Update model - STAGE3 Epoch 1 | Loss: 0.0000 | Acc: 9.71%

README.md

@@ -10,7 +10,7 @@ tags:
 - tiny-vlm
 - dinov2_small
 - smollm_135m
-- stage2
+- stage3
 base_model:
 - smollm_135m
 library_name: transformers
@@ -21,17 +21,23 @@ pipeline_tag: image-text-to-text
 
 **🔥 Efficient Vision-Language Model for Edge Deployment & Robotic Applications**
 
-This model is currently in training - **STAGE2 (Epoch 1)**.
+This model is currently in training - **STAGE3 (Epoch 1)**.
 
 ## 📊 Current Training Status
 
-- **Stage**: Multimodal Instruction Tuning - Following complex instructions
+- **Stage**: Robot Fleet Selection - Choosing optimal robots for tasks
 - **Epoch**: 1
-- **Last Updated**: 2026-02-20 09:13:46 UTC
+- **Last Updated**: 2026-02-20 09:19:56 UTC
 
 ### Latest Metrics
-- **instruction_loss**: 0.0000
-- **loss**: 6.4789
+- **Drone_f1**: 0.0000
+- **Humanoid_f1**: 0.0000
+- **Robot_with_Legs_f1**: 0.0000
+- **Robot_with_Wheels_f1**: 0.1770
+- **Underwater_Robot_f1**: 0.0000
+- **accuracy**: 0.0971
+- **loss**: 0.0000
+- **macro_f1**: 0.0354
 
 ## 🏗️ Model Architecture
 
@@ -50,7 +56,7 @@ EmberVLM follows a 4-stage training curriculum:
 3. ✅ **Stage 3: Robot Fleet Selection** - Task-robot matching
 4. ⏳ **Stage 4: Chain-of-Thought Reasoning** - Reasoning generation
 
-**Current Stage**: STAGE2
+**Current Stage**: STAGE3
 
 ## 💻 Usage
 
@@ -125,5 +131,5 @@ Apache 2.0
 
 ---
 
-**Note**: This is a checkpoint from stage2 training (epoch 1). 
+**Note**: This is a checkpoint from stage3 training (epoch 1). 
 The model will be updated after each epoch with improved performance.

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:30a7b733da85c9aa85f96ed70577a4a51ac3a4ee8d5c242a9502aaa2722ccc91
+oid sha256:871caca93b5cdba82011a19634d37b3ce537e11814a0d0331b1541a3fe0e4379
 size 388491879

training_info.json

@@ -1,12 +1,18 @@
 {
-  "stage": "stage2",
+  "stage": "stage3",
   "epoch": 1,
   "metrics": {
-    "loss": 6.4789011137826105,
-    "instruction_loss": 0.0
+    "loss": 0.0,
+    "accuracy": 0.0970873786407767,
+    "Drone_f1": 0.0,
+    "Underwater_Robot_f1": 0.0,
+    "Humanoid_f1": 0.0,
+    "Robot_with_Wheels_f1": 0.17699115044247787,
+    "Robot_with_Legs_f1": 0.0,
+    "macro_f1": 0.035398230088495575
   },
   "carbon_emissions_kg": 0.0,
-  "timestamp": "2026-02-20T09:13:46.607442",
+  "timestamp": "2026-02-20T09:19:56.476259",
   "vision_backbone": "dinov2_small",
   "language_backbone": "smollm_135m",
   "total_parameters": 164316833,