Training in progress, epoch 1

hyperparameters.yaml

@@ -87,14 +87,14 @@ include_tokens_per_second: false
 jit_mode_eval: false
 label_names: null
 label_smoothing_factor: 0.0
-learning_rate: 1.5730263550306148e-05
+learning_rate: 1.2045081648781836e-05
 length_column_name: length
 load_best_model_at_end: true
 local_rank: 0
 log_level: passive
 log_level_replica: warning
 log_on_each_node: true
-logging_dir: model_runs/donut_experiment_bayesian_trial_0/runs/May20_17-23-56_ip-172-16-167-107.ec2.internal
+logging_dir: model_runs/donut_experiment_bayesian_trial_0/runs/May30_15-29-21_ip-172-16-168-165.ec2.internal
 logging_first_step: false
 logging_nan_inf_filter: true
 logging_steps: 100
@@ -109,7 +109,7 @@ metric_for_best_model: loss
 mp_parameters: ''
 neftune_noise_alpha: null
 no_cuda: false
-num_train_epochs: 2
+num_train_epochs: 5
 optim: !!python/object/apply:transformers.training_args.OptimizerNames
 - adamw_torch
 optim_args: null
@@ -157,4 +157,4 @@ use_legacy_prediction_loop: false
 use_mps_device: false
 warmup_ratio: 0.0
 warmup_steps: 0
-weight_decay: 0.005642143457488303
+weight_decay: 0.008800425083056953

hyperparameters_tuned.json

@@ -1 +1 @@
-{"learning_rate": 1.5730263550306148e-05, "weight_decay": 0.005642143457488303, "num_train_epochs": 2}
+{"learning_rate": 1.2045081648781836e-05, "weight_decay": 0.008800425083056953, "num_train_epochs": 5}

metrics.jsonl

@@ -1,2 +1 @@
-{"eval_loss": 0.50071120262146, "eval_bleu": 0.06974875456826889, "eval_precisions": [0.8189300411522634, 0.7505827505827506, 0.7043010752688172, 0.6603174603174603], "eval_brevity_penalty": 0.095388787518951, "eval_length_ratio": 0.2985257985257985, "eval_translation_length": 486, "eval_reference_length": 1628, "eval_cer": 0.7575012392795679, "eval_wer": 0.8257353645358336, "eval_runtime": 71.6191, "eval_samples_per_second": 0.796, "eval_steps_per_second": 0.796, "epoch": 1.0}
-{"eval_loss": 0.47436320781707764, "eval_bleu": 0.0691840661962562, "eval_precisions": [0.8295218295218295, 0.7735849056603774, 0.7247956403269755, 0.6838709677419355], "eval_brevity_penalty": 0.09212440522976928, "eval_length_ratio": 0.29545454545454547, "eval_translation_length": 481, "eval_reference_length": 1628, "eval_cer": 0.7526673269798355, "eval_wer": 0.82323955242213, "eval_runtime": 71.6494, "eval_samples_per_second": 0.796, "eval_steps_per_second": 0.796, "epoch": 2.0}
+{"eval_loss": 0.6227948665618896, "eval_bleu": 0.04856205045021571, "eval_precisions": [0.7096774193548387, 0.6053921568627451, 0.5612535612535613, 0.5102040816326531], "eval_brevity_penalty": 0.08199678262097645, "eval_length_ratio": 0.2856265356265356, "eval_translation_length": 465, "eval_reference_length": 1628, "eval_cer": 0.775094682355145, "eval_wer": 0.8591598980283793, "eval_runtime": 187.5635, "eval_samples_per_second": 0.304, "eval_steps_per_second": 0.304, "epoch": 1.0}

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:75159014b3dc71b6da03ab3039c1a4e68683134d808e570d6c82f8265b097b5c
+oid sha256:727f7c42a1e454798ad4c84ec87ef1dd197a4148290db837d0d86c003f128542
 size 809103512

runs/May30_15-29-21_ip-172-16-168-165.ec2.internal/events.out.tfevents.1717082961.ip-172-16-168-165.ec2.internal.8916.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a00329ca35c2cb96744bbd74bcd59f96bb5d3047dd56e7f4f8fc25e872c5b5d9
+size 10299

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9cb66e8a1cdd6cff0e2b6b44220432c8e6152675d118cded394f4834682229b7
+oid sha256:78cdc1eb0aab87935dcb50385d3f8cd3ccb575e79c07b0a2828bd81d4871de6e
 size 5240