Update config of IAM lines

1 files changed, 36 insertions, 13 deletions

diff --git a/training/conf/experiment/cnn_htr_char_lines.yaml b/training/conf/experiment/cnn_htr_char_lines.yaml
index 08d9282..0f28ff9 100644
--- a/training/conf/experiment/cnn_htr_char_lines.yaml
+++ b/training/conf/experiment/cnn_htr_char_lines.yaml
@@ -11,13 +11,31 @@ defaults:
 
 
 criterion:
-  _target_: torch.nn.CrossEntropyLoss
-  ignore_index: 3
+  _target_: text_recognizer.criterions.label_smoothing.LabelSmoothingLoss
+  smoothing: 0.1 
+  ignore_index: 1000
     
 mapping:
-  _target_: text_recognizer.data.emnist_mapping.EmnistMapping
+  _target_: text_recognizer.data.word_piece_mapping.WordPieceMapping
+  num_features: 1000
+  tokens: iamdb_1kwp_tokens_1000.txt
+  lexicon: iamdb_1kwp_lex_1000.txt
+  data_dir: null
+  use_words: false
+  prepend_wordsep: false
+  special_tokens: [ <s>, <e>, <p> ]
+  # _target_: text_recognizer.data.emnist_mapping.EmnistMapping
   # extra_symbols: [ "\n" ]
 
+callbacks:
+  stochastic_weight_averaging:
+    _target_: pytorch_lightning.callbacks.StochasticWeightAveraging
+    swa_epoch_start: 0.8
+    swa_lrs: 0.05
+    annealing_epochs: 10
+    annealing_strategy: cos
+    device: null
+
 optimizers:
   madgrad:
     _target_: madgrad.MADGRAD
@@ -30,16 +48,21 @@ optimizers:
 
 lr_schedulers:
   network:
-    _target_: torch.optim.lr_scheduler.CosineAnnealingLR
-    T_max: 1024
-    eta_min: 4.5e-6
-    last_epoch: -1
+    _target_: torch.optim.lr_scheduler.ReduceLROnPlateau
+    mode: min
+    factor: 0.1
+    patience: 10
+    threshold: 1.0e-4
+    threshold_mode: rel
+    cooldown: 0
+    min_lr: 1.0e-7
+    eps: 1.0e-8
     interval: epoch
     monitor: val/loss
 
 datamodule:
   _target_: text_recognizer.data.iam_lines.IAMLines
-  batch_size: 24
+  batch_size: 16
   num_workers: 12
   train_fraction: 0.8
   augment: true
@@ -51,8 +74,8 @@ network:
   hidden_dim: 128
   encoder_dim: 1280
   dropout_rate: 0.2
-  num_classes: 58
-  pad_index: 3
+  num_classes: 1006
+  pad_index: 1000
   encoder:
     _target_: text_recognizer.networks.encoders.efficientnet.EfficientNet
     arch: b0
@@ -89,7 +112,7 @@ model:
 
 trainer:
   _target_: pytorch_lightning.Trainer
-  stochastic_weight_avg: false
+  stochastic_weight_avg: true
   auto_scale_batch_size: binsearch
   auto_lr_find: false
   gradient_clip_val: 0
@@ -98,7 +121,7 @@ trainer:
   precision: 16
   max_epochs: 1024
   terminate_on_nan: true
-  weights_summary: top
+  weights_summary: null
   limit_train_batches: 1.0 
   limit_val_batches: 1.0
   limit_test_batches: 1.0
@@ -106,4 +129,4 @@ trainer:
   accumulate_grad_batches: 4
   overfit_batches: 0.0
 
-# summary: [[1, 1, 56, 1024], [1, 89]]
+summary: [[1, 1, 56, 1024], [1, 89]]