mirror of https://github.com/commaai/tinygrad.git
46850a0269
* search: add a BEAM_COMPARE env to optionally not compare to hc/tc setting BEAM_COMPARE=0 will prevent additional memory allocation needed to do the timing tests assuming the BEAM result is in the diskcache. * change to optionally use Buffer.allocate |
||
|---|---|---|
| .. | ||
| accel | ||
| assembly | ||
| backends | ||
| datasets | ||
| gemm | ||
| hip_gpu_driver | ||
| hiprtc | ||
| junk | ||
| models | ||
| nv_gpu_driver | ||
| optimization | ||
| qcom_gpu_driver | ||
| archprobe.py | ||
| augment.py | ||
| autopad.py | ||
| disk_read_speed.py | ||
| dump_cache.py | ||
| export_model.py | ||
| gradcheck.py | ||
| hip_events.py | ||
| introspection.py | ||
| lr_scheduler.py | ||
| multitensor.py | ||
| onnx.py | ||
| onnx_ops.py | ||
| ring_copy.py | ||
| thneed.py | ||
| to_movement_ops.py | ||
| training.py | ||
| transfer_speed.py | ||