79 lines
3.3 KiB
Python
79 lines
3.3 KiB
Python
import os
|
|
import glob
|
|
|
|
Import('env', 'arch')
|
|
lenv = env.Clone()
|
|
CHUNK_BYTES = int(os.environ.get("TG_CHUNK_BYTES", str(45 * 1024 * 1024)))
|
|
|
|
tinygrad_root = env.Dir("#").abspath
|
|
tinygrad_files = ["#"+x for x in glob.glob(env.Dir("#tinygrad_repo").relpath + "/**", recursive=True, root_dir=tinygrad_root)
|
|
if 'pycache' not in x and os.path.isfile(os.path.join(tinygrad_root, x))]
|
|
|
|
# Get model metadata
|
|
for model_name in ['driving_vision', 'driving_policy', 'dmonitoring_model']:
|
|
fn = File(f"models/{model_name}").abspath
|
|
script_files = [File(Dir("#selfdrive/modeld").File("get_model_metadata.py").abspath)]
|
|
cmd = f'python3 {Dir("#selfdrive/modeld").abspath}/get_model_metadata.py {fn}.onnx'
|
|
lenv.Command(fn + "_metadata.pkl", [fn + ".onnx"] + tinygrad_files + script_files, cmd)
|
|
|
|
# compile warp
|
|
# THREADS=0 is need to prevent bug: https://github.com/tinygrad/tinygrad/issues/14689
|
|
tg_flags = {
|
|
'larch64': 'DEV=QCOM FLOAT16=1 NOLOCALS=1 JIT_BATCH_SIZE=0',
|
|
'Darwin': f'DEV=CPU THREADS=0 HOME={os.path.expanduser("~")}', # tinygrad calls brew which needs a $HOME in the env
|
|
}.get(arch, 'DEV=CPU CPU_LLVM=1 THREADS=0')
|
|
image_flag = {
|
|
'larch64': 'IMAGE=2',
|
|
}.get(arch, 'IMAGE=0')
|
|
script_files = [File(Dir("#selfdrive/modeld").File("compile_warp.py").abspath)]
|
|
cmd = f'{tg_flags} python3 {Dir("#selfdrive/modeld").abspath}/compile_warp.py '
|
|
from openpilot.common.transformations.camera import _ar_ox_fisheye, _os_fisheye
|
|
warp_targets = []
|
|
for cam in [_ar_ox_fisheye, _os_fisheye]:
|
|
w, h = cam.width, cam.height
|
|
warp_targets += [File(f"models/warp_{w}x{h}_tinygrad.pkl").abspath, File(f"models/dm_warp_{w}x{h}_tinygrad.pkl").abspath]
|
|
lenv.Command(warp_targets, tinygrad_files + script_files, cmd)
|
|
|
|
def tg_compile(flags, model_name):
|
|
pythonpath_string = 'PYTHONPATH="${PYTHONPATH}:' + env.Dir("#tinygrad_repo").abspath + '"'
|
|
fn = File(f"models/{model_name}").abspath
|
|
|
|
out = fn + "_tinygrad.pkl"
|
|
full = out + ".full"
|
|
parts = out + ".parts"
|
|
|
|
full_node = lenv.Command(
|
|
full,
|
|
[fn + ".onnx"] + tinygrad_files,
|
|
f'{pythonpath_string} {flags} {image_flag} python3 {Dir("#tinygrad_repo").abspath}/examples/openpilot/compile3.py {fn}.onnx {full}'
|
|
)
|
|
|
|
split_script = File(Dir("#selfdrive/modeld").File("external_pickle.py").abspath)
|
|
parts_node = lenv.Command(
|
|
parts,
|
|
[full_node, split_script, Value(str(CHUNK_BYTES))],
|
|
[f'python3 {split_script.abspath} {full} {out} {CHUNK_BYTES}', Delete(full)],
|
|
)
|
|
|
|
lenv.NoCache(parts_node)
|
|
lenv.AlwaysBuild(parts_node)
|
|
return parts_node
|
|
|
|
# Compile small models
|
|
for model_name in ['driving_vision', 'driving_policy', 'dmonitoring_model']:
|
|
tg_compile(tg_flags, model_name)
|
|
|
|
# Compile BIG model if USB GPU is available
|
|
if "USBGPU" in os.environ:
|
|
import subprocess
|
|
# because tg doesn't support multi-process
|
|
devs = subprocess.check_output('python3 -c "from tinygrad import Device; print(list(Device.get_available_devices()))"', shell=True, cwd=env.Dir('#').abspath)
|
|
if b"AMD" in devs:
|
|
print("USB GPU detected... building")
|
|
flags = "DEV=AMD AMD_IFACE=USB AMD_LLVM=1 NOLOCALS=0 IMAGE=0"
|
|
bp = tg_compile(flags, "big_driving_policy")
|
|
bv = tg_compile(flags, "big_driving_vision")
|
|
lenv.SideEffect('lock', [bp, bv]) # tg doesn't support multi-process so build serially
|
|
else:
|
|
print("USB GPU not detected... skipping")
|