update dataset configuration
Browse files- README.md +0 -2
- conf/config.yaml +3 -2
README.md
CHANGED
|
@@ -63,8 +63,6 @@ The model has been originally trained on the MSP-Podcast (v1.7) dataset by [audE
|
|
| 63 |
The model was further fine-tuned on the MSP-Podcast (v1.11) dataset, a large corpus of spontaneous emotional speech collected from various podcast recordings.
|
| 64 |
The dataset includes natural emotional expressions which cover a broad range of speakers, recording conditions, and conversation topics.
|
| 65 |
|
| 66 |
-
**Note:** The MSP-Podcast dataset is not yet included in the autrainer 0.5.0 release but can be found in [this Pull Request](https://github.com/autrainer/autrainer/pull/46).
|
| 67 |
-
|
| 68 |
### Training Process
|
| 69 |
|
| 70 |
The model has been fine-tuned for 5 epochs.
|
|
|
|
| 63 |
The model was further fine-tuned on the MSP-Podcast (v1.11) dataset, a large corpus of spontaneous emotional speech collected from various podcast recordings.
|
| 64 |
The dataset includes natural emotional expressions which cover a broad range of speakers, recording conditions, and conversation topics.
|
| 65 |
|
|
|
|
|
|
|
| 66 |
### Training Process
|
| 67 |
|
| 68 |
The model has been fine-tuned for 5 epochs.
|
conf/config.yaml
CHANGED
|
@@ -6,13 +6,14 @@ results_dir: results
|
|
| 6 |
experiment_id: default
|
| 7 |
iterations: 5
|
| 8 |
|
|
|
|
|
|
|
| 9 |
hydra:
|
| 10 |
sweeper:
|
| 11 |
params:
|
| 12 |
+seed: 1
|
| 13 |
+batch_size: 4
|
| 14 |
-
+inference_batch_size: 1
|
| 15 |
+learning_rate: 0.01
|
| 16 |
optimizer: SGD
|
| 17 |
-
dataset:
|
| 18 |
model: w2v2-l-emo
|
|
|
|
| 6 |
experiment_id: default
|
| 7 |
iterations: 5
|
| 8 |
|
| 9 |
+
inference_batch_size: 1
|
| 10 |
+
|
| 11 |
hydra:
|
| 12 |
sweeper:
|
| 13 |
params:
|
| 14 |
+seed: 1
|
| 15 |
+batch_size: 4
|
|
|
|
| 16 |
+learning_rate: 0.01
|
| 17 |
optimizer: SGD
|
| 18 |
+
dataset: MSPPodcast-EmoClass-big4-wav
|
| 19 |
model: w2v2-l-emo
|