Update README.md
Browse files
README.md
CHANGED
|
@@ -40,10 +40,12 @@ Ruby Code Generator is a versatile tool crafted to streamline the interaction be
|
|
| 40 |
- Load the bigcode/the-stack-smol dataset using the Hugging Face Datasets library.
|
| 41 |
- Filter for the specified subset (data/ruby) and split (train).
|
| 42 |
- Load the bigcode/starcoder2-3b model from the Hugging Face Hub with '4-bit' quantization.
|
|
|
|
| 43 |
**2. Data Preprocessing:**
|
| 44 |
- Tokenize the code text using the appropriate tokenizer for the chosen model.
|
| 45 |
- Apply necessary cleaning or normalization (e.g., removing comments, handling indentation).
|
| 46 |
- Create input examples suitable for the model's architecture (e.g., with masked language modeling objectives).
|
|
|
|
| 47 |
**3. Configure Training:**
|
| 48 |
- Initialize a Trainer object (likely from a library like Transformers).
|
| 49 |
- Set training arguments based on the provided args:
|
|
|
|
| 40 |
- Load the bigcode/the-stack-smol dataset using the Hugging Face Datasets library.
|
| 41 |
- Filter for the specified subset (data/ruby) and split (train).
|
| 42 |
- Load the bigcode/starcoder2-3b model from the Hugging Face Hub with '4-bit' quantization.
|
| 43 |
+
|
| 44 |
**2. Data Preprocessing:**
|
| 45 |
- Tokenize the code text using the appropriate tokenizer for the chosen model.
|
| 46 |
- Apply necessary cleaning or normalization (e.g., removing comments, handling indentation).
|
| 47 |
- Create input examples suitable for the model's architecture (e.g., with masked language modeling objectives).
|
| 48 |
+
|
| 49 |
**3. Configure Training:**
|
| 50 |
- Initialize a Trainer object (likely from a library like Transformers).
|
| 51 |
- Set training arguments based on the provided args:
|