Question Answering
Safetensors
phi3
biology
medical
custom_code
BossRui commited on
Commit
f22f582
1 Parent(s): 2a1e16f

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +22 -19
README.md CHANGED
@@ -76,12 +76,23 @@ Covering 12 Major Languages including English, Chinese, French, Hindi, Spanish,
76
  * **[2024.10.15]** ApolloMoE repo is published!🎉
77
 
78
 
 
 
 
 
 
 
 
 
 
 
 
79
  ## Architecture
80
 
81
  <details>
82
  <summary>Click to view the MoE routing image</summary>
83
 
84
- ![ApolloMoE](/assets/hybrid_routing.png)
85
 
86
  </details>
87
 
@@ -188,17 +199,17 @@ Covering 12 Major Languages including English, Chinese, French, Hindi, Spanish,
188
  <details><summary>Click to expand</summary>
189
 
190
 
191
- We take Gemma-2b as example
192
  1. Download Dataset for project:
193
 
194
  ```
195
- bash 0.download_data.sh
196
  ```
197
 
198
- 2. Prepare test and dev for specific model:
199
 
200
 
201
- - Create test data for with special token, you can use ./util/check.ipynb to check models' special tokens
202
 
203
  ```
204
  bash 1.data_process_test&dev.sh
@@ -206,23 +217,21 @@ Covering 12 Major Languages including English, Chinese, French, Hindi, Spanish,
206
 
207
  3. Prepare train data for specific model (Create tokenized data in advance):
208
 
209
-
210
- - You can adjust data Training order and Training Epoch in this step
211
 
 
 
212
  ```
213
  bash 2.data_process_train.sh
214
  ```
215
-
216
  4. Train the model
217
 
218
-
219
- - If you want to train in Multi Nodes please refer to ./scripts/multi_node_train_*.sh
220
-
221
-
222
 
223
 
224
  ```
225
- bash 3.single_node_train_gemma.sh
226
  ```
227
 
228
 
@@ -232,12 +241,6 @@ Covering 12 Major Languages including English, Chinese, French, Hindi, Spanish,
232
  bash 4.eval.sh
233
  ```
234
 
235
- 6. Evaluate your model: Play with your ckpts in bash
236
-
237
- ```
238
- python ./src/evaluate/cli_demo.py --model_name='./ckpts/your/path/tfmr'
239
- ```
240
-
241
  </details>
242
 
243
 
 
76
  * **[2024.10.15]** ApolloMoE repo is published!🎉
77
 
78
 
79
+ ## Languages Coverage
80
+ 12 Major Languages and 38 Minor Languages
81
+
82
+ <details>
83
+ <summary>Click to view the Languages Coverage</summary>
84
+
85
+ ![ApolloMoE](assets/languages.png)
86
+
87
+ </details>
88
+
89
+
90
  ## Architecture
91
 
92
  <details>
93
  <summary>Click to view the MoE routing image</summary>
94
 
95
+ ![ApolloMoE](assets/hybrid_routing.png)
96
 
97
  </details>
98
 
 
199
  <details><summary>Click to expand</summary>
200
 
201
 
202
+ We take Apollo2-7B or Apollo-MoE-0.5B as example
203
  1. Download Dataset for project:
204
 
205
  ```
206
+ bash 0.download_data.sh 
207
  ```
208
 
209
+ 2. Prepare test and dev data for specific model:
210
 
211
 
212
+ - Create test data for with special token
213
 
214
  ```
215
  bash 1.data_process_test&dev.sh
 
217
 
218
  3. Prepare train data for specific model (Create tokenized data in advance):
219
 
 
 
220
 
221
+ - You can adjust data Training order and Training Epoch in this step
222
+
223
  ```
224
  bash 2.data_process_train.sh
225
  ```
226
+
227
  4. Train the model
228
 
229
+
230
+ - If you want to train in Multi Nodes please refer to ./src/sft/training_config/zero_multi.yaml
 
 
231
 
232
 
233
  ```
234
+ bash 3.single_node_train.sh
235
  ```
236
 
237
 
 
241
  bash 4.eval.sh
242
  ```
243
 
 
 
 
 
 
 
244
  </details>
245
 
246