End of training

README.md

@@ -3,6 +3,9 @@ library_name: transformers
 license: cc-by-nc-4.0
 base_model: facebook/mms-300m
 tags:
+- automatic-speech-recognition
+- libri10h
+- mms
 - generated_from_trainer
 metrics:
 - wer
@@ -16,9 +19,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # wav2vec2-300m-librispeech-test
 
-This model is a fine-tuned version of [facebook/mms-300m](https://huggingface.co/facebook/mms-300m) on an unknown dataset.
+This model is a fine-tuned version of [facebook/mms-300m](https://huggingface.co/facebook/mms-300m) on the LIBRI10H - ENG dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.8867
+- Loss: 2.8909
 - Wer: 1.0
 
 ## Model description

adapter.eng.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0607c191653db5229f1b1327c634e798f0d851bfe935cfb07ca3071fb396db1a
+size 3586804

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 2.0,
+    "eval_loss": 2.890918731689453,
+    "eval_runtime": 144.4166,
+    "eval_samples": 2604,
+    "eval_samples_per_second": 18.031,
+    "eval_steps_per_second": 4.508,
+    "eval_wer": 1.0,
+    "total_flos": 2.1389390179654113e+18,
+    "train_loss": 3.3086385533429574,
+    "train_runtime": 2151.9759,
+    "train_samples": 2759,
+    "train_samples_per_second": 2.564,
+    "train_steps_per_second": 0.321
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 2.0,
+    "eval_loss": 2.890918731689453,
+    "eval_runtime": 144.4166,
+    "eval_samples": 2604,
+    "eval_samples_per_second": 18.031,
+    "eval_steps_per_second": 4.508,
+    "eval_wer": 1.0
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 2.0,
+    "total_flos": 2.1389390179654113e+18,
+    "train_loss": 3.3086385533429574,
+    "train_runtime": 2151.9759,
+    "train_samples": 2759,
+    "train_samples_per_second": 2.564,
+    "train_steps_per_second": 0.321
+}

trainer_state.json

@@ -0,0 +1,138 @@
+{
+  "best_metric": 2.881359577178955,
+  "best_model_checkpoint": "/scratch/skscla001/speech/results/wav2vec2-300m-librispeech-test/checkpoint-500",
+  "epoch": 2.0,
+  "eval_steps": 100,
+  "global_step": 690,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.2898550724637681,
+      "grad_norm": 0.9473298192024231,
+      "learning_rate": 0.00029699999999999996,
+      "loss": 6.0416,
+      "step": 100
+    },
+    {
+      "epoch": 0.2898550724637681,
+      "eval_loss": 2.919609785079956,
+      "eval_runtime": 144.5606,
+      "eval_samples_per_second": 18.013,
+      "eval_steps_per_second": 4.503,
+      "eval_wer": 1.0,
+      "step": 100
+    },
+    {
+      "epoch": 0.5797101449275363,
+      "grad_norm": 0.6596241593360901,
+      "learning_rate": 0.0002496610169491525,
+      "loss": 2.8561,
+      "step": 200
+    },
+    {
+      "epoch": 0.5797101449275363,
+      "eval_loss": 2.887662410736084,
+      "eval_runtime": 143.5989,
+      "eval_samples_per_second": 18.134,
+      "eval_steps_per_second": 4.533,
+      "eval_wer": 1.0,
+      "step": 200
+    },
+    {
+      "epoch": 0.8695652173913043,
+      "grad_norm": 0.790766716003418,
+      "learning_rate": 0.0001988135593220339,
+      "loss": 2.8553,
+      "step": 300
+    },
+    {
+      "epoch": 0.8695652173913043,
+      "eval_loss": 2.8884549140930176,
+      "eval_runtime": 143.4069,
+      "eval_samples_per_second": 18.158,
+      "eval_steps_per_second": 4.54,
+      "eval_wer": 1.0,
+      "step": 300
+    },
+    {
+      "epoch": 1.1594202898550725,
+      "grad_norm": 0.22183042764663696,
+      "learning_rate": 0.00014796610169491524,
+      "loss": 2.8423,
+      "step": 400
+    },
+    {
+      "epoch": 1.1594202898550725,
+      "eval_loss": 2.9169750213623047,
+      "eval_runtime": 143.8922,
+      "eval_samples_per_second": 18.097,
+      "eval_steps_per_second": 4.524,
+      "eval_wer": 1.0,
+      "step": 400
+    },
+    {
+      "epoch": 1.4492753623188406,
+      "grad_norm": 0.5317383408546448,
+      "learning_rate": 9.71186440677966e-05,
+      "loss": 2.8446,
+      "step": 500
+    },
+    {
+      "epoch": 1.4492753623188406,
+      "eval_loss": 2.881359577178955,
+      "eval_runtime": 143.963,
+      "eval_samples_per_second": 18.088,
+      "eval_steps_per_second": 4.522,
+      "eval_wer": 1.0,
+      "step": 500
+    },
+    {
+      "epoch": 1.7391304347826086,
+      "grad_norm": 0.1836908459663391,
+      "learning_rate": 4.6271186440677964e-05,
+      "loss": 2.8382,
+      "step": 600
+    },
+    {
+      "epoch": 1.7391304347826086,
+      "eval_loss": 2.886660575866699,
+      "eval_runtime": 144.326,
+      "eval_samples_per_second": 18.042,
+      "eval_steps_per_second": 4.511,
+      "eval_wer": 1.0,
+      "step": 600
+    },
+    {
+      "epoch": 2.0,
+      "step": 690,
+      "total_flos": 2.1389390179654113e+18,
+      "train_loss": 3.3086385533429574,
+      "train_runtime": 2151.9759,
+      "train_samples_per_second": 2.564,
+      "train_steps_per_second": 0.321
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 690,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 400,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.1389390179654113e+18,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}