Update README.md
Browse files
README.md
CHANGED
|
@@ -31,8 +31,12 @@ The Digital Navigator model is designed to assist users by generating natural la
|
|
| 31 |
|
| 32 |
### Direct Use
|
| 33 |
|
| 34 |
-
|
| 35 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 36 |
### Out-of-Scope Use
|
| 37 |
|
| 38 |
- Generating harmful, biased, or misleading content
|
|
@@ -66,7 +70,7 @@ The model was trained using standard procedures for fine-tuning GPT-2 models.
|
|
| 66 |
|
| 67 |
#### Training Hyperparameters
|
| 68 |
|
| 69 |
-
- **Training regime:**
|
| 70 |
- Learning Rate: 5e-5
|
| 71 |
- Batch Size: 16
|
| 72 |
- Epochs: 30
|
|
@@ -97,7 +101,7 @@ Due to the high complexity and small variety of the data:
|
|
| 97 |
|
| 98 |
- **Hardware Type:** NVIDIA T4
|
| 99 |
- **Hours used:** 18 minutes and 47 seconds
|
| 100 |
-
- **Cloud Provider:** Google
|
| 101 |
|
| 102 |
|
| 103 |
|
|
|
|
| 31 |
|
| 32 |
### Direct Use
|
| 33 |
|
| 34 |
+
The Digital Navigator model can be directly used for generating conversational responses in English. It is intended for use in chatbots, virtual assistants, and other applications requiring natural language understanding and generation.
|
| 35 |
|
| 36 |
+
- Conversational AI
|
| 37 |
+
- Customer Support
|
| 38 |
+
- Virtual Assistance
|
| 39 |
+
|
| 40 |
### Out-of-Scope Use
|
| 41 |
|
| 42 |
- Generating harmful, biased, or misleading content
|
|
|
|
| 70 |
|
| 71 |
#### Training Hyperparameters
|
| 72 |
|
| 73 |
+
- **Training regime:** fp32
|
| 74 |
- Learning Rate: 5e-5
|
| 75 |
- Batch Size: 16
|
| 76 |
- Epochs: 30
|
|
|
|
| 101 |
|
| 102 |
- **Hardware Type:** NVIDIA T4
|
| 103 |
- **Hours used:** 18 minutes and 47 seconds
|
| 104 |
+
- **Cloud Provider:** Google Colab
|
| 105 |
|
| 106 |
|
| 107 |
|