Commit
•
9579804
1
Parent(s):
f504d80
Update README.md
Browse files
README.md
CHANGED
@@ -36,19 +36,24 @@ More information needed
|
|
36 |
|
37 |
Evaluation: https://github.com/frank-morales2020/MLxDL/blob/main/FineTunning_Testing_For_EmotionQADataset.ipynb
|
38 |
|
|
|
39 |
The following hyperparameters were used during training:
|
40 |
learning_rate: 0.0002 train_batch_size: 3 eval_batch_size: 8 seed: 42 gradient_accumulation_steps: 2 total_train_batch_size: 6 optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 lr_scheduler_type: constant lr_scheduler_warmup_ratio: 0.03
|
41 |
num_epochs: 1
|
|
|
42 |
NOTE: test - Accuracy (Eval dataset and predict) for a sample of 2000: 59.45%
|
|
|
43 |
|
44 |
The following hyperparameters were used during training:
|
45 |
learning_rate: 0.0002 train_batch_size: 3 eval_batch_size: 8 seed: 42 gradient_accumulation_steps: 2 total_train_batch_size: 6 optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 lr_scheduler_type: constant lr_scheduler_warmup_ratio: 0.03
|
46 |
num_epochs: 25
|
|
|
47 |
NOTE: test - Accuracy (Eval dataset and predict) for a sample of 2000: 79.95%
|
48 |
|
49 |
The following hyperparameters were used during training:
|
50 |
learning_rate: 0.0002 train_batch_size: 3 eval_batch_size: 8 seed: 42 gradient_accumulation_steps: 2 total_train_batch_size: 6 optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 lr_scheduler_type: constant lr_scheduler_warmup_ratio: 0.03
|
51 |
num_epochs: 40
|
|
|
52 |
NOTE: test - Accuracy (Eval dataset and predict) for a sample of 2000: 80.70%
|
53 |
|
54 |
## Training procedure
|
|
|
36 |
|
37 |
Evaluation: https://github.com/frank-morales2020/MLxDL/blob/main/FineTunning_Testing_For_EmotionQADataset.ipynb
|
38 |
|
39 |
+
-------
|
40 |
The following hyperparameters were used during training:
|
41 |
learning_rate: 0.0002 train_batch_size: 3 eval_batch_size: 8 seed: 42 gradient_accumulation_steps: 2 total_train_batch_size: 6 optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 lr_scheduler_type: constant lr_scheduler_warmup_ratio: 0.03
|
42 |
num_epochs: 1
|
43 |
+
|
44 |
NOTE: test - Accuracy (Eval dataset and predict) for a sample of 2000: 59.45%
|
45 |
+
--------------
|
46 |
|
47 |
The following hyperparameters were used during training:
|
48 |
learning_rate: 0.0002 train_batch_size: 3 eval_batch_size: 8 seed: 42 gradient_accumulation_steps: 2 total_train_batch_size: 6 optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 lr_scheduler_type: constant lr_scheduler_warmup_ratio: 0.03
|
49 |
num_epochs: 25
|
50 |
+
|
51 |
NOTE: test - Accuracy (Eval dataset and predict) for a sample of 2000: 79.95%
|
52 |
|
53 |
The following hyperparameters were used during training:
|
54 |
learning_rate: 0.0002 train_batch_size: 3 eval_batch_size: 8 seed: 42 gradient_accumulation_steps: 2 total_train_batch_size: 6 optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 lr_scheduler_type: constant lr_scheduler_warmup_ratio: 0.03
|
55 |
num_epochs: 40
|
56 |
+
|
57 |
NOTE: test - Accuracy (Eval dataset and predict) for a sample of 2000: 80.70%
|
58 |
|
59 |
## Training procedure
|