Ezaf 4067 | ray tune/train reports depricated fix

Data-Science/Ray-CPU/run_cpu.ipynb

@@ -75,7 +75,10 @@
     "get_job_logs = client.get_job_logs(job_id)\n",
     "get_job_info = client.get_job_info(job_id)\n",
     "async for lines in client.tail_job_logs(job_id):\n",
-    "    print(lines, end=\"\") "
+    "    print(lines, end=\"\") \n",
+    "\n",
+    "# Disconnect from the Ray cluster\n",
+    "ray.shutdown()"
    ]
   },
   {

Data-Science/Ray-GPU/core.py

@@ -0,0 +1,24 @@
+import tensorflow as tf
+
+def main():
+    for iter in range(1, 5):
+        # Confirm that TensorFlow is using the GPU.
+        print("Num GPUs Available: ", len(tf.config.list_physical_devices('GPU')))
+
+        # If GPU is available and accessible for TF, resulting list should not be empty
+        tf.config.list_physical_devices('GPU')
+
+        # Enabling logging device placement to find out which devices TF operations and tensors are assigned to
+        tf.debugging.set_log_device_placement(True)
+
+        # Create some tensors
+        a = tf.constant([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]])
+        b = tf.constant([[1.0, 2.0], [3.0, 4.0], [5.0, 6.0]])
+        c = tf.matmul(a, b)
+
+        # The cell above should print an indication that the MatMul op was executed on GPU:0
+        print("RAY FRAMEWORK: GPU RESULT", c)
+
+if __name__ == "__main__":
+    main()
+

Data-Science/Ray-GPU/readme copy.md

@@ -0,0 +1,26 @@
+# RAY Fibonacci Example
+
+## Prerequisites
+- Python 3.7 or higher
+- The Current Release RAY framework installed (version 2.7.0)
+- Additional dependencies listed in requirements.txt
+
+## Validation Steps
+1. **Setup Environment**: Set up the required environment variables. Please make sure you have updated right head node hostport Ip and port.
+    ray_head_ip = "10.224.226.40" or 
+    ray_head_ip = "kuberay-head-svc.kuberay.svc.cluster.local"
+    ray_head_port = 8265
+
+2. **Validation Script**: Run the validation script.
+With defined custom NAS/network storage path
+To complete the tutorial follow simple steps below:
+1. Login to you EzAF cluster.
+2. Create a new notebook server using the `jupyter-data-science` image or any desired image, choice the kernel type "RAY".
+3. Clone the repository locally.
+4. Launch the `run.ipynb` notebook file.
+5. keep the `core.py` along with the above nortebook (same location). 
+5. This example will executes a log on top of RAY cluster.
+
+## Issues and Contributions
+
+If you encounter any issues during the validation process or would like to contribute to the validation scripts, please open an issue or submit a pull request to EzUA Dev Team.

Data-Science/Ray-GPU/requirements.txt

@@ -0,0 +1,7 @@
+ray[client]==2.7.0
+ray[default]==2.7.0
+time
+warnings
+tensorrt
+tensorflow
+nvidia-tensorrt==99.0.0