Upload dataset
Browse files
README.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
---
|
| 2 |
dataset_info:
|
| 3 |
-
|
| 4 |
features:
|
| 5 |
- name: task
|
| 6 |
dtype: string
|
|
@@ -52,9 +52,65 @@ dataset_info:
|
|
| 52 |
num_examples: 200
|
| 53 |
download_size: 321584878
|
| 54 |
dataset_size: 328893148.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 55 |
configs:
|
| 56 |
- config_name: bg_all_consistency
|
| 57 |
data_files:
|
| 58 |
- split: train
|
| 59 |
path: bg_all_consistency/train-*
|
|
|
|
|
|
|
|
|
|
|
|
|
| 60 |
---
|
|
|
|
| 1 |
---
|
| 2 |
dataset_info:
|
| 3 |
+
- config_name: bg_all_consistency
|
| 4 |
features:
|
| 5 |
- name: task
|
| 6 |
dtype: string
|
|
|
|
| 52 |
num_examples: 200
|
| 53 |
download_size: 321584878
|
| 54 |
dataset_size: 328893148.0
|
| 55 |
+
- config_name: bg_domain_consistency
|
| 56 |
+
features:
|
| 57 |
+
- name: task
|
| 58 |
+
dtype: string
|
| 59 |
+
- name: ind
|
| 60 |
+
dtype: int64
|
| 61 |
+
- name: positive_audio
|
| 62 |
+
dtype: audio
|
| 63 |
+
- name: negative_audio
|
| 64 |
+
dtype: audio
|
| 65 |
+
- name: audio_transition_s
|
| 66 |
+
dtype: int64
|
| 67 |
+
- name: prompt_audio
|
| 68 |
+
dtype:
|
| 69 |
+
audio:
|
| 70 |
+
sampling_rate: 16000
|
| 71 |
+
- name: continuation_audio_positive
|
| 72 |
+
dtype:
|
| 73 |
+
audio:
|
| 74 |
+
sampling_rate: 16000
|
| 75 |
+
- name: continuation_audio_negative
|
| 76 |
+
dtype:
|
| 77 |
+
audio:
|
| 78 |
+
sampling_rate: 16000
|
| 79 |
+
- name: negative_audio_sanity
|
| 80 |
+
dtype:
|
| 81 |
+
audio:
|
| 82 |
+
sampling_rate: 16000
|
| 83 |
+
- name: postive_sample_tokenwise_loss
|
| 84 |
+
sequence: float32
|
| 85 |
+
- name: negative_sample_tokenwise_loss
|
| 86 |
+
sequence: float32
|
| 87 |
+
- name: prompt_sample_tokenwise_loss
|
| 88 |
+
sequence: float32
|
| 89 |
+
- name: model_generated_continuation
|
| 90 |
+
dtype:
|
| 91 |
+
audio:
|
| 92 |
+
sampling_rate: 22050
|
| 93 |
+
- name: code_frame_rate
|
| 94 |
+
sequence: int64
|
| 95 |
+
- name: code_depth
|
| 96 |
+
dtype: int64
|
| 97 |
+
- name: model_sampling_rate
|
| 98 |
+
sequence: int64
|
| 99 |
+
- name: ppl_sanity
|
| 100 |
+
dtype: int64
|
| 101 |
+
splits:
|
| 102 |
+
- name: train
|
| 103 |
+
num_bytes: 332347830.0
|
| 104 |
+
num_examples: 200
|
| 105 |
+
download_size: 324695408
|
| 106 |
+
dataset_size: 332347830.0
|
| 107 |
configs:
|
| 108 |
- config_name: bg_all_consistency
|
| 109 |
data_files:
|
| 110 |
- split: train
|
| 111 |
path: bg_all_consistency/train-*
|
| 112 |
+
- config_name: bg_domain_consistency
|
| 113 |
+
data_files:
|
| 114 |
+
- split: train
|
| 115 |
+
path: bg_domain_consistency/train-*
|
| 116 |
---
|
bg_domain_consistency/train-00000-of-00001.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:23286ecbb38d1f757fa16059b5ec35a5d2ab2921487c4be24f4e3b20e0625164
|
| 3 |
+
size 324695408
|