Upload export_job.py with huggingface_hub
Browse files- export_job.py +36 -42
export_job.py
CHANGED
|
@@ -1,31 +1,25 @@
|
|
| 1 |
# /// script
|
| 2 |
# requires-python = ">=3.11"
|
| 3 |
-
# dependencies = ["torch", "torchvision", "pillow", "numpy", "
|
| 4 |
# ///
|
| 5 |
-
"""HF-job credibility gate: convert the wave-3 winner checkpoint
|
| 6 |
-
and upload the artifact
|
| 7 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
from huggingface_hub import HfApi, hf_hub_download
|
| 9 |
|
| 10 |
tok = os.environ["HF_TOKEN"]
|
| 11 |
api = HfApi(token=tok)
|
| 12 |
ckpt = hf_hub_download("Tonic/challenge5-acoustic-squeezenet1_1-s7", "best.pt",
|
| 13 |
repo_type="model", token=tok)
|
| 14 |
-
print("ckpt:",
|
| 15 |
|
| 16 |
here = os.path.dirname(os.path.abspath(__file__))
|
| 17 |
-
# calibration spectrograms from the challenge pack are not in the job; synthesize rep data
|
| 18 |
-
# from the unified dataset windows instead (same rendering):
|
| 19 |
-
from datasets import load_dataset # noqa
|
| 20 |
-
sys.path.insert(0, here)
|
| 21 |
-
# write a tiny driver that reuses esp32_demo/tools/export_tflite.py logic but with
|
| 22 |
-
# synthetic calibration from unified windows
|
| 23 |
-
import numpy as np, torch
|
| 24 |
-
import ai_edge_torch
|
| 25 |
-
from PIL import Image
|
| 26 |
-
import io
|
| 27 |
-
from datasets import load_dataset
|
| 28 |
-
|
| 29 |
ck = torch.load(ckpt, map_location="cpu", weights_only=False)
|
| 30 |
sd = ck.get("state_dict", ck) if isinstance(ck, dict) else ck
|
| 31 |
arch = ck.get("arch", "squeezenet1_1") if isinstance(ck, dict) else "squeezenet1_1"
|
|
@@ -39,31 +33,31 @@ model.load_state_dict(sd, strict=False)
|
|
| 39 |
model.eval()
|
| 40 |
print("arch:", arch, "params:", sum(p.numel() for p in model.parameters()), flush=True)
|
| 41 |
|
| 42 |
-
|
| 43 |
-
|
| 44 |
-
|
| 45 |
-
|
| 46 |
-
|
| 47 |
-
|
| 48 |
-
|
| 49 |
-
|
| 50 |
-
|
| 51 |
-
|
| 52 |
-
|
| 53 |
-
|
| 54 |
|
| 55 |
-
|
| 56 |
-
conv =
|
| 57 |
-
|
| 58 |
-
|
| 59 |
-
|
| 60 |
-
except AttributeError:
|
| 61 |
-
tfl = conv.convert_with_dynamic_range_quantization()
|
| 62 |
-
mode = "dynamic-range"
|
| 63 |
-
out = f"/tmp/model_full_224_{mode.replace('-', '_')}.tflite"
|
| 64 |
open(out, "wb").write(tfl)
|
| 65 |
-
print("
|
| 66 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 67 |
repo_id="Tonic/challenge5-acoustic-code", repo_type="model", token=tok,
|
| 68 |
-
commit_message=
|
| 69 |
-
print("UPLOADED esp32/
|
|
|
|
| 1 |
# /// script
|
| 2 |
# requires-python = ">=3.11"
|
| 3 |
+
# dependencies = ["torch", "torchvision", "pillow", "numpy", "onnx", "onnx2tf", "tensorflow-cpu", "huggingface_hub>=0.30", "timm", "datasets>=4.0"]
|
| 4 |
# ///
|
| 5 |
+
"""HF-job credibility gate: convert the wave-3 winner checkpoint (best.pt) to a
|
| 6 |
+
quantized TFLite model for ESP32 TFLite-Micro and upload the artifact to the code repo.
|
| 7 |
+
|
| 8 |
+
Path: best.pt -> ONNX (torch.onnx) -> SavedModel (onnx2tf, float) -> TFLite with
|
| 9 |
+
dynamic-range int8 weights (Optimize.DEFAULT). TFLite-Micro runs this fine on ESP32.
|
| 10 |
+
"""
|
| 11 |
+
import os, sys, io
|
| 12 |
+
import numpy as np
|
| 13 |
+
import torch
|
| 14 |
from huggingface_hub import HfApi, hf_hub_download
|
| 15 |
|
| 16 |
tok = os.environ["HF_TOKEN"]
|
| 17 |
api = HfApi(token=tok)
|
| 18 |
ckpt = hf_hub_download("Tonic/challenge5-acoustic-squeezenet1_1-s7", "best.pt",
|
| 19 |
repo_type="model", token=tok)
|
| 20 |
+
print("ckpt:", os.path.getsize(ckpt) // 1024, "KB", flush=True)
|
| 21 |
|
| 22 |
here = os.path.dirname(os.path.abspath(__file__))
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 23 |
ck = torch.load(ckpt, map_location="cpu", weights_only=False)
|
| 24 |
sd = ck.get("state_dict", ck) if isinstance(ck, dict) else ck
|
| 25 |
arch = ck.get("arch", "squeezenet1_1") if isinstance(ck, dict) else "squeezenet1_1"
|
|
|
|
| 33 |
model.eval()
|
| 34 |
print("arch:", arch, "params:", sum(p.numel() for p in model.parameters()), flush=True)
|
| 35 |
|
| 36 |
+
SIZE = 224
|
| 37 |
+
onnx_path = "/tmp/model_full_224.onnx"
|
| 38 |
+
torch.onnx.export(model, torch.zeros(1, 3, SIZE, SIZE), onnx_path, opset_version=17,
|
| 39 |
+
input_names=["spectrogram"], output_names=["wake_logit", "sound_class_logits"],
|
| 40 |
+
dynamic_axes={"spectrogram": {0: "batch"}})
|
| 41 |
+
print("onnx exported", flush=True)
|
| 42 |
+
|
| 43 |
+
import onnx2tf
|
| 44 |
+
sm_dir = "/tmp/saved_model"
|
| 45 |
+
onnx2tf.convert(input_onnx_file_path=onnx_path, output_folder_path=sm_dir,
|
| 46 |
+
not_use_onnxsim=True, verbosity="error")
|
| 47 |
+
print("onnx2tf done", flush=True)
|
| 48 |
|
| 49 |
+
import tensorflow as tf
|
| 50 |
+
conv = tf.lite.TFLiteConverter.from_saved_model(sm_dir)
|
| 51 |
+
conv.optimizations = [tf.lite.Optimize.DEFAULT] # dynamic-range int8 weights
|
| 52 |
+
tfl = conv.convert()
|
| 53 |
+
out = "/tmp/model_full_224_dynamic_int8.tflite"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 54 |
open(out, "wb").write(tfl)
|
| 55 |
+
print("tflite dynamic-int8:", os.path.getsize(out) // 1024, "KB", flush=True)
|
| 56 |
+
|
| 57 |
+
api.upload_file(path_or_fileobj=out, path_in_repo="esp32/model_full_224_dynamic_int8.tflite",
|
| 58 |
+
repo_id="Tonic/challenge5-acoustic-code", repo_type="model", token=tok,
|
| 59 |
+
commit_message=f"esp32: dynamic-int8 tflite of squeezenet1_1-s7 ({os.path.getsize(out)//1024} KB)")
|
| 60 |
+
api.upload_file(path_or_fileobj=onnx_path, path_in_repo="esp32/model_full_224.onnx",
|
| 61 |
repo_id="Tonic/challenge5-acoustic-code", repo_type="model", token=tok,
|
| 62 |
+
commit_message="esp32: onnx of squeezenet1_1-s7")
|
| 63 |
+
print("UPLOADED esp32/ artifacts", flush=True)
|