\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mv\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 157 \u001b[0;31m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 158 \u001b[0;31m \u001b[0;31m# Normalize\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\n", + "ipdb> up\n", + "> \u001b[0;32m/mnt/scratch/kashif/pytorch/torch/nn/modules/module.py\u001b[0m(1110)\u001b[0;36m_call_impl\u001b[0;34m()\u001b[0m\n", + "\u001b[0;32m 1108 \u001b[0;31m if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n", + "\u001b[0m\u001b[0;32m 1109 \u001b[0;31m or _global_forward_hooks or _global_forward_pre_hooks):\n", + "\u001b[0m\u001b[0;32m-> 1110 \u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 1111 \u001b[0;31m \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 1112 \u001b[0;31m \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\n", + "ipdb> up\n", + "> \u001b[0;32m/home/kashif/.env/pytorch/lib/python3.8/site-packages/xformers/components/multi_head_dispatch.py\u001b[0m(178)\u001b[0;36mforward\u001b[0;34m()\u001b[0m\n", + "\u001b[0;32m 176 \u001b[0;31m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 177 \u001b[0;31m \u001b[0;31m# Self-attend\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m--> 178 \u001b[0;31m \u001b[0my\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mattention\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mq\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mq\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mk\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mk\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mv\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mv\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0matt_mask\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0matt_mask\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 179 \u001b[0;31m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 180 \u001b[0;31m \u001b[0;31m# Re-assemble all head outputs side by side\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\n", + "ipdb> up\n", + "> \u001b[0;32m/mnt/scratch/kashif/pytorch/torch/nn/modules/module.py\u001b[0m(1110)\u001b[0;36m_call_impl\u001b[0;34m()\u001b[0m\n", + "\u001b[0;32m 1108 \u001b[0;31m if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n", + "\u001b[0m\u001b[0;32m 1109 \u001b[0;31m or _global_forward_hooks or _global_forward_pre_hooks):\n", + "\u001b[0m\u001b[0;32m-> 1110 \u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 1111 \u001b[0;31m \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 1112 \u001b[0;31m \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\n", + "ipdb> up\n", + "> \u001b[0;32m/home/kashif/.env/pytorch/lib/python3.8/site-packages/xformers/components/residual.py\u001b[0m(78)\u001b[0;36mforward\u001b[0;34m()\u001b[0m\n", + "\u001b[0;32m 76 \u001b[0;31m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 77 \u001b[0;31m \u001b[0mx_norm\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnorm\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx_\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mx_\u001b[0m \u001b[0;32min\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m---> 78 \u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msublayer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mx_norm\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 79 \u001b[0;31m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 80 \u001b[0;31m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\n", + "ipdb> up\n", + "> \u001b[0;32m/mnt/scratch/kashif/pytorch/torch/nn/modules/module.py\u001b[0m(1110)\u001b[0;36m_call_impl\u001b[0;34m()\u001b[0m\n", + "\u001b[0;32m 1108 \u001b[0;31m if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n", + "\u001b[0m\u001b[0;32m 1109 \u001b[0;31m or _global_forward_hooks or _global_forward_pre_hooks):\n", + "\u001b[0m\u001b[0;32m-> 1110 \u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 1111 \u001b[0;31m \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 1112 \u001b[0;31m \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\n", + "ipdb> up\n", + "> \u001b[0;32m/home/kashif/.env/pytorch/lib/python3.8/site-packages/xformers/components/residual.py\u001b[0m(57)\u001b[0;36mforward\u001b[0;34m()\u001b[0m\n", + "\u001b[0;32m 55 \u001b[0;31m \u001b[0minputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_to_tensor_list\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 56 \u001b[0;31m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m---> 57 \u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0minputs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlayer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 58 \u001b[0;31m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 59 \u001b[0;31m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\n", + "ipdb> up\n", + "> \u001b[0;32m/mnt/scratch/kashif/pytorch/torch/nn/modules/module.py\u001b[0m(1110)\u001b[0;36m_call_impl\u001b[0;34m()\u001b[0m\n", + "\u001b[0;32m 1108 \u001b[0;31m if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n", + "\u001b[0m\u001b[0;32m 1109 \u001b[0;31m or _global_forward_hooks or _global_forward_pre_hooks):\n", + "\u001b[0m\u001b[0;32m-> 1110 \u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 1111 \u001b[0;31m \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 1112 \u001b[0;31m \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\n", + "ipdb> up\n", + "> \u001b[0;32m/home/kashif/.env/pytorch/lib/python3.8/site-packages/xformers/factory/block_factory.py\u001b[0m(368)\u001b[0;36mforward\u001b[0;34m()\u001b[0m\n", + "\u001b[0;32m 366 \u001b[0;31m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 367 \u001b[0;31m \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwrap_att\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mtarget_q\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtarget_k\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtarget_v\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0matt_mask\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdecoder_att_mask\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m--> 368 \u001b[0;31m \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwrap_cross\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmemory\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmemory\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0matt_mask\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mencoder_att_mask\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 369 \u001b[0;31m \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwrap_ff\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 370 \u001b[0;31m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "ipdb> up\n", + "> \u001b[0;32m/mnt/scratch/kashif/pytorch/torch/nn/modules/module.py\u001b[0m(1110)\u001b[0;36m_call_impl\u001b[0;34m()\u001b[0m\n", + "\u001b[0;32m 1108 \u001b[0;31m if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n", + "\u001b[0m\u001b[0;32m 1109 \u001b[0;31m or _global_forward_hooks or _global_forward_pre_hooks):\n", + "\u001b[0m\u001b[0;32m-> 1110 \u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 1111 \u001b[0;31m \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 1112 \u001b[0;31m \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\n", + "ipdb> up\n", + "> \u001b[0;32m/home/kashif/.env/pytorch/lib/python3.8/site-packages/xformers/factory/model_factory.py\u001b[0m(200)\u001b[0;36mforward\u001b[0;34m()\u001b[0m\n", + "\u001b[0;32m 198 \u001b[0;31m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 199 \u001b[0;31m \u001b[0;32mfor\u001b[0m \u001b[0mdecoder\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdecoders\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m--> 200 \u001b[0;31m tgt = decoder(\n", + "\u001b[0m\u001b[0;32m 201 \u001b[0;31m \u001b[0mtarget\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtgt\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\u001b[0;32m 202 \u001b[0;31m \u001b[0;31m# pyre-fixme[61]: `memory` is not always initialized here.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", + "\u001b[0m\n", + "ipdb> decoder\n", + "xFormerDecoderBlock(\n", + " (mha): MultiHeadDispatch(\n", + " (attention): NystromAttention(\n", + " (attn_drop): Dropout(p=0.2, inplace=False)\n", + " )\n", + " (in_proj_container): InProjContainer()\n", + " (resid_drop): Dropout(p=0.2, inplace=False)\n", + " (proj): Linear(in_features=48, out_features=48, bias=True)\n", + " )\n", + " (cross_mha): MultiHeadDispatch(\n", + " (attention): FavorAttention(\n", + " (attn_drop): Dropout(p=0.2, inplace=True)\n", + " (feature_map_query): SMReg()\n", + " (feature_map_key): SMReg()\n", + " )\n", + " (in_proj_container): InProjContainer()\n", + " (resid_drop): Dropout(p=0.2, inplace=False)\n", + " (proj): Linear(in_features=48, out_features=48, bias=True)\n", + " )\n", + " (feedforward): MLP(\n", + " (mlp): Sequential(\n", + " (0): Linear(in_features=48, out_features=96, bias=True)\n", + " (1): GELU()\n", + " (2): Dropout(p=0.2, inplace=False)\n", + " (3): Linear(in_features=96, out_features=48, bias=True)\n", + " (4): Dropout(p=0.2, inplace=False)\n", + " )\n", + " )\n", + " (wrap_att): Residual(\n", + " (layer): PreNorm(\n", + " (norm): FusedLayerNorm()\n", + " (sublayer): MultiHeadDispatch(\n", + " (attention): NystromAttention(\n", + " (attn_drop): Dropout(p=0.2, inplace=False)\n", + " )\n", + " (in_proj_container): InProjContainer()\n", + " (resid_drop): Dropout(p=0.2, inplace=False)\n", + " (proj): Linear(in_features=48, out_features=48, bias=True)\n", + " )\n", + " )\n", + " )\n", + " (wrap_cross): Residual(\n", + " (layer): PreNorm(\n", + " (norm): FusedLayerNorm()\n", + " (sublayer): MultiHeadDispatch(\n", + " (attention): FavorAttention(\n", + " (attn_drop): Dropout(p=0.2, inplace=True)\n", + " (feature_map_query): SMReg()\n", + " (feature_map_key): SMReg()\n", + " )\n", + " (in_proj_container): InProjContainer()\n", + " (resid_drop): Dropout(p=0.2, inplace=False)\n", + " (proj): Linear(in_features=48, out_features=48, bias=True)\n", + " )\n", + " )\n", + " )\n", + " (wrap_ff): Residual(\n", + " (layer): PreNorm(\n", + " (norm): FusedLayerNorm()\n", + " (sublayer): MLP(\n", + " (mlp): Sequential(\n", + " (0): Linear(in_features=48, out_features=96, bias=True)\n", + " (1): GELU()\n", + " (2): Dropout(p=0.2, inplace=False)\n", + " (3): Linear(in_features=96, out_features=48, bias=True)\n", + " (4): Dropout(p=0.2, inplace=False)\n", + " )\n", + " )\n", + " )\n", + " )\n", + ")\n", + "ipdb> exit()\n" + ] + } + ], + "source": [ + "%debug" + ] + }, + { + "cell_type": "code", + "execution_count": 81, + "id": "8f776698", + "metadata": {}, + "outputs": [], + "source": [ + "forecast_it, ts_it = make_evaluation_predictions(\n", + " dataset=dataset.test,\n", + " predictor=predictor,\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": 82, + "id": "8a162f3c", + "metadata": { + "scrolled": true + }, + "outputs": [], + "source": [ + "forecasts = list(forecast_it)" + ] + }, + { + "cell_type": "code", + "execution_count": 83, + "id": "50635a96", + "metadata": {}, + "outputs": [], + "source": [ + "tss = list(ts_it)" + ] + }, + { + "cell_type": "code", + "execution_count": 84, + "id": "cadfc930", + "metadata": {}, + "outputs": [], + "source": [ + "evaluator = Evaluator()" + ] + }, + { + "cell_type": "code", + "execution_count": 85, + "id": "b4ea2090", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Running evaluation: 266it [00:00, 1089.83it/s]/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/_base.py:305: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " date_before_forecast = forecast.index[0] - forecast.index[0].freq\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/_base.py:305: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " date_before_forecast = forecast.index[0] - forecast.index[0].freq\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/_base.py:305: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " date_before_forecast = forecast.index[0] - forecast.index[0].freq\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/_base.py:305: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " date_before_forecast = forecast.index[0] - forecast.index[0].freq\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/_base.py:305: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " date_before_forecast = forecast.index[0] - forecast.index[0].freq\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/_base.py:305: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " date_before_forecast = forecast.index[0] - forecast.index[0].freq\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/_base.py:305: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " date_before_forecast = forecast.index[0] - forecast.index[0].freq\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/_base.py:305: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " date_before_forecast = forecast.index[0] - forecast.index[0].freq\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/_base.py:305: FutureWarning: Timestamp.freq is deprecated and will be removed in a future version\n", + " date_before_forecast = forecast.index[0] - forecast.index[0].freq\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/metrics.py:102: RuntimeWarning: divide by zero encountered in double_scalars\n", + " return np.mean(np.abs(target - forecast)) / seasonal_error\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/metrics.py:102: RuntimeWarning: divide by zero encountered in double_scalars\n", + " return np.mean(np.abs(target - forecast)) / seasonal_error\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/metrics.py:102: RuntimeWarning: divide by zero encountered in double_scalars\n", + " return np.mean(np.abs(target - forecast)) / seasonal_error\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/metrics.py:102: RuntimeWarning: divide by zero encountered in double_scalars\n", + " return np.mean(np.abs(target - forecast)) / seasonal_error\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/metrics.py:150: RuntimeWarning: divide by zero encountered in double_scalars\n", + " return numerator / seasonal_error\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/metrics.py:150: RuntimeWarning: divide by zero encountered in double_scalars\n", + " return numerator / seasonal_error\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/metrics.py:150: RuntimeWarning: divide by zero encountered in double_scalars\n", + " return numerator / seasonal_error\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/metrics.py:150: RuntimeWarning: divide by zero encountered in double_scalars\n", + " return numerator / seasonal_error\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/metrics.py:102: RuntimeWarning: divide by zero encountered in double_scalars\n", + " return np.mean(np.abs(target - forecast)) / seasonal_error\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/metrics.py:150: RuntimeWarning: divide by zero encountered in double_scalars\n", + " return numerator / seasonal_error\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/metrics.py:102: RuntimeWarning: divide by zero encountered in double_scalars\n", + " return np.mean(np.abs(target - forecast)) / seasonal_error\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/metrics.py:150: RuntimeWarning: divide by zero encountered in double_scalars\n", + " return numerator / seasonal_error\n", + "\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/metrics.py:102: RuntimeWarning: divide by zero encountered in double_scalars\n", + " return np.mean(np.abs(target - forecast)) / seasonal_error\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/metrics.py:150: RuntimeWarning: divide by zero encountered in double_scalars\n", + " return numerator / seasonal_error\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/metrics.py:102: RuntimeWarning: divide by zero encountered in double_scalars\n", + " return np.mean(np.abs(target - forecast)) / seasonal_error\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/metrics.py:150: RuntimeWarning: divide by zero encountered in double_scalars\n", + " return numerator / seasonal_error\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/metrics.py:102: RuntimeWarning: divide by zero encountered in double_scalars\n", + " return np.mean(np.abs(target - forecast)) / seasonal_error\n", + "/mnt/scratch/kashif/gluon-ts/src/gluonts/evaluation/metrics.py:150: RuntimeWarning: divide by zero encountered in double_scalars\n", + " return numerator / seasonal_error\n", + "/home/kashif/.env/pytorch/lib/python3.8/site-packages/pandas/core/construction.py:759: UserWarning: Warning: converting a masked element to nan.\n", + " subarr = np.array(arr, dtype=dtype, copy=copy)\n" + ] + } + ], + "source": [ + "agg_metrics, ts_metrics = evaluator(iter(tss), iter(forecasts))" + ] + }, + { + "cell_type": "code", + "execution_count": 86, + "id": "998dfc78", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "{'MSE': 4005462.1576438956,\n", + " 'abs_error': 2070024.706854701,\n", + " 'abs_target_sum': 21209029.0,\n", + " 'abs_target_mean': 2657.77305764411,\n", + " 'seasonal_error': 12.825281435633281,\n", + " 'MASE': inf,\n", + " 'MAPE': 0.14089255149942254,\n", + " 'sMAPE': 0.4196852070683226,\n", + " 'MSIS': inf,\n", + " 'QuantileLoss[0.1]': 808467.211132447,\n", + " 'Coverage[0.1]': 0.24185463659147874,\n", + " 'QuantileLoss[0.2]': 1278887.5805553913,\n", + " 'Coverage[0.2]': 0.36829573934837095,\n", + " 'QuantileLoss[0.3]': 1594186.2012551553,\n", + " 'Coverage[0.3]': 0.44172932330827075,\n", + " 'QuantileLoss[0.4]': 1853766.8644387266,\n", + " 'Coverage[0.4]': 0.5226817042606516,\n", + " 'QuantileLoss[0.5]': 2070024.6042572262,\n", + " 'Coverage[0.5]': 0.6552631578947369,\n", + " 'QuantileLoss[0.6]': 2180911.444006557,\n", + " 'Coverage[0.6]': 0.7090225563909774,\n", + " 'QuantileLoss[0.7]': 2142682.1592610786,\n", + " 'Coverage[0.7]': 0.7506265664160402,\n", + " 'QuantileLoss[0.8]': 2049024.134861926,\n", + " 'Coverage[0.8]': 0.7953634085213033,\n", + " 'QuantileLoss[0.9]': 1718981.390833828,\n", + " 'Coverage[0.9]': 0.8374686716791979,\n", + " 'RMSE': 2001.3650735545216,\n", + " 'NRMSE': 0.7530233131825641,\n", + " 'ND': 0.09760110690851057,\n", + " 'wQuantileLoss[0.1]': 0.03811901106516696,\n", + " 'wQuantileLoss[0.2]': 0.06029920467152887,\n", + " 'wQuantileLoss[0.3]': 0.07516544964199706,\n", + " 'wQuantileLoss[0.4]': 0.08740460793555077,\n", + " 'wQuantileLoss[0.5]': 0.09760110207106729,\n", + " 'wQuantileLoss[0.6]': 0.10282938667331527,\n", + " 'wQuantileLoss[0.7]': 0.10102688620309203,\n", + " 'wQuantileLoss[0.8]': 0.09661093560020716,\n", + " 'wQuantileLoss[0.9]': 0.08104950918940362,\n", + " 'mean_absolute_QuantileLoss': 1744103.5100669262,\n", + " 'mean_wQuantileLoss': 0.08223401033903656,\n", + " 'MAE_Coverage': 0.1062935115566695,\n", + " 'OWA': nan}" + ] + }, + "execution_count": 86, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "agg_metrics" + ] + }, + { + "cell_type": "code", + "execution_count": 119, + "id": "ce74d777", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "{'MSE': 2799751.1131833945,\n", + " 'abs_error': 8817300.178627014,\n", + " 'abs_target_sum': 128632956.0,\n", + " 'abs_target_mean': 2385.272140631954,\n", + " 'seasonal_error': 189.49338196116761,\n", + " 'MASE': 0.7286814231375043,\n", + " 'MAPE': 0.09201469989787364,\n", + " 'sMAPE': 0.1058379271666041,\n", + " 'MSIS': 6.154579991270963,\n", + " 'QuantileLoss[0.1]': 3854111.318776969,\n", + " 'Coverage[0.1]': 0.07155837412846758,\n", + " 'QuantileLoss[0.2]': 6027256.327012077,\n", + " 'Coverage[0.2]': 0.1468253968253968,\n", + " 'QuantileLoss[0.3]': 7489274.377354384,\n", + " 'Coverage[0.3]': 0.23153093012906098,\n", + " 'QuantileLoss[0.4]': 8386504.2699774,\n", + " 'Coverage[0.4]': 0.32309746328437916,\n", + " 'QuantileLoss[0.5]': 8817300.231568864,\n", + " 'Coverage[0.5]': 0.42143228007713984,\n", + " 'QuantileLoss[0.6]': 8857007.398170901,\n", + " 'Coverage[0.6]': 0.5182465509568314,\n", + " 'QuantileLoss[0.7]': 8313090.016379234,\n", + " 'Coverage[0.7]': 0.6296172674677347,\n", + " 'QuantileLoss[0.8]': 7147210.871046787,\n", + " 'Coverage[0.8]': 0.744288681204569,\n", + " 'QuantileLoss[0.9]': 4967439.278312072,\n", + " 'Coverage[0.9]': 0.8599243435692033,\n", + " 'RMSE': 1673.24568225452,\n", + " 'NRMSE': 0.7014904730372654,\n", + " 'ND': 0.06854619883435636,\n", + " 'wQuantileLoss[0.1]': 0.02996208311326507,\n", + " 'wQuantileLoss[0.2]': 0.04685623742497278,\n", + " 'wQuantileLoss[0.3]': 0.05822204985598235,\n", + " 'wQuantileLoss[0.4]': 0.06519716665748862,\n", + " 'wQuantileLoss[0.5]': 0.06854619924592935,\n", + " 'wQuantileLoss[0.6]': 0.06885488504338579,\n", + " 'wQuantileLoss[0.7]': 0.06462644002660745,\n", + " 'wQuantileLoss[0.8]': 0.05556282847955999,\n", + " 'wQuantileLoss[0.9]': 0.038617158718735126,\n", + " 'mean_absolute_QuantileLoss': 7095466.009844299,\n", + " 'mean_wQuantileLoss': 0.05516056095176961,\n", + " 'MAE_Coverage': 0.06149763470635747,\n", + " 'OWA': nan}" + ] + }, + "execution_count": 119, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "agg_metrics" + ] + }, + { + "cell_type": "code", + "execution_count": 87, + "id": "10ea7d83", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "plt.figure(figsize=(20, 15))\n", + "date_formater = mdates.DateFormatter('%b, %d')\n", + "plt.rcParams.update({'font.size': 15})\n", + "\n", + "for idx, (forecast, ts) in islice(enumerate(zip(forecasts, tss)), 9):\n", + " ax = plt.subplot(3, 3, idx+1)\n", + "\n", + " plt.plot(ts[-4 * dataset.metadata.prediction_length:], label=\"target\", )\n", + " forecast.plot( color='g')\n", + " plt.xticks(rotation=60)\n", + " ax.xaxis.set_major_formatter(date_formater)\n", + "\n", + "plt.gcf().tight_layout()\n", + "plt.legend()\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 88, + "id": "59b42f21", + "metadata": {}, + "outputs": [], + "source": [ + "def plot_prob_forecasts(ts_entry, forecast_entry):\n", + " plot_length = 50\n", + " prediction_intervals = (50.0, 90.0)\n", + " legend = [\"observations\", \"median prediction\"] + [f\"{k}% prediction interval\" for k in prediction_intervals][::-1]\n", + "\n", + " fig, ax = plt.subplots(1, 1, figsize=(10, 7))\n", + " ts_entry[-plot_length:].plot(ax=ax) # plot the time series\n", + " forecast_entry.plot(prediction_intervals=prediction_intervals, color='g')\n", + " plt.grid(which=\"both\")\n", + " plt.legend(legend, loc=\"upper left\")\n", + " plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 91, + "id": "35eda4db", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "index = 136\n", + "plot_prob_forecasts(tss[index], forecasts[index])" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "a1f978b1", + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.9.10" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +}