Training in progress, step 300

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f19397a77dfbf603c0e1c72990858c57d2da9d95e1055ccbb435dc5329d4be8
+oid sha256:731dc59a1e280e969912dcb444e05f6d7cd74c469c771cc391ba8d004b163203
 size 151098921

run_1.sh

@@ -6,13 +6,13 @@ python run_speech_recognition_seq2seq_streaming.py \
 --train_split_name="train+validation" \
 --eval_split_name="test" \
 --model_index_name="Whisper Tiny Pashto" \
---max_steps="200" \
+--max_steps="500" \
 --output_dir="./" \
 --per_device_train_batch_size="64" \
 --per_device_eval_batch_size="32" \
 --gradient_accumulation_steps="1" \
 --logging_steps="10" \
---learning_rate="3e-7" \
+--learning_rate="1e-5" \
 --warmup_steps="40" \
 --evaluation_strategy="steps" \
 --eval_steps="100" \
@@ -27,7 +27,7 @@ python run_speech_recognition_seq2seq_streaming.py \
 --report_to="tensorboard" \
 --gradient_checkpointing \
 --fp16 \
---overwrite_output_dir="True" \
+--overwrite_output_dir="False" \
 --do_train \
 --do_eval \
 --predict_with_generate \

runs/Dec18_15-08-41_129-146-179-188/1671376168.991209/events.out.tfevents.1671376168.129-146-179-188.129306.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:390c2edf5276aceb867282ca1ab088644057eb35d5d06d4d91655f2bde8c6954
+size 5872

runs/Dec18_15-08-41_129-146-179-188/events.out.tfevents.1671376168.129-146-179-188.129306.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ccbed61b41602bb65beccd78c4f3eda74609ee7ff0c13ce4d477f18a7689c343
+size 6148

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5cc4720d9aed8e3dfa8de40898a3a0253a9be1fbdae340a19e1743e520f212e6
+oid sha256:1fe94ef099a4a3b2b829dc44e8618397ff53dca3a1f192feca2b2d6fd1d4aa2f
 size 3579