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
 - generated_from_trainer
 metrics:
 - wer
@@ -15,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # wav2vec2-xlsr-53-ft-btb-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 - DEFAULT dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.5400
 - Wer: 0.4066

all_results.json

@@ -1,15 +1,15 @@
 {
-    "epoch": 1.366742596810934,
-    "eval_loss": 0.5374971628189087,
-    "eval_runtime": 177.247,
+    "epoch": 2.277904328018223,
+    "eval_loss": 0.5399738550186157,
+    "eval_runtime": 177.5085,
     "eval_samples": 7022,
-    "eval_samples_per_second": 39.617,
-    "eval_steps_per_second": 4.954,
-    "eval_wer": 0.4141431589577326,
-    "total_flos": 4.530371419942491e+18,
-    "train_loss": 1.872582066853841,
-    "train_runtime": 2251.6929,
+    "eval_samples_per_second": 39.559,
+    "eval_steps_per_second": 4.946,
+    "eval_wer": 0.40660917631113885,
+    "total_flos": 7.50335828249239e+18,
+    "train_loss": 1.4263343353271485,
+    "train_runtime": 3812.7792,
     "train_samples": 28086,
-    "train_samples_per_second": 17.054,
-    "train_steps_per_second": 0.533
+    "train_samples_per_second": 16.786,
+    "train_steps_per_second": 0.525
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 1.366742596810934,
-    "eval_loss": 0.5374971628189087,
-    "eval_runtime": 177.247,
+    "epoch": 2.277904328018223,
+    "eval_loss": 0.5399738550186157,
+    "eval_runtime": 177.5085,
     "eval_samples": 7022,
-    "eval_samples_per_second": 39.617,
-    "eval_steps_per_second": 4.954,
-    "eval_wer": 0.4141431589577326
+    "eval_samples_per_second": 39.559,
+    "eval_steps_per_second": 4.946,
+    "eval_wer": 0.40660917631113885
 }

runs/Aug15_10-29-01_a142c9bbb9ba/events.out.tfevents.1723718214.a142c9bbb9ba.948.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b51cf3f92089086f467cb7f968ca4f563fc1aff9703b93be03bff2c3d486c23
+size 406

train_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 1.366742596810934,
-    "total_flos": 4.530371419942491e+18,
-    "train_loss": 1.872582066853841,
-    "train_runtime": 2251.6929,
+    "epoch": 2.277904328018223,
+    "total_flos": 7.50335828249239e+18,
+    "train_loss": 1.4263343353271485,
+    "train_runtime": 3812.7792,
     "train_samples": 28086,
-    "train_samples_per_second": 17.054,
-    "train_steps_per_second": 0.533
+    "train_samples_per_second": 16.786,
+    "train_steps_per_second": 0.525
 }

trainer_state.json

@@ -1,95 +1,145 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.366742596810934,
+  "epoch": 2.277904328018223,
   "eval_steps": 200,
-  "global_step": 1200,
+  "global_step": 2000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.22779043280182232,
-      "eval_loss": 2.9538023471832275,
-      "eval_runtime": 176.5728,
-      "eval_samples_per_second": 39.768,
-      "eval_steps_per_second": 4.972,
+      "eval_loss": 2.9681296348571777,
+      "eval_runtime": 179.5569,
+      "eval_samples_per_second": 39.107,
+      "eval_steps_per_second": 4.89,
       "eval_wer": 1.0,
       "step": 200
     },
     {
       "epoch": 0.45558086560364464,
-      "eval_loss": 1.4731117486953735,
-      "eval_runtime": 177.0417,
-      "eval_samples_per_second": 39.663,
-      "eval_steps_per_second": 4.959,
-      "eval_wer": 0.8217488938200083,
+      "eval_loss": 1.3578158617019653,
+      "eval_runtime": 178.971,
+      "eval_samples_per_second": 39.235,
+      "eval_steps_per_second": 4.906,
+      "eval_wer": 0.8299738237287235,
       "step": 400
     },
     {
       "epoch": 0.5694760820045558,
-      "grad_norm": 2.272386074066162,
+      "grad_norm": 2.982609748840332,
       "learning_rate": 0.00029699999999999996,
-      "loss": 3.3429,
+      "loss": 3.3875,
       "step": 500
     },
     {
       "epoch": 0.683371298405467,
-      "eval_loss": 0.852441132068634,
-      "eval_runtime": 176.02,
-      "eval_samples_per_second": 39.893,
-      "eval_steps_per_second": 4.988,
-      "eval_wer": 0.6397905898297878,
+      "eval_loss": 0.9213220477104187,
+      "eval_runtime": 178.9628,
+      "eval_samples_per_second": 39.237,
+      "eval_steps_per_second": 4.906,
+      "eval_wer": 0.6513373815756255,
       "step": 600
     },
     {
       "epoch": 0.9111617312072893,
-      "eval_loss": 0.7107498645782471,
-      "eval_runtime": 176.239,
-      "eval_samples_per_second": 39.844,
-      "eval_steps_per_second": 4.982,
-      "eval_wer": 0.5182237340384539,
+      "eval_loss": 0.7288344502449036,
+      "eval_runtime": 180.4166,
+      "eval_samples_per_second": 38.921,
+      "eval_steps_per_second": 4.867,
+      "eval_wer": 0.5341155210672478,
       "step": 800
     },
     {
       "epoch": 1.1389521640091116,
-      "grad_norm": 0.8374969959259033,
-      "learning_rate": 9.042857142857142e-05,
-      "loss": 0.8743,
+      "grad_norm": 1.1159461736679077,
+      "learning_rate": 0.0002022,
+      "loss": 0.8947,
       "step": 1000
     },
     {
       "epoch": 1.1389521640091116,
-      "eval_loss": 0.5667783617973328,
-      "eval_runtime": 177.0088,
-      "eval_samples_per_second": 39.67,
-      "eval_steps_per_second": 4.96,
-      "eval_wer": 0.4325462735353911,
+      "eval_loss": 0.6245301961898804,
+      "eval_runtime": 180.7702,
+      "eval_samples_per_second": 38.845,
+      "eval_steps_per_second": 4.857,
+      "eval_wer": 0.46335986393653916,
       "step": 1000
     },
     {
       "epoch": 1.366742596810934,
-      "eval_loss": 0.5374971628189087,
-      "eval_runtime": 176.8891,
-      "eval_samples_per_second": 39.697,
-      "eval_steps_per_second": 4.964,
-      "eval_wer": 0.4141431589577326,
+      "eval_loss": 0.5951615571975708,
+      "eval_runtime": 177.8647,
+      "eval_samples_per_second": 39.479,
+      "eval_steps_per_second": 4.936,
+      "eval_wer": 0.445833720883881,
       "step": 1200
     },
     {
-      "epoch": 1.366742596810934,
-      "step": 1200,
-      "total_flos": 4.530371419942491e+18,
-      "train_loss": 1.872582066853841,
-      "train_runtime": 2251.6929,
-      "train_samples_per_second": 17.054,
-      "train_steps_per_second": 0.533
+      "epoch": 1.5945330296127562,
+      "eval_loss": 0.578143835067749,
+      "eval_runtime": 178.6987,
+      "eval_samples_per_second": 39.295,
+      "eval_steps_per_second": 4.913,
+      "eval_wer": 0.4384857625001661,
+      "step": 1400
+    },
+    {
+      "epoch": 1.7084282460136673,
+      "grad_norm": 1.4777103662490845,
+      "learning_rate": 0.00010279999999999999,
+      "loss": 0.7523,
+      "step": 1500
+    },
+    {
+      "epoch": 1.8223234624145785,
+      "eval_loss": 0.562013566493988,
+      "eval_runtime": 178.104,
+      "eval_samples_per_second": 39.426,
+      "eval_steps_per_second": 4.93,
+      "eval_wer": 0.42830757783122286,
+      "step": 1600
+    },
+    {
+      "epoch": 2.050113895216401,
+      "eval_loss": 0.5437431931495667,
+      "eval_runtime": 177.6652,
+      "eval_samples_per_second": 39.524,
+      "eval_steps_per_second": 4.942,
+      "eval_wer": 0.4140368593789447,
+      "step": 1800
+    },
+    {
+      "epoch": 2.277904328018223,
+      "grad_norm": 3.888875961303711,
+      "learning_rate": 2.9999999999999997e-06,
+      "loss": 0.6707,
+      "step": 2000
+    },
+    {
+      "epoch": 2.277904328018223,
+      "eval_loss": 0.5399738550186157,
+      "eval_runtime": 178.027,
+      "eval_samples_per_second": 39.443,
+      "eval_steps_per_second": 4.932,
+      "eval_wer": 0.40660917631113885,
+      "step": 2000
+    },
+    {
+      "epoch": 2.277904328018223,
+      "step": 2000,
+      "total_flos": 7.50335828249239e+18,
+      "train_loss": 1.4263343353271485,
+      "train_runtime": 3812.7792,
+      "train_samples_per_second": 16.786,
+      "train_steps_per_second": 0.525
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1200,
+  "max_steps": 2000,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 2,
+  "num_train_epochs": 3,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -103,7 +153,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 4.530371419942491e+18,
+  "total_flos": 7.50335828249239e+18,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null