End of training

README.md

@@ -14,13 +14,13 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model was trained from scratch on an unknown dataset.
 It achieves the following results on the evaluation set:
-- eval_loss: 2.3722
+- eval_loss: nan
 - eval_model_preparation_time: 0.0054
-- eval_cer: 0.3572
-- eval_wer: 0.5040
-- eval_runtime: 47.3827
-- eval_samples_per_second: 12.072
-- eval_steps_per_second: 0.76
+- eval_cer: 1.0
+- eval_wer: 1.0
+- eval_runtime: 46.54
+- eval_samples_per_second: 12.29
+- eval_steps_per_second: 1.547
 - step: 0
 
 ## Model description
@@ -41,11 +41,11 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0006
-- train_batch_size: 16
-- eval_batch_size: 16
+- train_batch_size: 8
+- eval_batch_size: 8
 - seed: 300
 - gradient_accumulation_steps: 2
-- total_train_batch_size: 32
+- total_train_batch_size: 16
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 500

all_results.json

@@ -1,10 +1,10 @@
 {
-    "eval_cer": 0.35724199025542597,
-    "eval_loss": 2.372227191925049,
+    "eval_cer": 1.0,
+    "eval_loss": NaN,
     "eval_model_preparation_time": 0.0054,
-    "eval_runtime": 47.3827,
+    "eval_runtime": 46.54,
     "eval_samples": 572,
-    "eval_samples_per_second": 12.072,
-    "eval_steps_per_second": 0.76,
-    "eval_wer": 0.5040222319730876
+    "eval_samples_per_second": 12.29,
+    "eval_steps_per_second": 1.547,
+    "eval_wer": 1.0
 }

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/m/triton/scratch/elec/puhe/p/palp3/MUCS/indicwav2vec-hindi",
+  "_name_or_path": "/scratch/elec/puhe/p/palp3/MUCS/indicwav2vec_outputs/pd_warmup500_rerun_latest/batchsize8/output",
   "activation_dropout": 0.0,
   "adapter_attn_dim": null,
   "adapter_kernel_size": 3,

eval_results.json

@@ -1,10 +1,10 @@
 {
-    "eval_cer": 0.35724199025542597,
-    "eval_loss": 2.372227191925049,
+    "eval_cer": 1.0,
+    "eval_loss": NaN,
     "eval_model_preparation_time": 0.0054,
-    "eval_runtime": 47.3827,
+    "eval_runtime": 46.54,
     "eval_samples": 572,
-    "eval_samples_per_second": 12.072,
-    "eval_steps_per_second": 0.76,
-    "eval_wer": 0.5040222319730876
+    "eval_samples_per_second": 12.29,
+    "eval_steps_per_second": 1.547,
+    "eval_wer": 1.0
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:88d5758118fce348aebadc5cf828d5353f39c3dd822ba51c3d9b83fec216ffa6
+oid sha256:0c825e3070fcd897d9b76e2c6776787b1422865718ef077585ebc5e5c964ac84
 size 5496