Merge pull request #2440 from krzeszew/krzeszew/numba-dpex-api-update-genetic-sample

Update Genetic Algorithm code sample according to new numba-dpex API syntax

Author: jimmytwei

AI-and-Analytics/Features-and-Functionality/IntelPython_GPU_numba-dpex_Genetic_Algorithm/IntelPython_GPU_numba-dpex_Genetic_Algorithm.ipynb

@@ -348,7 +348,7 @@
     "\n",
     "The only par that differs form the standard implementation is the evaluation function.\n",
     "\n",
-    "The most important part is to specify the global index of the computation. This is the current index of the computed chromosomes. This serves as a loop function across all chromosomes."
+    "The most important part is to specify the index of the computation. This is the current index of the computed chromosomes. This serves as a loop function across all chromosomes."
    ]
   },
   {
@@ -365,10 +365,11 @@
    "outputs": [],
    "source": [
     "import numba_dpex\n",
+    "from numba_dpex import kernel_api\n",
     "\n",
     "@numba_dpex.kernel\n",
-    "def eval_genomes_sycl_kernel(chromosomes, fitnesses, chrom_length):\n",
-    "  pos = numba_dpex.get_global_id(0)\n",
+    "def eval_genomes_sycl_kernel(item: kernel_api.Item, chromosomes, fitnesses, chrom_length):\n",
+    "  pos = item.get_id(0)\n",
     "  num_loops = 3000\n",
     "  for i in range(num_loops):\n",
     "    fitnesses[pos] += chromosomes[pos*chrom_length + 1]\n",
@@ -409,8 +410,9 @@
     "  chromosomes_flat = chromosomes.flatten()\n",
     "  chromosomes_flat_dpctl = dpnp.asarray(chromosomes_flat, device=\"gpu\")\n",
     "  fitnesses_dpctl = dpnp.asarray(fitnesses, device=\"gpu\")\n",
-    "\n",
-    "  eval_genomes_sycl_kernel[numba_dpex.Range(pop_size)](chromosomes_flat_dpctl, fitnesses_dpctl, chrom_size)\n",
+    "  \n",
+    "  exec_range = kernel_api.Range(pop_size)\n",
+    "  numba_dpex.call_kernel(eval_genomes_sycl_kernel, exec_range, chromosomes_flat_dpctl, fitnesses_dpctl, chrom_size)\n",
     "  fitnesses = dpnp.asnumpy(fitnesses_dpctl)\n",
     "  chromosomes = next_generation(chromosomes, fitnesses)\n",
     "  fitnesses = np.zeros(pop_size, dtype=np.float32)\n",
@@ -544,7 +546,7 @@
     "\n",
     "The evaluate created generation we are calculating the full distance of the given path (chromosome). In this example, the lower the fitness value is, the better the chromosome. That's different from the general GA that we implemented.\n",
     "\n",
-    "As in this example we are also using numba-dpex, we are using a global index like before."
+    "As in this example we are also using numba-dpex, we are using an index like before."
    ]
   },
   {
@@ -554,8 +556,8 @@
    "outputs": [],
    "source": [
     "@numba_dpex.kernel\n",
-    "def eval_genomes_plain_TSP_SYCL(chromosomes, fitnesses, distances, pop_length):\n",
-    "  pos = numba_dpex.get_global_id(0)\n",
+    "def eval_genomes_plain_TSP_SYCL(item: kernel_api.Item, chromosomes, fitnesses, distances, pop_length):\n",
+    "  pos = item.get_id(1)\n",
     "  for j in range(pop_length-1):\n",
     "    fitnesses[pos] += distances[int(chromosomes[pos, j]), int(chromosomes[pos, j+1])]\n"
    ]
@@ -708,7 +710,8 @@
     "  chromosomes_flat_dpctl = dpnp.asarray(chromosomes, device=\"gpu\")\n",
     "  fitnesses_dpctl = dpnp.asarray(fitnesses.copy(), device=\"gpu\")\n",
     "\n",
-    "  eval_genomes_plain_TSP_SYCL[numba_dpex.Range(pop_size)](chromosomes_flat_dpctl, fitnesses_dpctl, distances_dpctl, pop_size)\n",
+    "  exec_range = kernel_api.Range(pop_size)\n",
+    "  numba_dpex.call_kernel(eval_genomes_plain_TSP_SYCL, exec_range, chromosomes_flat_dpctl, fitnesses_dpctl, distances_dpctl, pop_size)\n",
     "  fitnesses = dpnp.asnumpy(fitnesses_dpctl)\n",
     "  chromosomes = next_generation_TSP(chromosomes, fitnesses)\n",
     "  fitnesses = np.zeros(pop_size, dtype=np.float32)\n",

AI-and-Analytics/Features-and-Functionality/IntelPython_GPU_numba-dpex_Genetic_Algorithm/IntelPython_GPU_numba-dpex_Genetic_Algorithm.py

@@ -260,16 +260,17 @@ def next_generation(chromosomes, fitnesses):
 # 
 # The only par that differs form the standard implementation is the evaluation function.
 # 
-# The most important part is to specify the global index of the computation. This is the current index of the computed chromosomes. This serves as a loop function across all chromosomes.
+# The most important part is to specify the index of the computation. This is the current index of the computed chromosomes. This serves as a loop function across all chromosomes.
 
 # In[ ]:
 
 
 import numba_dpex
+from numba_dpex import kernel_api
 
 @numba_dpex.kernel
-def eval_genomes_sycl_kernel(chromosomes, fitnesses, chrom_length):
-  pos = numba_dpex.get_global_id(0)
+def eval_genomes_sycl_kernel(item: kernel_api.Item, chromosomes, fitnesses, chrom_length):
+  pos = item.get_id(0)
   num_loops = 3000
   for i in range(num_loops):
     fitnesses[pos] += chromosomes[pos*chrom_length + 1]
@@ -300,7 +301,8 @@ def eval_genomes_sycl_kernel(chromosomes, fitnesses, chrom_length):
   chromosomes_flat_dpctl = dpnp.asarray(chromosomes_flat, device="gpu")
   fitnesses_dpctl = dpnp.asarray(fitnesses, device="gpu")
 
-  eval_genomes_sycl_kernel[numba_dpex.Range(pop_size)](chromosomes_flat_dpctl, fitnesses_dpctl, chrom_size)
+  exec_range = kernel_api.Range(pop_size)
+  numba_dpex.call_kernel(eval_genomes_sycl_kernel, exec_range, chromosomes_flat_dpctl, fitnesses_dpctl, chrom_size)
   fitnesses = dpnp.asnumpy(fitnesses_dpctl)
   chromosomes = next_generation(chromosomes, fitnesses)
   fitnesses = np.zeros(pop_size, dtype=np.float32)
@@ -398,14 +400,14 @@ def eval_genomes_sycl_kernel(chromosomes, fitnesses, chrom_length):
 # 
 # The evaluate created generation we are calculating the full distance of the given path (chromosome). In this example, the lower the fitness value is, the better the chromosome. That's different from the general GA that we implemented.
 # 
-# As in this example we are also using numba-dpex, we are using a global index like before.
+# As in this example we are also using numba-dpex, we are using an index like before.
 
 # In[ ]:
 
 
 @numba_dpex.kernel
-def eval_genomes_plain_TSP_SYCL(chromosomes, fitnesses, distances, pop_length):
-  pos = numba_dpex.get_global_id(0)
+def eval_genomes_plain_TSP_SYCL(item: kernel_api.Item, chromosomes, fitnesses, distances, pop_length):
+  pos = item.get_id(1)
   for j in range(pop_length-1):
     fitnesses[pos] += distances[int(chromosomes[pos, j]), int(chromosomes[pos, j+1])]
 
@@ -526,7 +528,8 @@ def next_generation_TSP(chromosomes, fitnesses):
   chromosomes_flat_dpctl = dpnp.asarray(chromosomes, device="gpu")
   fitnesses_dpctl = dpnp.asarray(fitnesses.copy(), device="gpu")
 
-  eval_genomes_plain_TSP_SYCL[numba_dpex.Range(pop_size)](chromosomes_flat_dpctl, fitnesses_dpctl, distances_dpctl, pop_size)
+  exec_range = kernel_api.Range(pop_size)
+  numba_dpex.call_kernel(eval_genomes_plain_TSP_SYCL, exec_range, chromosomes_flat_dpctl, fitnesses_dpctl, distances_dpctl, pop_size)
   fitnesses = dpnp.asnumpy(fitnesses_dpctl)
   chromosomes = next_generation_TSP(chromosomes, fitnesses)
   fitnesses = np.zeros(pop_size, dtype=np.float32)