End of training

README.md

@@ -2,6 +2,8 @@
 license: apache-2.0
 base_model: facebook/wav2vec2-large-xlsr-53
 tags:
+- automatic-speech-recognition
+- DewiBrynJones/banc-trawsgrifiadau-bangor-clean-with-ccv
 - generated_from_trainer
 metrics:
 - wer
@@ -15,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # wav2vec2-xlsr-53-ft-btb-ccv-cy
 
-This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on an unknown dataset.
+This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the DEWIBRYNJONES/BANC-TRAWSGRIFIADAU-BANGOR-CLEAN-WITH-CCV - DEFAULT dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.5418
 - Wer: 0.4178

all_results.json

@@ -1,15 +1,15 @@
 {
     "epoch": 1.0080645161290323,
-    "eval_loss": 0.4942198395729065,
-    "eval_runtime": 177.2206,
+    "eval_loss": 0.5417820811271667,
+    "eval_runtime": 188.0972,
     "eval_samples": 7022,
-    "eval_samples_per_second": 39.623,
-    "eval_steps_per_second": 2.477,
-    "eval_wer": 0.3917272352808302,
+    "eval_samples_per_second": 37.332,
+    "eval_steps_per_second": 2.334,
+    "eval_wer": 0.4178370693206128,
     "total_flos": 6.709869296482936e+18,
-    "train_loss": 1.1787635701497396,
-    "train_runtime": 4443.9077,
+    "train_loss": 1.3100875549316406,
+    "train_runtime": 4950.9591,
     "train_samples": 47607,
-    "train_samples_per_second": 10.801,
-    "train_steps_per_second": 0.675
+    "train_samples_per_second": 9.695,
+    "train_steps_per_second": 0.606
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 1.0080645161290323,
-    "eval_loss": 0.4942198395729065,
-    "eval_runtime": 177.2206,
+    "eval_loss": 0.5417820811271667,
+    "eval_runtime": 188.0972,
     "eval_samples": 7022,
-    "eval_samples_per_second": 39.623,
-    "eval_steps_per_second": 2.477,
-    "eval_wer": 0.3917272352808302
+    "eval_samples_per_second": 37.332,
+    "eval_steps_per_second": 2.334,
+    "eval_wer": 0.4178370693206128
 }

runs/Aug29_20-04-14_6d77da852b30/events.out.tfevents.1724964390.6d77da852b30.1700.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c9a9b676fc791d60bb726636f5c4bf380faff55248219f6ccafe191364c4580
+size 406

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 1.0080645161290323,
     "total_flos": 6.709869296482936e+18,
-    "train_loss": 1.1787635701497396,
-    "train_runtime": 4443.9077,
+    "train_loss": 1.3100875549316406,
+    "train_runtime": 4950.9591,
     "train_samples": 47607,
-    "train_samples_per_second": 10.801,
-    "train_steps_per_second": 0.675
+    "train_samples_per_second": 9.695,
+    "train_steps_per_second": 0.606
 }

trainer_state.json

@@ -10,189 +10,189 @@
   "log_history": [
     {
       "epoch": 0.06720430107526881,
-      "eval_loss": 2.9829530715942383,
-      "eval_runtime": 176.5397,
-      "eval_samples_per_second": 39.776,
-      "eval_steps_per_second": 2.487,
+      "eval_loss": 3.1445324420928955,
+      "eval_runtime": 187.2734,
+      "eval_samples_per_second": 37.496,
+      "eval_steps_per_second": 2.344,
       "eval_wer": 1.0,
       "step": 200
     },
     {
       "epoch": 0.13440860215053763,
-      "eval_loss": 1.5613375902175903,
-      "eval_runtime": 174.5758,
-      "eval_samples_per_second": 40.223,
-      "eval_steps_per_second": 2.515,
-      "eval_wer": 0.9657582481829415,
+      "eval_loss": 2.7407212257385254,
+      "eval_runtime": 185.5858,
+      "eval_samples_per_second": 37.837,
+      "eval_steps_per_second": 2.365,
+      "eval_wer": 0.9999867125526515,
       "step": 400
     },
     {
       "epoch": 0.16801075268817203,
-      "grad_norm": 2.138428211212158,
-      "learning_rate": 0.00027833333333333334,
-      "loss": 3.6118,
+      "grad_norm": 8.384466171264648,
+      "learning_rate": 0.00027866666666666665,
+      "loss": 4.0188,
       "step": 500
     },
     {
       "epoch": 0.20161290322580644,
-      "eval_loss": 1.0701438188552856,
-      "eval_runtime": 174.6767,
-      "eval_samples_per_second": 40.2,
-      "eval_steps_per_second": 2.513,
-      "eval_wer": 0.7648786191684716,
+      "eval_loss": 1.2700377702713013,
+      "eval_runtime": 187.6926,
+      "eval_samples_per_second": 37.412,
+      "eval_steps_per_second": 2.339,
+      "eval_wer": 0.8484433755431244,
       "step": 600
     },
     {
       "epoch": 0.26881720430107525,
-      "eval_loss": 0.8867517113685608,
-      "eval_runtime": 174.8944,
-      "eval_samples_per_second": 40.15,
-      "eval_steps_per_second": 2.51,
-      "eval_wer": 0.6947474720631419,
+      "eval_loss": 0.9953192472457886,
+      "eval_runtime": 187.642,
+      "eval_samples_per_second": 37.422,
+      "eval_steps_per_second": 2.34,
+      "eval_wer": 0.7435389787267968,
       "step": 800
     },
     {
       "epoch": 0.33602150537634407,
-      "grad_norm": 2.852606773376465,
-      "learning_rate": 0.00022288888888888887,
-      "loss": 0.9333,
+      "grad_norm": 5.077725410461426,
+      "learning_rate": 0.0002232222222222222,
+      "loss": 1.0707,
       "step": 1000
     },
     {
       "epoch": 0.33602150537634407,
-      "eval_loss": 0.7679557204246521,
-      "eval_runtime": 175.5538,
-      "eval_samples_per_second": 39.999,
-      "eval_steps_per_second": 2.501,
-      "eval_wer": 0.6070503195631087,
+      "eval_loss": 0.8646696209907532,
+      "eval_runtime": 187.682,
+      "eval_samples_per_second": 37.414,
+      "eval_steps_per_second": 2.339,
+      "eval_wer": 0.6541277455188084,
       "step": 1000
     },
     {
       "epoch": 0.4032258064516129,
-      "eval_loss": 0.7223904132843018,
-      "eval_runtime": 175.6961,
-      "eval_samples_per_second": 39.967,
-      "eval_steps_per_second": 2.499,
-      "eval_wer": 0.5453965638661157,
+      "eval_loss": 0.7888585925102234,
+      "eval_runtime": 187.1558,
+      "eval_samples_per_second": 37.52,
+      "eval_steps_per_second": 2.346,
+      "eval_wer": 0.5784025830797646,
       "step": 1200
     },
     {
       "epoch": 0.47043010752688175,
-      "eval_loss": 0.6732765436172485,
-      "eval_runtime": 175.6064,
-      "eval_samples_per_second": 39.987,
-      "eval_steps_per_second": 2.5,
-      "eval_wer": 0.5121646580475425,
+      "eval_loss": 0.7465152740478516,
+      "eval_runtime": 185.9206,
+      "eval_samples_per_second": 37.769,
+      "eval_steps_per_second": 2.361,
+      "eval_wer": 0.5440412442365697,
       "step": 1400
     },
     {
       "epoch": 0.5040322580645161,
-      "grad_norm": 4.027652740478516,
-      "learning_rate": 0.00016744444444444443,
-      "loss": 0.7446,
+      "grad_norm": 7.621553897857666,
+      "learning_rate": 0.00016777777777777776,
+      "loss": 0.8175,
       "step": 1500
     },
     {
       "epoch": 0.5376344086021505,
-      "eval_loss": 0.6437448859214783,
-      "eval_runtime": 175.5304,
-      "eval_samples_per_second": 40.004,
-      "eval_steps_per_second": 2.501,
-      "eval_wer": 0.4966449195445063,
+      "eval_loss": 0.68277907371521,
+      "eval_runtime": 187.6331,
+      "eval_samples_per_second": 37.424,
+      "eval_steps_per_second": 2.34,
+      "eval_wer": 0.5042453394278426,
       "step": 1600
     },
     {
       "epoch": 0.6048387096774194,
-      "eval_loss": 0.6063565015792847,
-      "eval_runtime": 175.877,
-      "eval_samples_per_second": 39.926,
-      "eval_steps_per_second": 2.496,
-      "eval_wer": 0.4774312706785899,
+      "eval_loss": 0.6549283862113953,
+      "eval_runtime": 186.6705,
+      "eval_samples_per_second": 37.617,
+      "eval_steps_per_second": 2.352,
+      "eval_wer": 0.4952098752308694,
       "step": 1800
     },
     {
       "epoch": 0.6720430107526881,
-      "grad_norm": 3.676745653152466,
-      "learning_rate": 0.000112,
-      "loss": 0.6579,
+      "grad_norm": 6.983826637268066,
+      "learning_rate": 0.00011233333333333333,
+      "loss": 0.7148,
       "step": 2000
     },
     {
       "epoch": 0.6720430107526881,
-      "eval_loss": 0.5673760771751404,
-      "eval_runtime": 176.4608,
-      "eval_samples_per_second": 39.794,
-      "eval_steps_per_second": 2.488,
-      "eval_wer": 0.44613933217289625,
+      "eval_loss": 0.6289859414100647,
+      "eval_runtime": 188.3084,
+      "eval_samples_per_second": 37.29,
+      "eval_steps_per_second": 2.331,
+      "eval_wer": 0.4905858435535949,
       "step": 2000
     },
     {
       "epoch": 0.739247311827957,
-      "eval_loss": 0.5556111931800842,
-      "eval_runtime": 175.9454,
-      "eval_samples_per_second": 39.91,
-      "eval_steps_per_second": 2.495,
-      "eval_wer": 0.4325196986406941,
+      "eval_loss": 0.6112708449363708,
+      "eval_runtime": 188.0636,
+      "eval_samples_per_second": 37.338,
+      "eval_steps_per_second": 2.334,
+      "eval_wer": 0.45763297412934,
       "step": 2200
     },
     {
       "epoch": 0.8064516129032258,
-      "eval_loss": 0.5264282822608948,
-      "eval_runtime": 176.4778,
-      "eval_samples_per_second": 39.79,
-      "eval_steps_per_second": 2.488,
-      "eval_wer": 0.4180363810308402,
+      "eval_loss": 0.5718730688095093,
+      "eval_runtime": 187.8424,
+      "eval_samples_per_second": 37.382,
+      "eval_steps_per_second": 2.337,
+      "eval_wer": 0.4404788796024396,
       "step": 2400
     },
     {
       "epoch": 0.8400537634408602,
-      "grad_norm": 2.7548441886901855,
-      "learning_rate": 5.666666666666666e-05,
-      "loss": 0.5823,
+      "grad_norm": 6.760587215423584,
+      "learning_rate": 5.6999999999999996e-05,
+      "loss": 0.6374,
       "step": 2500
     },
     {
       "epoch": 0.8736559139784946,
-      "eval_loss": 0.5129852890968323,
-      "eval_runtime": 176.2054,
-      "eval_samples_per_second": 39.851,
-      "eval_steps_per_second": 2.491,
-      "eval_wer": 0.4022110312387887,
+      "eval_loss": 0.5643858313560486,
+      "eval_runtime": 188.0058,
+      "eval_samples_per_second": 37.35,
+      "eval_steps_per_second": 2.335,
+      "eval_wer": 0.431376978168724,
       "step": 2600
     },
     {
       "epoch": 0.9408602150537635,
-      "eval_loss": 0.49821802973747253,
-      "eval_runtime": 176.2887,
-      "eval_samples_per_second": 39.832,
-      "eval_steps_per_second": 2.49,
-      "eval_wer": 0.39360076535696725,
+      "eval_loss": 0.5482733249664307,
+      "eval_runtime": 188.1104,
+      "eval_samples_per_second": 37.329,
+      "eval_steps_per_second": 2.334,
+      "eval_wer": 0.41896650234523447,
       "step": 2800
     },
     {
       "epoch": 1.0080645161290323,
-      "grad_norm": 0.7031016945838928,
-      "learning_rate": 1.111111111111111e-06,
-      "loss": 0.5426,
+      "grad_norm": 1.7233390808105469,
+      "learning_rate": 1.4444444444444445e-06,
+      "loss": 0.6013,
       "step": 3000
     },
     {
       "epoch": 1.0080645161290323,
-      "eval_loss": 0.4942198395729065,
-      "eval_runtime": 177.2018,
-      "eval_samples_per_second": 39.627,
-      "eval_steps_per_second": 2.477,
-      "eval_wer": 0.3917272352808302,
+      "eval_loss": 0.5417820811271667,
+      "eval_runtime": 189.023,
+      "eval_samples_per_second": 37.149,
+      "eval_steps_per_second": 2.322,
+      "eval_wer": 0.4178370693206128,
       "step": 3000
     },
     {
       "epoch": 1.0080645161290323,
       "step": 3000,
       "total_flos": 6.709869296482936e+18,
-      "train_loss": 1.1787635701497396,
-      "train_runtime": 4443.9077,
-      "train_samples_per_second": 10.801,
-      "train_steps_per_second": 0.675
+      "train_loss": 1.3100875549316406,
+      "train_runtime": 4950.9591,
+      "train_samples_per_second": 9.695,
+      "train_steps_per_second": 0.606
     }
   ],
   "logging_steps": 500,