End of training
Browse files
README.md
CHANGED
|
@@ -9,22 +9,9 @@ tags:
|
|
| 9 |
- generated_from_trainer
|
| 10 |
datasets:
|
| 11 |
- chinese_english
|
| 12 |
-
metrics:
|
| 13 |
-
- wer
|
| 14 |
model-index:
|
| 15 |
- name: Whisper tiny Chinese
|
| 16 |
-
results:
|
| 17 |
-
- task:
|
| 18 |
-
name: Automatic Speech Recognition
|
| 19 |
-
type: automatic-speech-recognition
|
| 20 |
-
dataset:
|
| 21 |
-
name: Chinese English
|
| 22 |
-
type: chinese_english
|
| 23 |
-
args: 'config: default, split: test'
|
| 24 |
-
metrics:
|
| 25 |
-
- name: Wer
|
| 26 |
-
type: wer
|
| 27 |
-
value: 25.210484511517077
|
| 28 |
---
|
| 29 |
|
| 30 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -34,8 +21,13 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 34 |
|
| 35 |
This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the Chinese English dataset.
|
| 36 |
It achieves the following results on the evaluation set:
|
| 37 |
-
-
|
| 38 |
-
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 39 |
|
| 40 |
## Model description
|
| 41 |
|
|
@@ -59,17 +51,9 @@ The following hyperparameters were used during training:
|
|
| 59 |
- eval_batch_size: 1
|
| 60 |
- seed: 42
|
| 61 |
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 62 |
-
- lr_scheduler_type:
|
| 63 |
- lr_scheduler_warmup_steps: 500
|
| 64 |
-
-
|
| 65 |
-
|
| 66 |
-
### Training results
|
| 67 |
-
|
| 68 |
-
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 69 |
-
|:-------------:|:------:|:----:|:---------------:|:-------:|
|
| 70 |
-
| 0.0008 | 1.2837 | 1000 | 0.8132 | 25.7347 |
|
| 71 |
-
| 0.0005 | 2.5674 | 2000 | 0.8111 | 25.2105 |
|
| 72 |
-
|
| 73 |
|
| 74 |
### Framework versions
|
| 75 |
|
|
|
|
| 9 |
- generated_from_trainer
|
| 10 |
datasets:
|
| 11 |
- chinese_english
|
|
|
|
|
|
|
| 12 |
model-index:
|
| 13 |
- name: Whisper tiny Chinese
|
| 14 |
+
results: []
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
---
|
| 16 |
|
| 17 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 21 |
|
| 22 |
This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the Chinese English dataset.
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
+
- eval_loss: 0.9038
|
| 25 |
+
- eval_wer: 26.1795
|
| 26 |
+
- eval_runtime: 175.3742
|
| 27 |
+
- eval_samples_per_second: 3.587
|
| 28 |
+
- eval_steps_per_second: 3.587
|
| 29 |
+
- epoch: 2.5674
|
| 30 |
+
- step: 2000
|
| 31 |
|
| 32 |
## Model description
|
| 33 |
|
|
|
|
| 51 |
- eval_batch_size: 1
|
| 52 |
- seed: 42
|
| 53 |
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 54 |
+
- lr_scheduler_type: linear
|
| 55 |
- lr_scheduler_warmup_steps: 500
|
| 56 |
+
- num_epochs: 5
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 57 |
|
| 58 |
### Framework versions
|
| 59 |
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 151061672
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:13b3d629c922f8a2947a1983f1da9ba16eefd3733ad1ee0c22a2c3f3ea368ebf
|
| 3 |
size 151061672
|
runs/Mar16_20-18-31_d28da09ccdee/events.out.tfevents.1742156323.d28da09ccdee.882.3
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9f7130e4b871bdd8c1b04fc2a4769ac0cb04ae06f415529383e1eddb1f163b4e
|
| 3 |
+
size 31785
|