End of training

README.md

@@ -3,6 +3,8 @@ library_name: transformers
 license: apache-2.0
 base_model: facebook/wav2vec2-xls-r-1b
 tags:
+- automatic-speech-recognition
+- bigcgen
 - generated_from_trainer
 metrics:
 - wer
@@ -16,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # xls-r-1b-bigcgen-combined-10hrs-model
 
-This model is a fine-tuned version of [facebook/wav2vec2-xls-r-1b](https://huggingface.co/facebook/wav2vec2-xls-r-1b) on an unknown dataset.
+This model is a fine-tuned version of [facebook/wav2vec2-xls-r-1b](https://huggingface.co/facebook/wav2vec2-xls-r-1b) on the BIGCGEN - NA dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.5381
 - Wer: 0.5532

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 4.016913319238901,
+    "eval_loss": 0.5381172299385071,
+    "eval_runtime": 52.0034,
+    "eval_samples": 682,
+    "eval_samples_per_second": 13.115,
+    "eval_steps_per_second": 3.288,
+    "eval_wer": 0.5531826949572886,
+    "total_flos": 1.1255723951737272e+19,
+    "train_loss": 2.379649096037212,
+    "train_runtime": 3597.9031,
+    "train_samples": 3783,
+    "train_samples_per_second": 31.543,
+    "train_steps_per_second": 3.944
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 4.016913319238901,
+    "eval_loss": 0.5381172299385071,
+    "eval_runtime": 52.0034,
+    "eval_samples": 682,
+    "eval_samples_per_second": 13.115,
+    "eval_steps_per_second": 3.288,
+    "eval_wer": 0.5531826949572886
+}

runs/Dec26_23-15-15_srvrocgpu011.uct.ac.za/events.out.tfevents.1735251719.srvrocgpu011.uct.ac.za

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:447fb770eaba7dae7914b17a96d245940180ebfe8e0f30d1656be9c3201fbfa1
+size 40

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 4.016913319238901,
+    "total_flos": 1.1255723951737272e+19,
+    "train_loss": 2.379649096037212,
+    "train_runtime": 3597.9031,
+    "train_samples": 3783,
+    "train_samples_per_second": 31.543,
+    "train_steps_per_second": 3.944
+}

trainer_state.json

@@ -0,0 +1,243 @@
+{
+  "best_metric": 0.5338566303253174,
+  "best_model_checkpoint": "/scratch/skscla001/speech/results/xls-r-1b-bigcgen-combined-10hrs-model/checkpoint-1500",
+  "epoch": 4.016913319238901,
+  "eval_steps": 100,
+  "global_step": 1900,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.21141649048625794,
+      "eval_loss": 3.6878206729888916,
+      "eval_runtime": 51.4868,
+      "eval_samples_per_second": 13.246,
+      "eval_steps_per_second": 3.321,
+      "eval_wer": 1.0,
+      "step": 100
+    },
+    {
+      "epoch": 0.42283298097251587,
+      "eval_loss": 2.6382739543914795,
+      "eval_runtime": 51.8513,
+      "eval_samples_per_second": 13.153,
+      "eval_steps_per_second": 3.298,
+      "eval_wer": 1.0,
+      "step": 200
+    },
+    {
+      "epoch": 0.6342494714587738,
+      "eval_loss": 0.909231960773468,
+      "eval_runtime": 51.7378,
+      "eval_samples_per_second": 13.182,
+      "eval_steps_per_second": 3.305,
+      "eval_wer": 0.9116836594103058,
+      "step": 300
+    },
+    {
+      "epoch": 0.8456659619450317,
+      "eval_loss": 0.7954993844032288,
+      "eval_runtime": 51.8008,
+      "eval_samples_per_second": 13.166,
+      "eval_steps_per_second": 3.301,
+      "eval_wer": 0.7865803251584459,
+      "step": 400
+    },
+    {
+      "epoch": 1.0570824524312896,
+      "grad_norm": 4.6689229011535645,
+      "learning_rate": 3e-05,
+      "loss": 5.7153,
+      "step": 500
+    },
+    {
+      "epoch": 1.0570824524312896,
+      "eval_loss": 0.7447614669799805,
+      "eval_runtime": 52.2471,
+      "eval_samples_per_second": 13.053,
+      "eval_steps_per_second": 3.273,
+      "eval_wer": 0.8102783135850097,
+      "step": 500
+    },
+    {
+      "epoch": 1.2684989429175475,
+      "eval_loss": 0.7003403902053833,
+      "eval_runtime": 51.6189,
+      "eval_samples_per_second": 13.212,
+      "eval_steps_per_second": 3.313,
+      "eval_wer": 0.7529622485533205,
+      "step": 600
+    },
+    {
+      "epoch": 1.4799154334038054,
+      "eval_loss": 0.6151377558708191,
+      "eval_runtime": 51.9411,
+      "eval_samples_per_second": 13.13,
+      "eval_steps_per_second": 3.292,
+      "eval_wer": 0.6453568476164233,
+      "step": 700
+    },
+    {
+      "epoch": 1.6913319238900635,
+      "eval_loss": 0.6107520461082458,
+      "eval_runtime": 51.395,
+      "eval_samples_per_second": 13.27,
+      "eval_steps_per_second": 3.327,
+      "eval_wer": 0.6132543400385781,
+      "step": 800
+    },
+    {
+      "epoch": 1.9027484143763214,
+      "eval_loss": 0.5824649333953857,
+      "eval_runtime": 51.9182,
+      "eval_samples_per_second": 13.136,
+      "eval_steps_per_second": 3.294,
+      "eval_wer": 0.6153210250757785,
+      "step": 900
+    },
+    {
+      "epoch": 2.1141649048625792,
+      "grad_norm": 15.280961990356445,
+      "learning_rate": 2.8904309715120527e-05,
+      "loss": 1.3819,
+      "step": 1000
+    },
+    {
+      "epoch": 2.1141649048625792,
+      "eval_loss": 0.5743407011032104,
+      "eval_runtime": 51.5224,
+      "eval_samples_per_second": 13.237,
+      "eval_steps_per_second": 3.319,
+      "eval_wer": 0.5954808487186553,
+      "step": 1000
+    },
+    {
+      "epoch": 2.3255813953488373,
+      "eval_loss": 0.5597561001777649,
+      "eval_runtime": 51.6771,
+      "eval_samples_per_second": 13.197,
+      "eval_steps_per_second": 3.309,
+      "eval_wer": 0.5770184623863324,
+      "step": 1100
+    },
+    {
+      "epoch": 2.536997885835095,
+      "eval_loss": 0.5867146849632263,
+      "eval_runtime": 51.6196,
+      "eval_samples_per_second": 13.212,
+      "eval_steps_per_second": 3.313,
+      "eval_wer": 0.6104987599889777,
+      "step": 1200
+    },
+    {
+      "epoch": 2.748414376321353,
+      "eval_loss": 0.5733869671821594,
+      "eval_runtime": 52.0076,
+      "eval_samples_per_second": 13.113,
+      "eval_steps_per_second": 3.288,
+      "eval_wer": 0.5588316340589694,
+      "step": 1300
+    },
+    {
+      "epoch": 2.9598308668076108,
+      "eval_loss": 0.5357180833816528,
+      "eval_runtime": 51.4942,
+      "eval_samples_per_second": 13.244,
+      "eval_steps_per_second": 3.321,
+      "eval_wer": 0.5405070267291264,
+      "step": 1400
+    },
+    {
+      "epoch": 3.171247357293869,
+      "grad_norm": 3.271611213684082,
+      "learning_rate": 2.7808619430241053e-05,
+      "loss": 1.1226,
+      "step": 1500
+    },
+    {
+      "epoch": 3.171247357293869,
+      "eval_loss": 0.5338566303253174,
+      "eval_runtime": 51.5906,
+      "eval_samples_per_second": 13.219,
+      "eval_steps_per_second": 3.315,
+      "eval_wer": 0.5414714797464867,
+      "step": 1500
+    },
+    {
+      "epoch": 3.382663847780127,
+      "eval_loss": 0.558174192905426,
+      "eval_runtime": 51.6074,
+      "eval_samples_per_second": 13.215,
+      "eval_steps_per_second": 3.313,
+      "eval_wer": 0.5585560760540094,
+      "step": 1600
+    },
+    {
+      "epoch": 3.5940803382663846,
+      "eval_loss": 0.5485186576843262,
+      "eval_runtime": 52.0985,
+      "eval_samples_per_second": 13.091,
+      "eval_steps_per_second": 3.282,
+      "eval_wer": 0.5571782860292092,
+      "step": 1700
+    },
+    {
+      "epoch": 3.8054968287526427,
+      "eval_loss": 0.5342406034469604,
+      "eval_runtime": 51.6512,
+      "eval_samples_per_second": 13.204,
+      "eval_steps_per_second": 3.311,
+      "eval_wer": 0.5403692477266464,
+      "step": 1800
+    },
+    {
+      "epoch": 4.016913319238901,
+      "eval_loss": 0.5380779504776001,
+      "eval_runtime": 51.6402,
+      "eval_samples_per_second": 13.207,
+      "eval_steps_per_second": 3.311,
+      "eval_wer": 0.5531826949572886,
+      "step": 1900
+    },
+    {
+      "epoch": 4.016913319238901,
+      "step": 1900,
+      "total_flos": 1.1255723951737272e+19,
+      "train_loss": 2.379649096037212,
+      "train_runtime": 3597.9031,
+      "train_samples_per_second": 31.543,
+      "train_steps_per_second": 3.944
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 14190,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 30,
+  "save_steps": 400,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 4,
+        "early_stopping_threshold": 0.0
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 1
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.1255723951737272e+19,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}