create engine folder and move code (#3948)

* retry

* older tf

* that

test/external/external_benchmark_multitensor_allreduce.py

@@ -3,8 +3,8 @@ from tinygrad import Tensor, Device
 from tinygrad.lazy import LazyBuffer
 from tinygrad.ops import ReduceOps, GlobalCounters
 from tinygrad.features.multi import MultiLazyBuffer, all_reduce
-from tinygrad.features.jit import TinyJit
-from tinygrad.realize import create_schedule, run_schedule
+from tinygrad.engine.jit import TinyJit
+from tinygrad.engine.realize import create_schedule, run_schedule
 from tinygrad.helpers import getenv, Context, RING
 from typing import List, Union
 

test/external/external_model_benchmark.py

@@ -66,7 +66,7 @@ def benchmark_model(m, devices, validate_outs=False):
     tinygrad_model = get_run_onnx(onnx_model)
     benchmark(m, f"tinygrad_{device.lower()}_jitless", lambda: {k:v.numpy() for k,v in tinygrad_model(inputs).items()})
 
-    from tinygrad.features.jit import TinyJit
+    from tinygrad.engine.jit import TinyJit
     tinygrad_jitted_model = TinyJit(lambda **kwargs: {k:v.realize() for k,v in tinygrad_model(kwargs).items()})
     for _ in range(3): {k:v.numpy() for k,v in tinygrad_jitted_model(**inputs).items()}
     benchmark(m, f"tinygrad_{device.lower()}_jit", lambda: {k:v.numpy() for k,v in tinygrad_jitted_model(**inputs).items()}) # noqa: F821

test/external/external_test_hip_compile.py

@@ -2,7 +2,7 @@ import time, unittest
 from tinygrad.runtime.driver.hip_comgr import compile_hip
 from tinygrad import Tensor
 from tinygrad.device import Device
-from tinygrad.realize import create_schedule
+from tinygrad.engine.realize import create_schedule
 from tinygrad.codegen.linearizer import Linearizer
 
 class TestHIPCompileSpeed(unittest.TestCase):

test/external/external_test_hsa_driver.py

@@ -4,7 +4,7 @@ from tinygrad.device import Device, Buffer, BufferXfer
 from tinygrad.dtype import dtypes
 from tinygrad.runtime.driver.hsa import AQLQueue
 from tinygrad.runtime.graph.hsa import VirtAQLQueue, HSAGraph
-from tinygrad.features.jit import JitItem
+from tinygrad.engine.jit import JitItem
 
 def get_hsa_inc_prog(dev, inc=1):
   prg = f"""

test/external/external_test_jit_on_models.py

@@ -2,7 +2,7 @@
 import unittest
 import numpy as np
 from tinygrad import Tensor, dtypes
-from tinygrad.features.jit import TinyJit
+from tinygrad.engine.jit import TinyJit
 from tinygrad.helpers import CI
 from test.helpers import derandomize_model
 

test/external/external_test_opt.py

@@ -17,7 +17,7 @@ from tinygrad.helpers import getenv
 from tinygrad.nn import optim
 #from tinygrad.lazy import PUSH_PERMUTES
 PUSH_PERMUTES = False
-from tinygrad.features.jit import CacheCollector
+from tinygrad.engine.jit import CacheCollector
 
 class CLCache:
   def __init__(self, allowed=None, strict=False, preclear=True, var_vals=None):

test/external/external_test_uops_graphing.py

@@ -4,7 +4,7 @@ from tinygrad.tensor import Tensor
 from tinygrad.codegen.linearizer import Linearizer
 from tinygrad.renderer.cstyle import OpenCLRenderer
 from tinygrad.features.graph import graph_uops
-from tinygrad.realize import create_schedule
+from tinygrad.engine.realize import create_schedule
 from tinygrad.nn import Conv2d
 
 class TestUopsGraph(unittest.TestCase):