Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -1,20 +1,49 @@
|
|
| 1 |
-
#
|
| 2 |
-
import
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
import matplotlib.pyplot as plt
|
|
|
|
| 7 |
import shap
|
|
|
|
| 8 |
|
| 9 |
-
|
| 10 |
-
MODEL_PATH = "model_extratrees_ds4.joblib"
|
| 11 |
FEATS_PATH = "feature_names.json"
|
| 12 |
|
| 13 |
-
|
|
|
|
|
|
|
| 14 |
model = joblib.load(MODEL_PATH)
|
|
|
|
| 15 |
with open(FEATS_PATH) as f:
|
| 16 |
FEATURES = json.load(f)
|
| 17 |
|
|
|
|
| 18 |
def _clean_headers(df):
|
| 19 |
df.columns = (df.columns.astype(str)
|
| 20 |
.str.replace(r'[\u200b\ufeff]', '', regex=True)
|
|
@@ -32,6 +61,7 @@ def _read_table(file):
|
|
| 32 |
on_bad_lines="skip", low_memory=False,
|
| 33 |
encoding_errors="ignore")
|
| 34 |
except Exception:
|
|
|
|
| 35 |
for sep in [",", "\t", ";", "|"]:
|
| 36 |
try:
|
| 37 |
df = pd.read_csv(name, engine="python", sep=sep, dtype=str,
|
|
@@ -41,6 +71,8 @@ def _read_table(file):
|
|
| 41 |
break
|
| 42 |
except Exception:
|
| 43 |
pass
|
|
|
|
|
|
|
| 44 |
return _clean_headers(df)
|
| 45 |
|
| 46 |
def _coerce_numeric(df):
|
|
@@ -60,6 +92,7 @@ def _validate_columns(df):
|
|
| 60 |
extra = [c for c in df.columns if c not in FEATURES]
|
| 61 |
return missing, extra
|
| 62 |
|
|
|
|
| 63 |
def predict_csv(file):
|
| 64 |
try:
|
| 65 |
if file is None: return None, "No file."
|
|
@@ -69,7 +102,9 @@ def predict_csv(file):
|
|
| 69 |
X = _prepare(df)
|
| 70 |
proba = model.predict_proba(X)[:, 1]
|
| 71 |
pred = (proba > 0.5).astype(int)
|
| 72 |
-
out = pd.DataFrame({"row_index": df.index,
|
|
|
|
|
|
|
| 73 |
return out, "OK"
|
| 74 |
except Exception:
|
| 75 |
return None, "ERROR:\n" + traceback.format_exc()
|
|
@@ -82,7 +117,6 @@ def explain_first_row(file):
|
|
| 82 |
if missing: return None, f"Missing features for SHAP. First 10: {missing[:10]}"
|
| 83 |
X = _prepare(df)
|
| 84 |
if X.shape[0] == 0: return None, "No rows."
|
| 85 |
-
|
| 86 |
X_imp = pd.DataFrame(model.named_steps["imputer"].transform(X), columns=FEATURES)
|
| 87 |
clf = model.named_steps["clf"]
|
| 88 |
|
|
@@ -102,6 +136,7 @@ def explain_first_row(file):
|
|
| 102 |
except Exception:
|
| 103 |
return None, "ERROR:\n" + traceback.format_exc()
|
| 104 |
|
|
|
|
| 105 |
with gr.Blocks(title="Variant Pathogenicity Demo β ExtraTrees (Dataset-4)") as demo:
|
| 106 |
gr.Markdown("### Variant Pathogenicity Demo β ExtraTrees (Dataset-4)")
|
| 107 |
inp = gr.File(label="Upload CSV/XLSX containing the trained feature columns")
|
|
@@ -109,7 +144,7 @@ with gr.Blocks(title="Variant Pathogenicity Demo β ExtraTrees (Dataset-4)") as
|
|
| 109 |
btn_pred = gr.Button("Predict")
|
| 110 |
btn_expl = gr.Button("Explain first row")
|
| 111 |
out_df = gr.Dataframe(label="Predictions")
|
| 112 |
-
out_plot = gr.Plot(label="SHAP (row #1)")
|
| 113 |
out_txt = gr.Markdown()
|
| 114 |
btn_pred.click(fn=predict_csv, inputs=inp, outputs=[out_df, out_txt])
|
| 115 |
btn_expl.click(fn=explain_first_row, inputs=inp, outputs=[out_plot, out_txt])
|
|
|
|
| 1 |
+
# --- bootstrap: force-install deps if Hugging Face didn't ---
|
| 2 |
+
import sys, subprocess
|
| 3 |
+
def _need(modname):
|
| 4 |
+
try: __import__(modname); return False
|
| 5 |
+
except ImportError: return True
|
| 6 |
+
def _pip(pkg_spec):
|
| 7 |
+
subprocess.check_call([sys.executable, "-m", "pip", "install", pkg_spec])
|
| 8 |
+
|
| 9 |
+
PKGS = [
|
| 10 |
+
("joblib", "1.4.2", "joblib"),
|
| 11 |
+
("numpy", "1.26.4","numpy"),
|
| 12 |
+
("pandas", "2.2.2", "pandas"),
|
| 13 |
+
("scikit-learn", "1.4.2", "sklearn"),
|
| 14 |
+
("scipy", "1.11.4","scipy"),
|
| 15 |
+
("shap", "0.46.0","shap"),
|
| 16 |
+
("matplotlib", "3.8.4", "matplotlib"),
|
| 17 |
+
("openpyxl", "3.1.5", "openpyxl"),
|
| 18 |
+
("pillow", None, "PIL"),
|
| 19 |
+
("gradio", "4.44.1","gradio"),
|
| 20 |
+
]
|
| 21 |
+
for pkg, ver, mod in PKGS:
|
| 22 |
+
if _need(mod):
|
| 23 |
+
_pip(f"{pkg}=={ver}" if ver else pkg)
|
| 24 |
+
|
| 25 |
+
# now safe to import everything
|
| 26 |
+
import os, json, io, gzip, joblib, traceback
|
| 27 |
+
import numpy as np, pandas as pd
|
| 28 |
+
import matplotlib; matplotlib.use("Agg") # headless rendering
|
| 29 |
import matplotlib.pyplot as plt
|
| 30 |
+
from PIL import Image
|
| 31 |
import shap
|
| 32 |
+
import gradio as gr
|
| 33 |
|
| 34 |
+
# ---- artifacts in repo root ----
|
| 35 |
+
MODEL_PATH = "model_extratrees_ds4.joblib" # or "model_extratrees_ds4.gz"
|
| 36 |
FEATS_PATH = "feature_names.json"
|
| 37 |
|
| 38 |
+
# if you uploaded a .gz model, uncomment these two lines and delete the next joblib.load:
|
| 39 |
+
# with gzip.open(MODEL_PATH, "rb") as f:
|
| 40 |
+
# model = joblib.load(f)
|
| 41 |
model = joblib.load(MODEL_PATH)
|
| 42 |
+
|
| 43 |
with open(FEATS_PATH) as f:
|
| 44 |
FEATURES = json.load(f)
|
| 45 |
|
| 46 |
+
# ---------- helpers ----------
|
| 47 |
def _clean_headers(df):
|
| 48 |
df.columns = (df.columns.astype(str)
|
| 49 |
.str.replace(r'[\u200b\ufeff]', '', regex=True)
|
|
|
|
| 61 |
on_bad_lines="skip", low_memory=False,
|
| 62 |
encoding_errors="ignore")
|
| 63 |
except Exception:
|
| 64 |
+
df = None
|
| 65 |
for sep in [",", "\t", ";", "|"]:
|
| 66 |
try:
|
| 67 |
df = pd.read_csv(name, engine="python", sep=sep, dtype=str,
|
|
|
|
| 71 |
break
|
| 72 |
except Exception:
|
| 73 |
pass
|
| 74 |
+
if df is None:
|
| 75 |
+
raise
|
| 76 |
return _clean_headers(df)
|
| 77 |
|
| 78 |
def _coerce_numeric(df):
|
|
|
|
| 92 |
extra = [c for c in df.columns if c not in FEATURES]
|
| 93 |
return missing, extra
|
| 94 |
|
| 95 |
+
# ---------- app fns ----------
|
| 96 |
def predict_csv(file):
|
| 97 |
try:
|
| 98 |
if file is None: return None, "No file."
|
|
|
|
| 102 |
X = _prepare(df)
|
| 103 |
proba = model.predict_proba(X)[:, 1]
|
| 104 |
pred = (proba > 0.5).astype(int)
|
| 105 |
+
out = pd.DataFrame({"row_index": df.index,
|
| 106 |
+
"pred_prob_pathogenic": proba,
|
| 107 |
+
"pred_label": pred})
|
| 108 |
return out, "OK"
|
| 109 |
except Exception:
|
| 110 |
return None, "ERROR:\n" + traceback.format_exc()
|
|
|
|
| 117 |
if missing: return None, f"Missing features for SHAP. First 10: {missing[:10]}"
|
| 118 |
X = _prepare(df)
|
| 119 |
if X.shape[0] == 0: return None, "No rows."
|
|
|
|
| 120 |
X_imp = pd.DataFrame(model.named_steps["imputer"].transform(X), columns=FEATURES)
|
| 121 |
clf = model.named_steps["clf"]
|
| 122 |
|
|
|
|
| 136 |
except Exception:
|
| 137 |
return None, "ERROR:\n" + traceback.format_exc()
|
| 138 |
|
| 139 |
+
# ---------- UI ----------
|
| 140 |
with gr.Blocks(title="Variant Pathogenicity Demo β ExtraTrees (Dataset-4)") as demo:
|
| 141 |
gr.Markdown("### Variant Pathogenicity Demo β ExtraTrees (Dataset-4)")
|
| 142 |
inp = gr.File(label="Upload CSV/XLSX containing the trained feature columns")
|
|
|
|
| 144 |
btn_pred = gr.Button("Predict")
|
| 145 |
btn_expl = gr.Button("Explain first row")
|
| 146 |
out_df = gr.Dataframe(label="Predictions")
|
| 147 |
+
out_plot = gr.Plot(label="SHAP (row #1)")
|
| 148 |
out_txt = gr.Markdown()
|
| 149 |
btn_pred.click(fn=predict_csv, inputs=inp, outputs=[out_df, out_txt])
|
| 150 |
btn_expl.click(fn=explain_first_row, inputs=inp, outputs=[out_plot, out_txt])
|