Update README.md

README.md

@@ -3,18 +3,27 @@ license: apache-2.0
 ---
 **Base Model**: BLIP2-t5 pretrained version
 
-**Finetune data**: LLAVA 150k (sample one pair of instruction-answer if multi-round conversations)
+**Finetune data**: 
+* LLAVA 150k (sample one pair of instruction-answer if multi-round conversations)
+* MiniGPT4 3500 pairs
 
 **Hyper-parameters**: 
 
-v0:
-* lr = 2e-5 --> 0.0 with cosine lr scheduler
-* gbs = 32
-* image size = 480
-* weight decay = 0.05
+* BLIP2-flant5-xl + LLAVA (initial commits)
+  * v0:
+  * lr = 2e-5 --> 0.0 with cosine lr scheduler
+  * gbs = 32
+  * image size = 480
+  * weight decay = 0.05
 
-v1 (same as LLAVA):
-* lr = 2e-5
-* gbs = 32
-* image size = 480
-* weight decay = 0.0
+  * v1 (same as LLAVA):
+  * lr = 2e-5
+  * gbs = 32
+  * image size = 480
+  * weight decay = 0.0
+
+* BLIP2-flant5-xl + MiniGPT4
+  * lr = 2e-5
+  * gbs = 32
+  * image size = 480
+  * weight decay = 0.0