End of training

README.md

@@ -1,12 +1,25 @@
 ---
 base_model: nadsoft/hamsa_small
 tags:
+- whisper-event
 - generated_from_trainer
+datasets:
+- nadsoft/QASR-Speech-Resource
 metrics:
 - wer
 model-index:
 - name: hamsa-small-finetuned-qasr
-  results: []
+  results:
+  - task:
+      name: Automatic Speech Recognition
+      type: automatic-speech-recognition
+    dataset:
+      name: nadsoft/QASR-Speech-Resource default
+      type: nadsoft/QASR-Speech-Resource
+    metrics:
+    - name: Wer
+      type: wer
+      value: 22.587152044424403
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -14,10 +27,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hamsa-small-finetuned-qasr
 
-This model is a fine-tuned version of [nadsoft/hamsa_small](https://huggingface.co/nadsoft/hamsa_small) on an unknown dataset.
+This model is a fine-tuned version of [nadsoft/hamsa_small](https://huggingface.co/nadsoft/hamsa_small) on the nadsoft/QASR-Speech-Resource default dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2832
-- Wer: 21.0114
+- Loss: 0.2831
+- Wer: 22.5872
 
 ## Model description
 

all_results.json

@@ -1,7 +1,12 @@
 {
     "epoch": 0.3,
+    "eval_loss": 0.2830975353717804,
+    "eval_runtime": 2204.2826,
+    "eval_samples_per_second": 2.271,
+    "eval_steps_per_second": 0.142,
+    "eval_wer": 22.587152044424403,
     "train_loss": 1.1846045168919276e-05,
-    "train_runtime": 3.611,
-    "train_samples_per_second": 33231.537,
-    "train_steps_per_second": 2076.971
+    "train_runtime": 2.8319,
+    "train_samples_per_second": 112999.353,
+    "train_steps_per_second": 7062.46
 }

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.3,
+    "eval_loss": 0.2830975353717804,
+    "eval_runtime": 2204.2826,
+    "eval_samples_per_second": 2.271,
+    "eval_steps_per_second": 0.142,
+    "eval_wer": 22.587152044424403
+}

generation_config.json

@@ -1,5 +1,4 @@
 {
-  "_from_model_config": true,
   "begin_suppress_tokens": [
     220,
     50257

runs/Feb23_14-08-18_ip-10-0-3-5.eu-west-1.compute.internal/events.out.tfevents.1708699558.ip-10-0-3-5.eu-west-1.compute.internal.17482.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f770ad1cd91f0426b30031bfbd36e3802a1b1045843df97b9909aaa48879a297
+size 412

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 0.3,
     "train_loss": 1.1846045168919276e-05,
-    "train_runtime": 3.611,
-    "train_samples_per_second": 33231.537,
-    "train_steps_per_second": 2076.971
+    "train_runtime": 2.8319,
+    "train_samples_per_second": 112999.353,
+    "train_steps_per_second": 7062.46
 }

trainer_state.json

@@ -7321,13 +7321,13 @@
       "step": 30001,
       "total_flos": 1.3852560952000512e+20,
       "train_loss": 1.1846045168919276e-05,
-      "train_runtime": 3.611,
-      "train_samples_per_second": 33231.537,
-      "train_steps_per_second": 2076.971
+      "train_runtime": 2.8319,
+      "train_samples_per_second": 112999.353,
+      "train_steps_per_second": 7062.46
     }
   ],
   "logging_steps": 25,
-  "max_steps": 7500,
+  "max_steps": 20000,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 1,
   "save_steps": 2500,