diff --git a/nbs/models.tft.ipynb b/nbs/models.tft.ipynb
index ec38bf09a..2207fc64d 100644
--- a/nbs/models.tft.ipynb
+++ b/nbs/models.tft.ipynb
@@ -1,5 +1,22 @@
{
"cells": [
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "env: PYTORCH_ENABLE_MPS_FALLBACK=1\n"
+ ]
+ }
+ ],
+ "source": [
+ "%set_env PYTORCH_ENABLE_MPS_FALLBACK=1"
+ ]
+ },
{
"cell_type": "code",
"execution_count": null,
@@ -44,7 +61,7 @@
"outputs": [],
"source": [
"#| export\n",
- "from typing import Tuple, Optional\n",
+ "from typing import Tuple, Optional, Callable\n",
"\n",
"import torch\n",
"import torch.nn as nn\n",
@@ -124,6 +141,19 @@
"outputs": [],
"source": [
"#| exporti\n",
+ "def get_activation_fn(activation_str: str) -> Callable:\n",
+ " activation_map = {\n",
+ " 'ReLU': F.relu,\n",
+ " 'Softplus': F.softplus,\n",
+ " 'Tanh': F.tanh,\n",
+ " 'SELU': F.selu,\n",
+ " 'LeakyReLU': F.leaky_relu,\n",
+ " 'Sigmoid': F.sigmoid,\n",
+ " 'ELU': F.elu,\n",
+ " 'GLU': F.glu\n",
+ " }\n",
+ " return activation_map.get(activation_str, F.elu)\n",
+ "\n",
"class MaybeLayerNorm(nn.Module):\n",
" def __init__(self, output_size, hidden_size, eps):\n",
" super().__init__()\n",
@@ -149,12 +179,12 @@
"class GRN(nn.Module):\n",
" def __init__(self,\n",
" input_size,\n",
- " hidden_size, \n",
+ " hidden_size,\n",
" output_size=None,\n",
" context_hidden_size=None,\n",
- " dropout=0):\n",
+ " dropout=0,\n",
+ " activation='ELU',):\n",
" super().__init__()\n",
- " \n",
" self.layer_norm = MaybeLayerNorm(output_size, hidden_size, eps=1e-3)\n",
" self.lin_a = nn.Linear(input_size, hidden_size)\n",
" if context_hidden_size is not None:\n",
@@ -163,12 +193,13 @@
" self.glu = GLU(hidden_size, output_size if output_size else hidden_size)\n",
" self.dropout = nn.Dropout(dropout)\n",
" self.out_proj = nn.Linear(input_size, output_size) if output_size else None\n",
- "\n",
+ " self.activation_fn = get_activation_fn(activation)\n",
+ " \n",
" def forward(self, a: Tensor, c: Optional[Tensor] = None):\n",
" x = self.lin_a(a)\n",
" if c is not None:\n",
" x = x + self.lin_c(c).unsqueeze(1)\n",
- " x = F.elu(x)\n",
+ " x = self.activation_fn(x)\n",
" x = self.lin_i(x)\n",
" x = self.dropout(x)\n",
" x = self.glu(x)\n",
@@ -292,15 +323,16 @@
" return s_inp, k_inp, o_inp, target_inp\n",
"\n",
"class VariableSelectionNetwork(nn.Module):\n",
- " def __init__(self, hidden_size, num_inputs, dropout):\n",
+ " def __init__(self, hidden_size, num_inputs, dropout, grn_activation):\n",
" super().__init__()\n",
" self.joint_grn = GRN(input_size=hidden_size*num_inputs, \n",
" hidden_size=hidden_size, \n",
" output_size=num_inputs, \n",
- " context_hidden_size=hidden_size)\n",
+ " context_hidden_size=hidden_size,\n",
+ " activation=grn_activation)\n",
" self.var_grns = nn.ModuleList(\n",
" [GRN(input_size=hidden_size, \n",
- " hidden_size=hidden_size, dropout=dropout)\n",
+ " hidden_size=hidden_size, dropout=dropout, activation=grn_activation)\n",
" for _ in range(num_inputs)])\n",
"\n",
" def forward(self, x: Tensor, context: Optional[Tensor] = None):\n",
@@ -445,10 +477,10 @@
"source": [
"#| exporti\n",
"class StaticCovariateEncoder(nn.Module):\n",
- " def __init__(self, hidden_size, num_static_vars, dropout):\n",
+ " def __init__(self, hidden_size, num_static_vars, dropout, grn_activation):\n",
" super().__init__()\n",
" self.vsn = VariableSelectionNetwork(\n",
- " hidden_size=hidden_size, num_inputs=num_static_vars, dropout=dropout\n",
+ " hidden_size=hidden_size, num_inputs=num_static_vars, dropout=dropout, grn_activation=grn_activation\n",
" )\n",
" self.context_grns = nn.ModuleList(\n",
" [\n",
@@ -502,18 +534,18 @@
"source": [
"#| exporti\n",
"class TemporalCovariateEncoder(nn.Module):\n",
- " def __init__(self, hidden_size, num_historic_vars, num_future_vars, dropout):\n",
+ " def __init__(self, hidden_size, num_historic_vars, num_future_vars, dropout, grn_activation):\n",
" super(TemporalCovariateEncoder, self).__init__()\n",
"\n",
" self.history_vsn = VariableSelectionNetwork(\n",
- " hidden_size=hidden_size, num_inputs=num_historic_vars, dropout=dropout\n",
+ " hidden_size=hidden_size, num_inputs=num_historic_vars, dropout=dropout, grn_activation=grn_activation\n",
" )\n",
" self.history_encoder = nn.LSTM(\n",
" input_size=hidden_size, hidden_size=hidden_size, batch_first=True\n",
" )\n",
"\n",
" self.future_vsn = VariableSelectionNetwork(\n",
- " hidden_size=hidden_size, num_inputs=num_future_vars, dropout=dropout\n",
+ " hidden_size=hidden_size, num_inputs=num_future_vars, dropout=dropout, grn_activation=grn_activation\n",
" )\n",
" self.future_encoder = nn.LSTM(\n",
" input_size=hidden_size, hidden_size=hidden_size, batch_first=True\n",
@@ -567,7 +599,7 @@
"#| exporti\n",
"class TemporalFusionDecoder(nn.Module):\n",
" def __init__(\n",
- " self, n_head, hidden_size, example_length, encoder_length, attn_dropout, dropout\n",
+ " self, n_head, hidden_size, example_length, encoder_length, attn_dropout, dropout, grn_activation\n",
" ):\n",
" super(TemporalFusionDecoder, self).__init__()\n",
" self.encoder_length = encoder_length\n",
@@ -578,6 +610,7 @@
" hidden_size=hidden_size,\n",
" context_hidden_size=hidden_size,\n",
" dropout=dropout,\n",
+ " activation=grn_activation\n",
" )\n",
" self.attention = InterpretableMultiHeadAttention(\n",
" n_head=n_head,\n",
@@ -590,7 +623,7 @@
" self.attention_ln = LayerNorm(normalized_shape=hidden_size, eps=1e-3)\n",
"\n",
" self.positionwise_grn = GRN(\n",
- " input_size=hidden_size, hidden_size=hidden_size, dropout=dropout\n",
+ " input_size=hidden_size, hidden_size=hidden_size, dropout=dropout, activation=grn_activation\n",
" )\n",
"\n",
" # ---------------------- Decoder -----------------------#\n",
@@ -652,8 +685,7 @@
" `dropout`: float (0, 1), dropout of inputs VSNs.
\n",
" `n_head`: int=4, number of attention heads in temporal fusion decoder.
\n",
" `attn_dropout`: float (0, 1), dropout of fusion decoder's attention layer.
\n",
- " `shared_weights`: bool, If True, all blocks within each stack will share parameters.
\n",
- " `activation`: str, activation from ['ReLU', 'Softplus', 'Tanh', 'SELU', 'LeakyReLU', 'PReLU', 'Sigmoid'].
\n",
+ " `grn_activation`: str, activation for the GRN module from ['ReLU', 'Softplus', 'Tanh', 'SELU', 'LeakyReLU', 'Sigmoid', 'ELU', 'GLU'].
\n",
" `loss`: PyTorch module, instantiated train loss class from [losses collection](https://nixtla.github.io/neuralforecast/losses.pytorch.html).
\n",
" `valid_loss`: PyTorch module=`loss`, instantiated valid loss class from [losses collection](https://nixtla.github.io/neuralforecast/losses.pytorch.html).
\n",
" `max_steps`: int=1000, maximum number of training steps.
\n",
@@ -700,6 +732,7 @@
" hidden_size: int = 128,\n",
" n_head: int = 4,\n",
" attn_dropout: float = 0.0,\n",
+ " grn_activation: str = 'ELU',\n",
" dropout: float = 0.1,\n",
" loss=MAE(),\n",
" valid_loss=None,\n",
@@ -758,6 +791,7 @@
" self.example_length = input_size + h\n",
" self.interpretability_params = dict([]) # type: ignore\n",
" self.tgt_size = tgt_size\n",
+ " self.grn_activation = grn_activation\n",
" futr_exog_size = max(self.futr_exog_size, 1)\n",
" num_historic_vars = futr_exog_size + self.hist_exog_size + tgt_size\n",
"\n",
@@ -772,13 +806,15 @@
" self.static_encoder = StaticCovariateEncoder(\n",
" hidden_size=hidden_size,\n",
" num_static_vars=self.stat_exog_size,\n",
- " dropout=dropout)\n",
+ " dropout=dropout,\n",
+ " grn_activation=self.grn_activation)\n",
"\n",
" self.temporal_encoder = TemporalCovariateEncoder(\n",
" hidden_size=hidden_size,\n",
" num_historic_vars=num_historic_vars,\n",
" num_future_vars=futr_exog_size,\n",
" dropout=dropout,\n",
+ " grn_activation=self.grn_activation\n",
" )\n",
"\n",
" # ------------------------------ Decoders -----------------------------#\n",
@@ -789,6 +825,7 @@
" encoder_length=self.input_size,\n",
" attn_dropout=attn_dropout,\n",
" dropout=dropout,\n",
+ " grn_activation=self.grn_activation\n",
" )\n",
"\n",
" # Adapter with Loss dependent dimensions\n",
@@ -992,7 +1029,7 @@
"> TFT.fit (dataset, val_size=0, test_size=0, random_seed=None,\n",
"> distributed_config=None)\n",
"\n",
- "*Fit.\n",
+ "Fit.\n",
"\n",
"The `fit` method, optimizes the neural network's weights using the\n",
"initialization parameters (`learning_rate`, `windows_batch_size`, ...)\n",
@@ -1011,7 +1048,7 @@
"`dataset`: NeuralForecast's `TimeSeriesDataset`, see [documentation](https://nixtla.github.io/neuralforecast/tsdataset.html).
\n",
"`val_size`: int, validation size for temporal cross-validation.
\n",
"`random_seed`: int=None, random_seed for pytorch initializer and numpy generators, overwrites model.__init__'s.
\n",
- "`test_size`: int, test size for temporal cross-validation.
*"
+ "`test_size`: int, test size for temporal cross-validation.
"
],
"text/plain": [
"---\n",
@@ -1021,7 +1058,7 @@
"> TFT.fit (dataset, val_size=0, test_size=0, random_seed=None,\n",
"> distributed_config=None)\n",
"\n",
- "*Fit.\n",
+ "Fit.\n",
"\n",
"The `fit` method, optimizes the neural network's weights using the\n",
"initialization parameters (`learning_rate`, `windows_batch_size`, ...)\n",
@@ -1040,7 +1077,7 @@
"`dataset`: NeuralForecast's `TimeSeriesDataset`, see [documentation](https://nixtla.github.io/neuralforecast/tsdataset.html).
\n",
"`val_size`: int, validation size for temporal cross-validation.
\n",
"`random_seed`: int=None, random_seed for pytorch initializer and numpy generators, overwrites model.__init__'s.
\n",
- "`test_size`: int, test size for temporal cross-validation.
*"
+ "`test_size`: int, test size for temporal cross-validation.
"
]
},
"execution_count": null,
@@ -1067,7 +1104,7 @@
"> TFT.predict (dataset, test_size=None, step_size=1, random_seed=None,\n",
"> **data_module_kwargs)\n",
"\n",
- "*Predict.\n",
+ "Predict.\n",
"\n",
"Neural network prediction with PL's `Trainer` execution of `predict_step`.\n",
"\n",
@@ -1076,7 +1113,7 @@
"`test_size`: int=None, test size for temporal cross-validation.
\n",
"`step_size`: int=1, Step size between each window.
\n",
"`random_seed`: int=None, random_seed for pytorch initializer and numpy generators, overwrites model.__init__'s.
\n",
- "`**data_module_kwargs`: PL's TimeSeriesDataModule args, see [documentation](https://pytorch-lightning.readthedocs.io/en/1.6.1/extensions/datamodules.html#using-a-datamodule).*"
+ "`**data_module_kwargs`: PL's TimeSeriesDataModule args, see [documentation](https://pytorch-lightning.readthedocs.io/en/1.6.1/extensions/datamodules.html#using-a-datamodule)."
],
"text/plain": [
"---\n",
@@ -1086,7 +1123,7 @@
"> TFT.predict (dataset, test_size=None, step_size=1, random_seed=None,\n",
"> **data_module_kwargs)\n",
"\n",
- "*Predict.\n",
+ "Predict.\n",
"\n",
"Neural network prediction with PL's `Trainer` execution of `predict_step`.\n",
"\n",
@@ -1095,7 +1132,7 @@
"`test_size`: int=None, test size for temporal cross-validation.
\n",
"`step_size`: int=1, Step size between each window.
\n",
"`random_seed`: int=None, random_seed for pytorch initializer and numpy generators, overwrites model.__init__'s.
\n",
- "`**data_module_kwargs`: PL's TimeSeriesDataModule args, see [documentation](https://pytorch-lightning.readthedocs.io/en/1.6.1/extensions/datamodules.html#using-a-datamodule).*"
+ "`**data_module_kwargs`: PL's TimeSeriesDataModule args, see [documentation](https://pytorch-lightning.readthedocs.io/en/1.6.1/extensions/datamodules.html#using-a-datamodule)."
]
},
"execution_count": null,
@@ -1117,34 +1154,34 @@
"text/markdown": [
"---\n",
"\n",
- "[source](https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/tft.py#L630){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+ "[source](https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/tft.py#L678){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
"\n",
"### TFT.feature_importances,\n",
"\n",
"> TFT.feature_importances, ()\n",
"\n",
- "*Compute the feature importances for historical, future, and static features.\n",
+ "Compute the feature importances for historical, future, and static features.\n",
"\n",
"Returns:\n",
" dict: A dictionary containing the feature importances for each feature type.\n",
" The keys are 'hist_vsn', 'future_vsn', and 'static_vsn', and the values\n",
- " are pandas DataFrames with the corresponding feature importances.*"
+ " are pandas DataFrames with the corresponding feature importances."
],
"text/plain": [
"---\n",
"\n",
- "[source](https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/tft.py#L630){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+ "[source](https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/tft.py#L678){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
"\n",
"### TFT.feature_importances,\n",
"\n",
"> TFT.feature_importances, ()\n",
"\n",
- "*Compute the feature importances for historical, future, and static features.\n",
+ "Compute the feature importances for historical, future, and static features.\n",
"\n",
"Returns:\n",
" dict: A dictionary containing the feature importances for each feature type.\n",
" The keys are 'hist_vsn', 'future_vsn', and 'static_vsn', and the values\n",
- " are pandas DataFrames with the corresponding feature importances.*"
+ " are pandas DataFrames with the corresponding feature importances."
]
},
"execution_count": null,
@@ -1166,30 +1203,30 @@
"text/markdown": [
"---\n",
"\n",
- "[source](https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/tft.py#L688){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+ "[source](https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/tft.py#L736){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
"\n",
"### TFT.attention_weights\n",
"\n",
"> TFT.attention_weights ()\n",
"\n",
- "*Batch average attention weights\n",
+ "Batch average attention weights\n",
"\n",
"Returns:\n",
- "np.ndarray: A 1D array containing the attention weights for each time step.*"
+ "np.ndarray: A 1D array containing the attention weights for each time step."
],
"text/plain": [
"---\n",
"\n",
- "[source](https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/tft.py#L688){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+ "[source](https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/tft.py#L736){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
"\n",
"### TFT.attention_weights\n",
"\n",
"> TFT.attention_weights ()\n",
"\n",
- "*Batch average attention weights\n",
+ "Batch average attention weights\n",
"\n",
"Returns:\n",
- "np.ndarray: A 1D array containing the attention weights for each time step.*"
+ "np.ndarray: A 1D array containing the attention weights for each time step."
]
},
"execution_count": null,
@@ -1211,30 +1248,30 @@
"text/markdown": [
"---\n",
"\n",
- "[source](https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/tft.py#L688){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+ "[source](https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/tft.py#L736){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
"\n",
"### TFT.attention_weights\n",
"\n",
"> TFT.attention_weights ()\n",
"\n",
- "*Batch average attention weights\n",
+ "Batch average attention weights\n",
"\n",
"Returns:\n",
- "np.ndarray: A 1D array containing the attention weights for each time step.*"
+ "np.ndarray: A 1D array containing the attention weights for each time step."
],
"text/plain": [
"---\n",
"\n",
- "[source](https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/tft.py#L688){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+ "[source](https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/tft.py#L736){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
"\n",
"### TFT.attention_weights\n",
"\n",
"> TFT.attention_weights ()\n",
"\n",
- "*Batch average attention weights\n",
+ "Batch average attention weights\n",
"\n",
"Returns:\n",
- "np.ndarray: A 1D array containing the attention weights for each time step.*"
+ "np.ndarray: A 1D array containing the attention weights for each time step."
]
},
"execution_count": null,
@@ -1256,30 +1293,30 @@
"text/markdown": [
"---\n",
"\n",
- "[source](https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/tft.py#L706){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+ "[source](https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/tft.py#L754){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
"\n",
"### TFT.feature_importance_correlations\n",
"\n",
"> TFT.feature_importance_correlations ()\n",
"\n",
- "*Compute the correlation between the past and future feature importances and the mean attention weights.\n",
+ "Compute the correlation between the past and future feature importances and the mean attention weights.\n",
"\n",
"Returns:\n",
- "pd.DataFrame: A DataFrame containing the correlation coefficients between the past feature importances and the mean attention weights.*"
+ "pd.DataFrame: A DataFrame containing the correlation coefficients between the past feature importances and the mean attention weights."
],
"text/plain": [
"---\n",
"\n",
- "[source](https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/tft.py#L706){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
+ "[source](https://github.com/Nixtla/neuralforecast/blob/main/neuralforecast/models/tft.py#L754){target=\"_blank\" style=\"float:right; font-size:smaller\"}\n",
"\n",
"### TFT.feature_importance_correlations\n",
"\n",
"> TFT.feature_importance_correlations ()\n",
"\n",
- "*Compute the correlation between the past and future feature importances and the mean attention weights.\n",
+ "Compute the correlation between the past and future feature importances and the mean attention weights.\n",
"\n",
"Returns:\n",
- "pd.DataFrame: A DataFrame containing the correlation coefficients between the past feature importances and the mean attention weights.*"
+ "pd.DataFrame: A DataFrame containing the correlation coefficients between the past feature importances and the mean attention weights."
]
},
"execution_count": null,
@@ -1304,10 +1341,21 @@
"execution_count": null,
"metadata": {},
"outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "Seed set to 1\n",
+ "GPU available: True (mps), used: True\n",
+ "TPU available: False, using: 0 TPU cores\n",
+ "IPU available: False, using: 0 IPUs\n",
+ "HPU available: False, using: 0 HPUs\n"
+ ]
+ },
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "28d80883ad5c4fbcb09c287416508cff",
+ "model_id": "fb09f042057e45d181a21ef46b8d933d",
"version_major": 2,
"version_minor": 0
},
@@ -1321,7 +1369,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "164f652306bc4443b667b223fd216571",
+ "model_id": "e7853b68bc174f7bb1a022a466da2186",
"version_major": 2,
"version_minor": 0
},
@@ -1335,7 +1383,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "9f5d2857dc80499ab589276dc2ef2ef7",
+ "model_id": "86866ee63d314c69b955af148d6d28c9",
"version_major": 2,
"version_minor": 0
},
@@ -1349,7 +1397,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "a61af16db6fa430e8bad2e56cbb6e204",
+ "model_id": "f1e52f5aba3e40d9beca566cb62ef5b4",
"version_major": 2,
"version_minor": 0
},
@@ -1363,7 +1411,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "3d6a111c2c4a4310a668494267427ee3",
+ "model_id": "4a058b96deae42dabda97307249d15a6",
"version_major": 2,
"version_minor": 0
},
@@ -1377,7 +1425,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "053994a65f284431aabfd6ba7aa46e08",
+ "model_id": "779e686e6a86423ebda2172ea6268786",
"version_major": 2,
"version_minor": 0
},
@@ -1391,7 +1439,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "21573be21eea4e858a75e23f7f090e9c",
+ "model_id": "3866504eb4ac47a398f82b67b4528e68",
"version_major": 2,
"version_minor": 0
},
@@ -1405,7 +1453,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "65b34751133141388175b695fe42d76b",
+ "model_id": "652afb9d18c9401e800372fe47e3bd0c",
"version_major": 2,
"version_minor": 0
},
@@ -1419,7 +1467,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "87a7b895c8b14b41a0365e1e0eebab3f",
+ "model_id": "aaca1e1ab85e4988acfc52b976e06b62",
"version_major": 2,
"version_minor": 0
},
@@ -1433,7 +1481,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "45d33286f3ec4611bf269d991c984283",
+ "model_id": "17af4b56375e4eda9b42038b0ddca23b",
"version_major": 2,
"version_minor": 0
},
@@ -1447,7 +1495,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "a3df1d8c667b467a9c95a674e8e70790",
+ "model_id": "bc0cf020042b480e865786435fe86174",
"version_major": 2,
"version_minor": 0
},
@@ -1461,7 +1509,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "1676e1cec80a43feb6fff5b101aa9e74",
+ "model_id": "406fc612f09c45489dcfd57159646a9a",
"version_major": 2,
"version_minor": 0
},
@@ -1475,7 +1523,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "ce5140456f724787b8c521c71877161f",
+ "model_id": "9cfb437ad5fe4b60bc407ddfd3a54eef",
"version_major": 2,
"version_minor": 0
},
@@ -1489,7 +1537,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "bcf26a086e48436dbec942a16d1068e5",
+ "model_id": "4caea18544d44beca9f03f4ba5432051",
"version_major": 2,
"version_minor": 0
},
@@ -1503,7 +1551,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "82080ec388b64b36a9329573ba7f6c72",
+ "model_id": "c02d707ed0974a1ba562e6b261370815",
"version_major": 2,
"version_minor": 0
},
@@ -1517,7 +1565,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "9cb23c364290468982ab7db4066d41c4",
+ "model_id": "356f7ff7f9654864aa8a95d1516e0e03",
"version_major": 2,
"version_minor": 0
},
@@ -1531,7 +1579,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "b17d7126e58f4765aa5d5bf685fb1c76",
+ "model_id": "ac4a9d8fa3e1494ab986fe59b47c3bd2",
"version_major": 2,
"version_minor": 0
},
@@ -1545,7 +1593,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "b56b7debd7374198bf86a3c4ded57a40",
+ "model_id": "90906beaebd84ac6bd4b996816d0efe8",
"version_major": 2,
"version_minor": 0
},
@@ -1559,7 +1607,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "e25fceeff8aa41c0852b5f40c9a73489",
+ "model_id": "89d445aa28fd43568b55c5674577d986",
"version_major": 2,
"version_minor": 0
},
@@ -1573,7 +1621,7 @@
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "39a9ebe57449439992be2036b0311762",
+ "model_id": "1e0aa3e904954d89b1f7446da474c971",
"version_major": 2,
"version_minor": 0
},
@@ -1585,23 +1633,20 @@
"output_type": "display_data"
},
{
- "data": {
- "application/vnd.jupyter.widget-view+json": {
- "model_id": "9bdae404519d47769ed1940b9c7e26a0",
- "version_major": 2,
- "version_minor": 0
- },
- "text/plain": [
- "Validation: | | 0/? [00:00, ?it/s]"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "Trainer already configured with model summary callbacks: []. Skipping setting a default `ModelSummary` callback.\n",
+ "GPU available: True (mps), used: True\n",
+ "TPU available: False, using: 0 TPU cores\n",
+ "IPU available: False, using: 0 IPUs\n",
+ "HPU available: False, using: 0 HPUs\n"
+ ]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
- "model_id": "970de95dd7a84c40bf408a433189067a",
+ "model_id": "23e3f9ec08ab4b4697bf2479a6bd0bc4",
"version_major": 2,
"version_minor": 0
},
@@ -1624,7 +1669,7 @@
},
{
"data": {
- "image/png": "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",
+ "image/png": "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",
"text/plain": [
"