End of training

README.md

@@ -1,6 +1,10 @@
 ---
+language:
+- ps
 base_model: ylacombe/w2v-bert-2.0
 tags:
+- automatic-speech-recognition
+- mozilla-foundation/common_voice_16_0
 - generated_from_trainer
 datasets:
 - common_voice_16_0
@@ -13,11 +17,11 @@ model-index:
       name: Automatic Speech Recognition
       type: automatic-speech-recognition
     dataset:
-      name: common_voice_16_0
+      name: MOZILLA-FOUNDATION/COMMON_VOICE_16_0 - PS
       type: common_voice_16_0
       config: ps
       split: test
-      args: ps
+      args: 'Config: ps, Training split: train+validation, Eval split: test'
     metrics:
     - name: Wer
       type: wer
@@ -29,9 +33,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # wav2vec2-common_voice-ps-demo
 
-This model is a fine-tuned version of [ylacombe/w2v-bert-2.0](https://huggingface.co/ylacombe/w2v-bert-2.0) on the common_voice_16_0 dataset.
+This model is a fine-tuned version of [ylacombe/w2v-bert-2.0](https://huggingface.co/ylacombe/w2v-bert-2.0) on the MOZILLA-FOUNDATION/COMMON_VOICE_16_0 - PS dataset.
 It achieves the following results on the evaluation set:
-- Loss: 3.0949
+- Loss: 3.0510
 - Wer: 0.9484
 
 ## Model description

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 14.77,
+    "eval_loss": 3.051039218902588,
+    "eval_runtime": 4.949,
+    "eval_samples": 195,
+    "eval_samples_per_second": 39.402,
+    "eval_steps_per_second": 5.052,
+    "eval_wer": 0.9484029484029484,
+    "train_loss": 6.593796793619791,
+    "train_runtime": 936.6313,
+    "train_samples": 1027,
+    "train_samples_per_second": 16.447,
+    "train_steps_per_second": 0.512
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 14.77,
+    "eval_loss": 3.051039218902588,
+    "eval_runtime": 4.949,
+    "eval_samples": 195,
+    "eval_samples_per_second": 39.402,
+    "eval_steps_per_second": 5.052,
+    "eval_wer": 0.9484029484029484
+}

runs/Jan01_22-09-15_vorace/events.out.tfevents.1704148023.vorace.125432.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5674dd5f310deb160584533a32537701dce7577b642d8dcd00939a7eaa9b0c4
+size 358

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 14.77,
+    "train_loss": 6.593796793619791,
+    "train_runtime": 936.6313,
+    "train_samples": 1027,
+    "train_samples_per_second": 16.447,
+    "train_steps_per_second": 0.512
+}

trainer_state.json

@@ -0,0 +1,66 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 14.76923076923077,
+  "eval_steps": 100,
+  "global_step": 480,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 3.08,
+      "eval_loss": 10.369086265563965,
+      "eval_runtime": 5.0307,
+      "eval_samples_per_second": 38.762,
+      "eval_steps_per_second": 4.97,
+      "eval_wer": 1.0,
+      "step": 100
+    },
+    {
+      "epoch": 6.15,
+      "eval_loss": 3.5670173168182373,
+      "eval_runtime": 4.9472,
+      "eval_samples_per_second": 39.416,
+      "eval_steps_per_second": 5.053,
+      "eval_wer": 1.0,
+      "step": 200
+    },
+    {
+      "epoch": 9.23,
+      "eval_loss": 3.113863945007324,
+      "eval_runtime": 4.8913,
+      "eval_samples_per_second": 39.866,
+      "eval_steps_per_second": 5.111,
+      "eval_wer": 0.9484029484029484,
+      "step": 300
+    },
+    {
+      "epoch": 12.31,
+      "eval_loss": 3.094938039779663,
+      "eval_runtime": 4.8858,
+      "eval_samples_per_second": 39.912,
+      "eval_steps_per_second": 5.117,
+      "eval_wer": 0.9484029484029484,
+      "step": 400
+    },
+    {
+      "epoch": 14.77,
+      "step": 480,
+      "total_flos": 2.1645545063012813e+18,
+      "train_loss": 6.593796793619791,
+      "train_runtime": 936.6313,
+      "train_samples_per_second": 16.447,
+      "train_steps_per_second": 0.512
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 480,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 15,
+  "save_steps": 400,
+  "total_flos": 2.1645545063012813e+18,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}