jonghyunlee commited on
Commit
26004d1
·
verified ·
1 Parent(s): 9f2bf97

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -6
README.md CHANGED
@@ -5,9 +5,7 @@ tags:
5
  ---
6
 
7
 
8
- ```markdown
9
  # 🧬 MoLLaMA-Small
10
-
11
  MoLLaMA-Small is a lightweight LLaMA-based causal language model (57.2M parameters) trained from scratch to generate valid chemical molecules using SMILES strings.
12
 
13
  This model uses DeepChem's `SmilesTokenizer` and was trained on a combined dataset of ZINC15 and MuMOInstruct. It is designed for unconditional molecule generation.
@@ -31,11 +29,11 @@ A custom, scaled-down LLaMA architecture was used to optimize for chemical langu
31
  * **Max Position Embeddings**: 1024
32
 
33
  ## 🚀 How to Use
34
-
35
  You can easily load this model using the standard `transformers` library. The model generates SMILES strings by prompting it with the `[bos]` (Beginning of Sequence) token.
36
 
37
  ### Prerequisites
38
  Make sure you have the required libraries installed:
 
39
  ```bash
40
  pip install transformers torch deepchem
41
 
@@ -86,6 +84,4 @@ print(f"Generated SMILES: {generated_smiles}")
86
  * **Batch Size**: 512 (with gradient accumulation steps of 4)
87
  * **Learning Rate**: 1e-4 (Cosine scheduler, 10% Warmup)
88
  * **Precision**: bf16 (Mixed Precision)
89
- * **Early Stopping Patience**: 5 epochs
90
-
91
- ```
 
5
  ---
6
 
7
 
 
8
  # 🧬 MoLLaMA-Small
 
9
  MoLLaMA-Small is a lightweight LLaMA-based causal language model (57.2M parameters) trained from scratch to generate valid chemical molecules using SMILES strings.
10
 
11
  This model uses DeepChem's `SmilesTokenizer` and was trained on a combined dataset of ZINC15 and MuMOInstruct. It is designed for unconditional molecule generation.
 
29
  * **Max Position Embeddings**: 1024
30
 
31
  ## 🚀 How to Use
 
32
  You can easily load this model using the standard `transformers` library. The model generates SMILES strings by prompting it with the `[bos]` (Beginning of Sequence) token.
33
 
34
  ### Prerequisites
35
  Make sure you have the required libraries installed:
36
+
37
  ```bash
38
  pip install transformers torch deepchem
39
 
 
84
  * **Batch Size**: 512 (with gradient accumulation steps of 4)
85
  * **Learning Rate**: 1e-4 (Cosine scheduler, 10% Warmup)
86
  * **Precision**: bf16 (Mixed Precision)
87
+ * **Early Stopping Patience**: 5 epochs