unknown
commited on
Commit
·
1930271
1
Parent(s):
c665586
add files
Browse files
README.md
CHANGED
@@ -140,13 +140,13 @@ We only fine-tuned CodeT5+ across three tasks, and compare it with accuracy of A
|
|
140 |
| **Decrease** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** |
|
141 |
|
142 |
- LLVM
|
143 |
-
|
|
144 |
-
|------------------
|
145 |
-
|
|
146 |
-
|
|
147 |
-
| -w GPU and MPU
|
148 |
-
| -w/o GPU and MPU |
|
149 |
-
|
|
150 |
|
151 |
- `New_Targets/Itr_Expansion/*`: **Take data of RI5CY in LLVM as test set, split train/valid set in the ratio of 85%:15% of CPU targets excluding RISC-V and including RISC-V**
|
152 |
|
@@ -169,9 +169,9 @@ We only fine-tuned CodeT5+ across three tasks, and compare it with accuracy of A
|
|
169 |
|
170 |
We only fine-tuned CodeT5+ across three tasks, and compare it with accuracy of RI5CY between excluding and including RISC-V in dataset.
|
171 |
|
172 |
-
|
|
173 |
-
|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:----------:
|
174 |
-
|
|
175 |
-
| -w RISC-V
|
176 |
-
| -w/o RISC-V |
|
177 |
-
|
|
|
|
140 |
| **Decrease** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** |
|
141 |
|
142 |
- LLVM
|
143 |
+
| | Stmt. Comp. | Stmt. Comp. | Stmt. Comp. | Stmt. Comp. | Next. Sugg. | Next. Sugg. | Next. Sugg. | Next. Sugg. | Code. Gen. | Code. Gen. | Code. Gen. | Code. Gen. |
|
144 |
+
|------------------ |:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:----------: |:----------: |:----------: |:----------: |
|
145 |
+
| | ARC(MPU) | ARC(MPU) | NVPTX(GPU) | NVPTX(GPU) | ARC(MPU) | ARC(MPU) | NVPTX(GPU) | NVPTX(GPU) | ARC(MPU) | ARC(MPU) | NVPTX(GPU) | NVPTX(GPU) |
|
146 |
+
| Dataset | EM | ED | EM | ED | EM | ED | EM | ED | BLEU4 | ED | BLEU4 | ED |
|
147 |
+
| -w GPU and MPU | 71.34 | 85.98 | 64.45 | 81.53 | 58.68 | 74.57 | 47.81 | 65.5 | 55.38 | 74.41 | 44.33 | 66.36 |
|
148 |
+
| -w/o GPU and MPU | xxx | xxx | xxx | xxx | xxx | xxx | xxx | xxx | xxx | xxx | xxx | xxx |
|
149 |
+
| **Decrease** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** |
|
150 |
|
151 |
- `New_Targets/Itr_Expansion/*`: **Take data of RI5CY in LLVM as test set, split train/valid set in the ratio of 85%:15% of CPU targets excluding RISC-V and including RISC-V**
|
152 |
|
|
|
169 |
|
170 |
We only fine-tuned CodeT5+ across three tasks, and compare it with accuracy of RI5CY between excluding and including RISC-V in dataset.
|
171 |
|
172 |
+
| | Stmt. Comp. | Stmt. Comp. | Next. Sugg. | Next. Sugg. | Code. Gen. | Code. Gen. |
|
173 |
+
|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:----------: |:----------: |
|
174 |
+
| Dataset | EM | ED | EM | ED | BLEU4 | ED |
|
175 |
+
| -w RISC-V | xxx | xxx | xxx | xxx | xxx | xxx |
|
176 |
+
| -w/o RISC-V | xxx | xxx | xxx | xxx | xxx | xxx |
|
177 |
+
| **Diff** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** | **xxx** |
|