suryadev1 commited on
Commit
d58e19e
Β·
verified Β·
1 Parent(s): 2e36208

Update Astra Project Setup Instructions.md

Browse files

added finetune task description in setup description

Files changed (1) hide show
  1. Astra Project Setup Instructions.md +10 -3
Astra Project Setup Instructions.md CHANGED
@@ -122,10 +122,17 @@ Replace `<finetune_task>` with the actual fine-tuning task value.
122
  - Description: Specifies which fine-tuning strategy was applied to the model.
123
  - Options:
124
  - **ASTRA-FT-HGR** β†’ Fine-tuned with 10% data from schools that have a **High Graduation Rate (HGR)**.
125
- - **ASTRA-FT-Skills** β†’ Fine-tuned with 10% of initial problems from both **HGR + LGR schools**, with **Prior Skills encoded** using **Bayesian Knowledge Tracing (BKT)**.
126
- - **ASTRA-FT-Temporal** β†’ Fine-tuned with 10% of initial problems from both **HGR + LGR schools**, with **temporal features** measuring student engagement in MATHia.
127
- - **ASTRA-FT-Unified** β†’ Fine-tuned with 10% of initial problems from both **HGR + LGR schools**, combining **Prior Skills (BKT) + temporal features**.
128
 
 
 
 
 
 
 
 
129
  **`-test_dataset_path`**
130
  - Description: Path to the test dataset file that you want to use for evaluation.
131
  - Example: `"../../../../fileHandler/selected_rows.txt"`
 
122
  - Description: Specifies which fine-tuning strategy was applied to the model.
123
  - Options:
124
  - **ASTRA-FT-HGR** β†’ Fine-tuned with 10% data from schools that have a **High Graduation Rate (HGR)**.
125
+ - **ASTRA-FT-FIRST10-WSKILLS**
126
+ - Checkpoint: `first10/bert_fine_tuned.model.first10%.wskills.ep24`
127
+ - Description: Fine-tuned with 10% of initial problems from both **HGR + LGR schools**, with **Prior Skills encoded** using **Bayesian Knowledge Tracing (BKT)**.
128
 
129
+ - **ASTRA-FT-FIRST10-WTIME**
130
+ - Checkpoint: `first10/bert_fine_tuned.model.first10%.wfaopttime.wttime.wttopttime.wttnoopttime.ep23`
131
+ - Description: Fine-tuned with 10% of initial problems from both **HGR + LGR schools**, using **temporal features** measuring student engagement in MATHia.
132
+
133
+ - **ASTRA-FT-FIRST10-WSKILLS_WTIME**
134
+ - Checkpoint: `first10/bert_fine_tuned.model.first10%.wskills.wfaopttime.wttime.wttopttime.wttnoopttime.ep40`
135
+ - Description: Fine-tuned with 10% of initial problems from both **HGR + LGR schools**, combining **Prior Skills (BKT) + temporal features**.
136
  **`-test_dataset_path`**
137
  - Description: Path to the test dataset file that you want to use for evaluation.
138
  - Example: `"../../../../fileHandler/selected_rows.txt"`