PraisonAI 0.0.59rc5__tar.gz → 0.0.59rc7__tar.gz

{praisonai-0.0.59rc5 → praisonai-0.0.59rc7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: PraisonAI
-Version: 0.0.59rc5
+Version: 0.0.59rc7
 Summary: PraisonAI application combines AutoGen and CrewAI or similar frameworks into a low-code solution for building and managing multi-agent LLM systems, focusing on simplicity, customization, and efficient human-agent collaboration.
 Author: Mervin Praison
 Requires-Python: >=3.10,<3.13

{praisonai-0.0.59rc5 → praisonai-0.0.59rc7}/praisonai/cli.py

@@ -13,6 +13,7 @@ from .auto import AutoGenerator
 from .agents_generator import AgentsGenerator
 from .inbuilt_tools import *
 import shutil
+import subprocess
 import logging
 logging.basicConfig(level=os.environ.get('LOGLEVEL', 'INFO'), format='%(asctime)s - %(levelname)s - %(message)s')
 
@@ -27,6 +28,32 @@ try:
     GRADIO_AVAILABLE = True
 except ImportError:
     GRADIO_AVAILABLE = False
+    
+def stream_subprocess(command):
+    """
+    Execute a subprocess command and stream the output to the terminal in real-time.
+
+    Args:
+        command (list): A list containing the command and its arguments.
+    """
+    process = subprocess.Popen(
+        command,
+        stdout=subprocess.PIPE,
+        stderr=subprocess.STDOUT,
+        text=True,
+        bufsize=1,
+        universal_newlines=True
+    )
+
+    for line in iter(process.stdout.readline, ''):
+        print(line, end='')
+        sys.stdout.flush()  # Ensure output is flushed immediately
+
+    process.stdout.close()
+    return_code = process.wait()
+
+    if return_code != 0:
+        raise subprocess.CalledProcessError(return_code, command)
 
 class PraisonAI:
     def __init__(self, agent_file="agents.yaml", framework="", auto=False, init=False, agent_yaml=None):
@@ -100,12 +127,40 @@ class PraisonAI:
             return
         
         if args.agent_file == 'train':
+            package_root = os.path.dirname(os.path.abspath(__file__))
+            config_yaml_source = os.path.join(package_root, 'setup', 'config.yaml')
+            config_yaml_destination = os.path.join(os.getcwd(), 'config.yaml')
+
+            if not os.path.exists(config_yaml_destination):
+                try:
+                    shutil.copyfile(config_yaml_source, config_yaml_destination)
+                    print("config.yaml copied to the current directory.")
+                except FileExistsError:
+                    print("config.yaml already exists in the current directory. Skipping copy.")
+            else:
+                print("config.yaml already exists in the current directory. Skipping copy.")
+
             if 'init' in sys.argv:
                 from praisonai.setup.setup_conda_env import main as setup_conda_main
                 setup_conda_main()
-            from .train import main as train_main
-            train_args = sys.argv[2:]  # Get all arguments after 'train' 
-            train_main()
+                print("All packages installed")
+                return
+
+            try:
+                result = subprocess.check_output(['conda', 'env', 'list'])
+                if 'unsloth_env' in result.decode('utf-8'):
+                    print("Conda environment 'unsloth_env' found.")
+                else:
+                    raise subprocess.CalledProcessError(1, 'grep')
+            except subprocess.CalledProcessError:
+                print("Conda environment 'unsloth_env' not found. Setting it up...")
+                from praisonai.setup.setup_conda_env import main as setup_conda_main
+                setup_conda_main()
+                print("All packages installed.")
+
+            train_args = sys.argv[2:]  # Get all arguments after 'train'
+            train_script_path = os.path.join(package_root, 'train.py')
+            stream_subprocess(['conda', 'run', '--name', 'unsloth_env', 'python', train_script_path, 'train'])
             return
         
         invocation_cmd = "praisonai"

{praisonai-0.0.59rc5 → praisonai-0.0.59rc7}/praisonai/deploy.py

@@ -56,7 +56,7 @@ class CloudDeployer:
             file.write("FROM python:3.11-slim\n")
             file.write("WORKDIR /app\n")
             file.write("COPY . .\n")
-            file.write("RUN pip install flask praisonai==0.0.59rc5 gunicorn markdown\n")
+            file.write("RUN pip install flask praisonai==0.0.59rc7 gunicorn markdown\n")
             file.write("EXPOSE 8080\n")
             file.write('CMD ["gunicorn", "-b", "0.0.0.0:8080", "api:app"]\n')
             

praisonai-0.0.59rc7/praisonai/setup/config.yaml

@@ -0,0 +1,58 @@
+model_name: "unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit"
+hf_model_name: "mervinpraison/llama-3.1-tamilan-8B-test"
+max_seq_length: 2048
+load_in_4bit: true
+lora_r: 16
+lora_target_modules: 
+  - "q_proj"
+  - "k_proj"
+  - "v_proj"
+  - "o_proj"
+  - "gate_proj"
+  - "up_proj"
+  - "down_proj"
+lora_alpha: 16
+lora_dropout: 0
+lora_bias: "none"
+use_gradient_checkpointing: "unsloth"
+random_state: 3407
+use_rslora: false
+loftq_config: null
+
+dataset:
+  - name: "yahma/alpaca-cleaned"
+    split_type: "train"
+    processing_func: "format_prompts"
+    rename:
+      input: "input"
+      output: "output"
+      instruction: "instruction"
+    filter_data: false
+    filter_column_value: "id"
+    filter_value: "alpaca"
+    num_samples: 20000
+
+dataset_text_field: "text"
+dataset_num_proc: 2
+packing: false
+
+per_device_train_batch_size: 2
+gradient_accumulation_steps: 2
+warmup_steps: 5
+num_train_epochs: 1
+max_steps: 10
+learning_rate: 2.0e-4
+logging_steps: 1
+optim: "adamw_8bit"
+weight_decay: 0.01
+lr_scheduler_type: "linear"
+seed: 3407
+output_dir: "outputs"
+
+quantization_method: 
+  - "q4_k_m"
+  - "q8_0"
+  - "q5_k_m"
+
+ollama_model: "llama3.1-tamilan-test"
+model_parameters: "8b"

{praisonai-0.0.59rc5 → praisonai-0.0.59rc7}/praisonai/setup/setup_conda_env.sh

@@ -35,8 +35,16 @@ fi
 # Create and activate the Conda environment
 ENV_NAME="unsloth_env"
 if conda info --envs | grep -q $ENV_NAME; then
-    echo "Environment $ENV_NAME already exists. Activating..."
-    conda activate $ENV_NAME
+    echo "Environment $ENV_NAME already exists. Recreating..."
+    conda env remove -y -n $ENV_NAME  # Remove existing environment
+    if [[ "$OSTYPE" == "darwin"* ]]; then
+        # macOS (both Intel and M1/M2)
+        conda create --name $ENV_NAME python=3.10 pytorch=2.3.0 -c pytorch -y
+    elif [[ "$OSTYPE" == "linux-gnu"* ]]; then
+        # Linux
+        conda create --name $ENV_NAME python=3.10 pytorch=2.3.0 cudatoolkit=11.8 -c pytorch -c nvidia -y
+    fi
+    # conda activate $ENV_NAME
 else
     echo "Creating new environment $ENV_NAME..."
     if [[ "$OSTYPE" == "darwin"* ]]; then
@@ -45,16 +53,20 @@ else
     elif [[ "$OSTYPE" == "linux-gnu"* ]]; then
         # Linux
         conda create --name $ENV_NAME python=3.10 pytorch=2.3.0 cudatoolkit=11.8 -c pytorch -c nvidia -y
-        conda activate $ENV_NAME
-        pip install xformers==0.0.26.post1
     fi
+    # conda activate $ENV_NAME
 fi
 
-source $HOME/miniconda/bin/activate $ENV_NAME
+# source $HOME/miniconda/bin/activate $ENV_NAME
+
+# Get full path of pip
+PIP_FULL_PATH=$(conda run -n $ENV_NAME which pip)
 
-# Install other packages
-pip install --upgrade pip
-pip install "unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git@4e570be9ae4ced8cdc64e498125708e34942befc"
-pip install --no-deps "trl<0.9.0" peft accelerate bitsandbytes
+# Install other packages within the activated environment
+# Use PIP_FULL_PATH to run pip commands
+$PIP_FULL_PATH install --upgrade pip 
+$PIP_FULL_PATH install "xformers==0.0.26.post1"
+$PIP_FULL_PATH install "unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git@4e570be9ae4ced8cdc64e498125708e34942befc"
+$PIP_FULL_PATH install --no-deps "trl<0.9.0" peft accelerate bitsandbytes 
 
-echo "Setup completed successfully!"
+echo "Setup completed successfully!"

{praisonai-0.0.59rc5 → praisonai-0.0.59rc7}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "PraisonAI"
-version = "0.0.59rc5"
+version = "0.0.59rc7"
 description = "PraisonAI application combines AutoGen and CrewAI or similar frameworks into a low-code solution for building and managing multi-agent LLM systems, focusing on simplicity, customization, and efficient human-agent collaboration."
 authors = ["Mervin Praison"]
 license = ""