🔧 v3: Fix SFTConfig for trl>=0.8

app.py

@@ -1,5 +1,5 @@
 """
-🧬 HIVEMIND GPU WORKER v2
+🧬 HIVEMIND GPU WORKER v3
 Training LoRA adapters za Hivemind agente
 """
 import gradio as gr
@@ -14,7 +14,7 @@ def train_agent(agent_name: str, skill: str, epochs: int = 2):
         import torch
         from transformers import AutoModelForCausalLM, AutoTokenizer, TrainingArguments
         from peft import LoraConfig, get_peft_model
-        from trl import SFTTrainer
+        from trl import SFTTrainer, SFTConfig
         from datasets import Dataset
         from huggingface_hub import HfApi, login
         from datetime import datetime
@@ -22,14 +22,14 @@ def train_agent(agent_name: str, skill: str, epochs: int = 2):
         log = [f"🚀 Starting: {agent_name} - {skill}"]
         
         if not HF_TOKEN:
-            return "❌ HF_TOKEN not set in Space secrets! Go to Settings -> Repository secrets"
+            return "❌ HF_TOKEN not set in Space secrets!"
         
         login(token=HF_TOKEN)
         api = HfApi(token=HF_TOKEN)
         
         task_id = f"{agent_name[:8].lower().replace(' ','')}-{datetime.now().strftime('%m%d%H%M%S')}"
         
-        # Load model (CPU friendly small model)
+        # Load model
         log.append("📦 Loading TinyLlama...")
         model = AutoModelForCausalLM.from_pretrained(
             "TinyLlama/TinyLlama-1.1B-Chat-v1.0", 
@@ -51,34 +51,36 @@ def train_agent(agent_name: str, skill: str, epochs: int = 2):
         trainable = sum(p.numel() for p in model.parameters() if p.requires_grad)
         log.append(f"   Trainable params: {trainable:,}")
         
-        # Dataset
-        data = [
-            {"text": f"<|user|>\nWhat is {skill}?</s>\n<|assistant|>\n{skill} is a fundamental technique in machine learning and AI.</s>"},
-            {"text": f"<|user|>\nExplain {skill}</s>\n<|assistant|>\n{skill} helps optimize model performance and efficiency.</s>"},
-            {"text": f"<|user|>\nHow to implement {skill}?</s>\n<|assistant|>\nTo implement {skill}, apply proper techniques and best practices.</s>"},
-            {"text": f"<|user|>\nWhy is {skill} important?</s>\n<|assistant|>\n{skill} is crucial for building effective AI systems.</s>"},
+        # Dataset - format as list of strings
+        texts = [
+            f"<|user|>\nWhat is {skill}?</s>\n<|assistant|>\n{skill} is a fundamental technique in machine learning.</s>",
+            f"<|user|>\nExplain {skill}</s>\n<|assistant|>\n{skill} helps optimize model performance.</s>",
+            f"<|user|>\nHow to implement {skill}?</s>\n<|assistant|>\nTo implement {skill}, apply proper techniques.</s>",
+            f"<|user|>\nWhy is {skill} important?</s>\n<|assistant|>\n{skill} is crucial for effective AI systems.</s>",
         ]
-        dataset = Dataset.from_list(data)
+        dataset = Dataset.from_dict({"text": texts})
         log.append(f"📊 Dataset: {len(dataset)} examples")
         
-        # Train (minimal for CPU)
+        # Train with SFTConfig
         log.append(f"🏋️ Training {epochs} epoch(s)...")
-        trainer = SFTTrainer(
-            model=model, 
-            train_dataset=dataset, 
+        
+        training_args = SFTConfig(
+            output_dir="./out",
+            num_train_epochs=epochs,
+            per_device_train_batch_size=1,
+            learning_rate=2e-4,
+            save_strategy="no",
+            report_to="none",
+            logging_steps=1,
+            max_seq_length=128,
             dataset_text_field="text",
-            max_seq_length=128, 
+        )
+        
+        trainer = SFTTrainer(
+            model=model,
+            train_dataset=dataset,
             tokenizer=tokenizer,
-            args=TrainingArguments(
-                output_dir="./out", 
-                num_train_epochs=epochs, 
-                per_device_train_batch_size=1,
-                learning_rate=2e-4, 
-                save_strategy="no", 
-                report_to="none", 
-                fp16=False,
-                logging_steps=1,
-            )
+            args=training_args,
         )
         result = trainer.train()
         log.append(f"✅ Training complete! Loss: {result.training_loss:.4f}")
@@ -91,7 +93,7 @@ def train_agent(agent_name: str, skill: str, epochs: int = 2):
         log.append(f"📤 Uploading to {repo_id}...")
         
         api.create_repo(repo_id=repo_id, exist_ok=True, private=False)
-        api.upload_folder(folder_path="./lora", repo_id=repo_id, commit_message=f"🤖 {agent_name}: {skill}")
+        api.upload_folder(folder_path="./lora", repo_id=repo_id)
         log.append(f"✅ SUCCESS: https://huggingface.co/{repo_id}")
         
         return "\n".join(log)
@@ -101,7 +103,7 @@ def train_agent(agent_name: str, skill: str, epochs: int = 2):
 
 
 with gr.Blocks(title="Hivemind GPU Worker") as demo:
-    gr.Markdown("# 🧬 Hivemind GPU Worker v2\nTraining LoRA adapters for AI agents")
+    gr.Markdown("# 🧬 Hivemind GPU Worker v3\nTraining LoRA adapters for AI agents")
     
     with gr.Row():
         agent_input = gr.Textbox(label="Agent Name", value="TestAgent")