Saving train state of step 25

=0.4.1

@@ -11,7 +11,7 @@ Requirement already satisfied: tensorboard in /opt/conda/lib/python3.10/site-pac
 Requirement already satisfied: nltk in /opt/conda/lib/python3.10/site-packages (3.2.4)
 Collecting deepspeed
   Downloading deepspeed-0.15.1.tar.gz (1.4 MB)
-[?25l     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/1.4 MB ? eta -:--:--
+[?25l     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/1.4 MB ? eta -:--:--
 [?25h  Preparing metadata (setup.py): started
   Preparing metadata (setup.py): finished with status 'done'
 Requirement already satisfied: datasets[audio] in /opt/conda/lib/python3.10/site-packages (2.21.0)
@@ -100,13 +100,13 @@ Downloading jiwer-3.0.4-py3-none-any.whl (21 kB)
 Downloading evaluate-0.4.3-py3-none-any.whl (84 kB)
 [?25l   ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/84.0 kB ? eta -:--:--
 [?25hDownloading rapidfuzz-3.9.7-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (3.4 MB)
-[?25l   ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/3.4 MB ? eta -:--:--
+[?25l   ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/3.4 MB ? eta -:--:--
 [?25hDownloading hjson-3.1.0-py3-none-any.whl (54 kB)
-[?25l   ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/54.0 kB ? eta -:--:--
+[?25l   ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/54.0 kB ? eta -:--:--
 [?25hBuilding wheels for collected packages: deepspeed
   Building wheel for deepspeed (setup.py): started
   Building wheel for deepspeed (setup.py): finished with status 'done'
-  Created wheel for deepspeed: filename=deepspeed-0.15.1-py3-none-any.whl size=1483865 sha256=a2392a5cd494e5ddb590fc34407243343522a79cfdc33f16098f4b314c09864c
+  Created wheel for deepspeed: filename=deepspeed-0.15.1-py3-none-any.whl size=1483868 sha256=38b4ab969914052bc651b1dd6dc7d849b3942c6f54b7d89ee306fcdc98897498
   Stored in directory: /root/.cache/pip/wheels/da/cb/14/9cbba50c73df044eb32a7ca29e34844c5f8959e12d22ae8b60
 Successfully built deepspeed
 Installing collected packages: hjson, rapidfuzz, jiwer, deepspeed, evaluate

checkpoint-25-epoch-1/config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "./distil-medium-init",
+  "_name_or_path": "KevinKibe/working",
   "activation_dropout": 0.0,
   "activation_function": "gelu",
   "apply_spec_augment": false,

checkpoint-25-epoch-1/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0d1c17c292a62042e30703b06315a543a3530cabd5e5d095ded3f0acaee0a613
+oid sha256:c573391e90182ffb2b66b51adec40e372b6861763ddbcd569b849863c14f9939
 size 1577553712

checkpoint-25-epoch-1/optimizer.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a5bc121bf336ceb95cdb99f5fe4cbfd1aaae1b4acbfbf2ed0021235acf894975
+oid sha256:a1c9443d464aef2e44945dfd78d1e4d720747ee2a8b7c2357510321f68b81596
 size 693652474

distil-whisper/events.out.tfevents.1726494954.61332a8484de.245.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:beac2350217c024e1b967221f2e5bb1515ccc3dc050dc99f7390f94577bc9cd8
+size 392

distil-whisper/training/run_distillation.py

@@ -1648,6 +1648,7 @@ def main():
                         rotate_checkpoints(training_args.save_total_limit, output_dir=training_args.output_dir)
 
                         if training_args.push_to_hub:
+                            repo_name = training_args.hub_model_id
                             upload_folder(
                                 folder_path=training_args.output_dir,
                                 repo_id=repo_name,
@@ -1769,6 +1770,7 @@ def main():
                             accelerator.unwrap_model(student_model).save_pretrained(training_args.output_dir)
 
                             if training_args.push_to_hub:
+                                repo_name = training_args.hub_model_id
                                 upload_folder(
                                     folder_path=training_args.output_dir,
                                     repo_id=repo_name,
@@ -1795,6 +1797,7 @@ def main():
                     student_model.save_pretrained(final_weights_dir)
 
                     if training_args.push_to_hub:
+                        repo_name = training_args.hub_model_id
                         upload_folder(
                             folder_path=training_args.output_dir,
                             repo_id=repo_name,

distil-whisper/training/run_pseudo_labelling.py

@@ -508,6 +508,7 @@ def main():
                 token=token,
                 streaming=data_args.streaming,
                 num_proc=data_args.preprocessing_num_workers if not data_args.streaming else None,
+                trust_remote_code=True
             )
 
     if data_args.audio_column_name not in next(iter(raw_datasets.values())).column_names: