Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -1,5 +1,4 @@
|
|
| 1 |
import os
|
| 2 |
-
import json
|
| 3 |
import numpy as np
|
| 4 |
import pandas as pd
|
| 5 |
import gradio as gr
|
|
@@ -7,21 +6,26 @@ import pickle
|
|
| 7 |
import tensorflow as tf
|
| 8 |
|
| 9 |
# =========================================================
|
| 10 |
-
# Settings / Constants
|
| 11 |
# =========================================================
|
| 12 |
BEST_L = 3
|
| 13 |
CC_COL = "CarbonContents"
|
| 14 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
RPM_RANGE = (150, 650)
|
| 16 |
-
RPM_DELTAS = (-200, -150, -100, -50, 50, 100, 150, 200)
|
| 17 |
FORBID_BOUNDARY = True
|
| 18 |
|
| 19 |
DEFAULT_MIN_GAIN_PRED = 0.5
|
| 20 |
|
| 21 |
-
# μΆμ² λμ timepoint rule
|
| 22 |
-
START_HOURS = 12.0
|
| 23 |
-
START_BDO = 20.0
|
| 24 |
-
|
| 25 |
# =========================================================
|
| 26 |
# Gene set (must match training)
|
| 27 |
# =========================================================
|
|
@@ -64,9 +68,19 @@ for k in keys_to_drop:
|
|
| 64 |
medium_cols_carbon = list(Carbon_source_dict.keys())
|
| 65 |
medium_cols_salts = list(Base_dict.keys()) # salts = base
|
| 66 |
medium_cols_add = list(Additional_dict.keys()) # additives
|
| 67 |
-
|
| 68 |
medium_cols = medium_cols_carbon + medium_cols_salts + medium_cols_add
|
| 69 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 70 |
# =========================================================
|
| 71 |
# Medium display name mapping (hide "_i" in UI)
|
| 72 |
# =========================================================
|
|
@@ -102,8 +116,7 @@ else:
|
|
| 102 |
model = tf.keras.models.load_model(MODEL_PATH, compile=False)
|
| 103 |
|
| 104 |
# =========================================================
|
| 105 |
-
# Presets
|
| 106 |
-
# Preset format: dict[preset_key] = dict[internal_component -> value]
|
| 107 |
# =========================================================
|
| 108 |
def _safe_read_csv(path: str):
|
| 109 |
try:
|
|
@@ -114,21 +127,10 @@ def _safe_read_csv(path: str):
|
|
| 114 |
return None
|
| 115 |
|
| 116 |
def build_top5_presets_from_csv(csv_path: str, top_k: int = 5):
|
| 117 |
-
"""
|
| 118 |
-
If training CSV is available in Space, derive top-K medium combinations:
|
| 119 |
-
- For each BatchNo, take the first timepoint row's medium values
|
| 120 |
-
- Define combo by set of nonzero medium components
|
| 121 |
-
- Count combos, pick top-K
|
| 122 |
-
- For each combo, preset values = median of those components across batches in that combo
|
| 123 |
-
"""
|
| 124 |
df = _safe_read_csv(csv_path)
|
| 125 |
-
if df is None:
|
| 126 |
-
return {}
|
| 127 |
-
|
| 128 |
-
if "BatchNo" not in df.columns:
|
| 129 |
return {}
|
| 130 |
|
| 131 |
-
# ensure columns exist
|
| 132 |
for c in medium_cols:
|
| 133 |
if c not in df.columns:
|
| 134 |
df[c] = 0.0
|
|
@@ -143,23 +145,19 @@ def build_top5_presets_from_csv(csv_path: str, top_k: int = 5):
|
|
| 143 |
else:
|
| 144 |
first_rows = df.groupby("BatchNo", as_index=False).first()
|
| 145 |
|
| 146 |
-
# numeric
|
| 147 |
first_rows[medium_cols] = first_rows[medium_cols].apply(pd.to_numeric, errors="coerce").fillna(0.0)
|
| 148 |
|
| 149 |
-
# combo signature by nonzero components
|
| 150 |
def signature(row):
|
| 151 |
nz = [c for c in medium_cols if float(row[c]) != 0.0]
|
| 152 |
nz.sort()
|
| 153 |
return "|".join(nz)
|
| 154 |
|
| 155 |
first_rows["_sig"] = first_rows.apply(signature, axis=1)
|
| 156 |
-
# drop empty signature (no medium info)
|
| 157 |
first_rows = first_rows[first_rows["_sig"].astype(str).str.len() > 0].copy()
|
| 158 |
if len(first_rows) == 0:
|
| 159 |
return {}
|
| 160 |
|
| 161 |
-
|
| 162 |
-
top_sigs = sig_counts.head(top_k).index.tolist()
|
| 163 |
|
| 164 |
presets = {}
|
| 165 |
for i, sig in enumerate(top_sigs, start=1):
|
|
@@ -169,12 +167,9 @@ def build_top5_presets_from_csv(csv_path: str, top_k: int = 5):
|
|
| 169 |
for c in comps:
|
| 170 |
vals[c] = float(np.median(sub[c].to_numpy(dtype=float)))
|
| 171 |
presets[f"TopCombo{i}"] = vals
|
| 172 |
-
|
| 173 |
return presets
|
| 174 |
|
| 175 |
def fallback_example_presets():
|
| 176 |
-
# NOTE: CSVκ° Spaceμ μμ λλ dropdownμ΄ "νμ" 보μ΄λλ‘ μμ 5κ°λ₯Ό μ 곡
|
| 177 |
-
# κ°μ 0μ΄ μλ μμλ‘λ§ λ£μκ³ , μ€μ μ€ν 쑰건과 λ€λ₯Ό μ μμ.
|
| 178 |
ex = {
|
| 179 |
"Example1": {"Glucose_i": 1.0, "YE_i": 10.0, "K2HPO4_i": 3.0, "KH2PO4_i": 1.0, "MgSO4Β·7H2O_i": 0.5},
|
| 180 |
"Example2": {"Glucose_i": 1.0, "YE_i": 5.0, "NaCl_i": 1.0, "K2HPO4_i": 3.0, "MgSO4Β·7H2O_i": 0.5},
|
|
@@ -182,7 +177,6 @@ def fallback_example_presets():
|
|
| 182 |
"Example4": {"Glucose_i": 1.0, "YE_i": 10.0, "K2HPO4_i": 3.0, "MgSO4Β·7H2O_i": 0.5, "CSL_i": 5.0},
|
| 183 |
"Example5": {"Glucose_i": 1.0, "YE_i": 10.0, "K2HPO4_i": 3.0, "MgSO4Β·7H2O_i": 0.5, "IPTG_i": 0.1},
|
| 184 |
}
|
| 185 |
-
# keep only known keys
|
| 186 |
cleaned = {}
|
| 187 |
for k, d in ex.items():
|
| 188 |
dd = {}
|
|
@@ -192,7 +186,6 @@ def fallback_example_presets():
|
|
| 192 |
cleaned[k] = dd
|
| 193 |
return cleaned
|
| 194 |
|
| 195 |
-
# Try to build presets from CSV if it exists in Space
|
| 196 |
PRESETS = build_top5_presets_from_csv("BDO_Data_cleaned_with_Feed_deltat.csv", top_k=5)
|
| 197 |
if not PRESETS:
|
| 198 |
PRESETS = fallback_example_presets()
|
|
@@ -204,13 +197,10 @@ def preset_label(preset_key: str) -> str:
|
|
| 204 |
items = []
|
| 205 |
for internal, v in d.items():
|
| 206 |
disp = INTERNAL_TO_DISP.get(internal, internal)
|
| 207 |
-
|
| 208 |
-
fv = float(v)
|
| 209 |
-
except Exception:
|
| 210 |
-
fv = v
|
| 211 |
items.append((disp, fv))
|
| 212 |
items = sorted(items, key=lambda x: x[0])
|
| 213 |
-
inside = "; ".join([f"{n}={val:g}
|
| 214 |
return f"{preset_key} ({inside})"
|
| 215 |
|
| 216 |
PRESET_LABELS = {preset_label(k): k for k in PRESETS.keys()}
|
|
@@ -220,18 +210,13 @@ def apply_preset(preset_label_selected: str):
|
|
| 220 |
preset_key = PRESET_LABELS.get(preset_label_selected, "")
|
| 221 |
d = PRESETS.get(preset_key, {})
|
| 222 |
if not isinstance(d, dict) or len(d) == 0:
|
| 223 |
-
return [], [], [], pd.DataFrame(columns=["component", "value"])
|
| 224 |
|
| 225 |
-
# split by category and convert to display names
|
| 226 |
carbon_sel, salts_sel, add_sel = [], [], []
|
| 227 |
rows = []
|
| 228 |
-
|
| 229 |
for internal, v in d.items():
|
| 230 |
disp = INTERNAL_TO_DISP.get(internal, internal)
|
| 231 |
-
|
| 232 |
-
fv = float(v)
|
| 233 |
-
except Exception:
|
| 234 |
-
fv = 0.0
|
| 235 |
rows.append([disp, fv])
|
| 236 |
|
| 237 |
if internal in medium_cols_carbon:
|
|
@@ -241,11 +226,8 @@ def apply_preset(preset_label_selected: str):
|
|
| 241 |
elif internal in medium_cols_add:
|
| 242 |
add_sel.append(disp)
|
| 243 |
|
| 244 |
-
df = pd.DataFrame(rows, columns=["component", "value"])
|
| 245 |
-
|
| 246 |
-
salts_sel = sorted(list(set(salts_sel)))
|
| 247 |
-
add_sel = sorted(list(set(add_sel)))
|
| 248 |
-
return carbon_sel, salts_sel, add_sel, df
|
| 249 |
|
| 250 |
# =========================================================
|
| 251 |
# Gene embedding helpers
|
|
@@ -269,11 +251,9 @@ def build_gene_embedding(deleted_txt: str, added_txt: str, replaced_txt: str):
|
|
| 269 |
for g in deleted:
|
| 270 |
if g in g2i:
|
| 271 |
emb[g2i[g]] = -1.0
|
| 272 |
-
|
| 273 |
for g in added:
|
| 274 |
if g in g2i:
|
| 275 |
emb[g2i[g]] += 1.0
|
| 276 |
-
|
| 277 |
for item in replaced:
|
| 278 |
if "->" in item:
|
| 279 |
old, new = [x.strip() for x in item.split("->", 1)]
|
|
@@ -284,7 +264,6 @@ def build_gene_embedding(deleted_txt: str, added_txt: str, replaced_txt: str):
|
|
| 284 |
else:
|
| 285 |
if item in g2i:
|
| 286 |
emb[g2i[item]] = -1.0
|
| 287 |
-
|
| 288 |
return emb.tolist()
|
| 289 |
|
| 290 |
# =========================================================
|
|
@@ -296,51 +275,47 @@ def build_medium_vector_from_df(df: pd.DataFrame):
|
|
| 296 |
return vec.tolist()
|
| 297 |
|
| 298 |
dff = df.copy()
|
| 299 |
-
|
|
|
|
| 300 |
return vec.tolist()
|
| 301 |
|
| 302 |
dff["component"] = dff["component"].astype(str)
|
| 303 |
-
dff[
|
| 304 |
|
| 305 |
m2i = {m: i for i, m in enumerate(medium_cols)}
|
| 306 |
-
|
| 307 |
for _, r in dff.iterrows():
|
| 308 |
disp = r["component"]
|
| 309 |
internal = DISP_TO_INTERNAL.get(disp, None)
|
| 310 |
if internal is None:
|
| 311 |
continue
|
| 312 |
-
|
| 313 |
-
vec[m2i[internal]] = float(r["value"])
|
| 314 |
-
|
| 315 |
return vec.tolist()
|
| 316 |
|
| 317 |
def merge_selected_to_table(existing_df, carbon_sel, salts_sel, add_sel):
|
| 318 |
-
"""
|
| 319 |
-
Table rows = union of selected components across 3 categories.
|
| 320 |
-
Keep existing values if present.
|
| 321 |
-
"""
|
| 322 |
all_sel = []
|
| 323 |
for arr in [carbon_sel, salts_sel, add_sel]:
|
| 324 |
if arr:
|
| 325 |
all_sel.extend(list(arr))
|
| 326 |
all_sel = sorted(list(dict.fromkeys(all_sel)))
|
| 327 |
|
|
|
|
|
|
|
| 328 |
if existing_df is None or not isinstance(existing_df, pd.DataFrame) or len(existing_df) == 0:
|
| 329 |
-
return pd.DataFrame([[c, 0.0] for c in all_sel], columns=["component",
|
| 330 |
|
| 331 |
dff = existing_df.copy()
|
| 332 |
-
if "component" not in dff.columns or
|
| 333 |
-
return pd.DataFrame([[c, 0.0] for c in all_sel], columns=["component",
|
| 334 |
|
| 335 |
dff["component"] = dff["component"].astype(str)
|
| 336 |
-
dff[
|
| 337 |
|
| 338 |
-
have = {c: float(v) for c, v in zip(dff["component"], dff[
|
| 339 |
rows = [[c, have.get(c, 0.0)] for c in all_sel]
|
| 340 |
-
return pd.DataFrame(rows, columns=["component",
|
| 341 |
|
| 342 |
# =========================================================
|
| 343 |
-
# Inference
|
| 344 |
# =========================================================
|
| 345 |
def _prepare_window(win_raw: np.ndarray, L: int) -> np.ndarray:
|
| 346 |
if win_raw.shape[0] < L:
|
|
@@ -368,6 +343,9 @@ def build_window_matrix(
|
|
| 368 |
for c in required:
|
| 369 |
df[c] = pd.to_numeric(df[c], errors="coerce").fillna(0.0)
|
| 370 |
|
|
|
|
|
|
|
|
|
|
| 371 |
t = df["Time"].to_numpy(dtype=float)
|
| 372 |
delta_t = np.zeros_like(t, dtype=float)
|
| 373 |
if len(t) >= 2:
|
|
@@ -392,7 +370,7 @@ def build_window_matrix(
|
|
| 392 |
rows.append([row_dict[c] for c in feature_cols])
|
| 393 |
return np.asarray(rows, dtype=np.float32)
|
| 394 |
|
| 395 |
-
def
|
| 396 |
timepoints: list,
|
| 397 |
gene_vec: list,
|
| 398 |
medium_vec: list,
|
|
@@ -401,12 +379,14 @@ def predict_next_bdo_and_suggest_rpm_pm200(
|
|
| 401 |
feature_cols: list,
|
| 402 |
gene_cols: list,
|
| 403 |
medium_cols: list,
|
| 404 |
-
seq_len: int
|
| 405 |
-
min_gain_pred: float
|
| 406 |
):
|
| 407 |
df_last = pd.DataFrame(timepoints).copy()
|
| 408 |
-
|
| 409 |
-
|
|
|
|
|
|
|
| 410 |
t_last = float(df_last.iloc[-1]["Time"])
|
| 411 |
bdo_last = float(df_last.iloc[-1]["BDO"])
|
| 412 |
|
|
@@ -415,8 +395,8 @@ def predict_next_bdo_and_suggest_rpm_pm200(
|
|
| 415 |
"decision": "NO_ACTION",
|
| 416 |
"reason": (
|
| 417 |
f"μΆμ² λμ ꡬκ°μ΄ μλλλ€.\n"
|
| 418 |
-
f"- νμ¬ Time={t_last:.1f}
|
| 419 |
-
f"- μΆμ²μ Time β₯ {START_HOURS:.1f}
|
| 420 |
),
|
| 421 |
}
|
| 422 |
|
|
@@ -433,7 +413,28 @@ def predict_next_bdo_and_suggest_rpm_pm200(
|
|
| 433 |
X_keep = _safe_transform(scaler, win, feature_cols)
|
| 434 |
pred_keep = float(model.predict(X_keep[None, :, :], verbose=0)[0, 0])
|
| 435 |
|
| 436 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 437 |
rpm_feat_idx = feature_cols.index("rpm")
|
| 438 |
|
| 439 |
best_pred = pred_keep
|
|
@@ -457,26 +458,54 @@ def predict_next_bdo_and_suggest_rpm_pm200(
|
|
| 457 |
best_pred = pred
|
| 458 |
best_rpm = float(rpm_cand)
|
| 459 |
|
| 460 |
-
|
| 461 |
-
|
| 462 |
-
if (abs(best_rpm - curr_rpm) < 1e-6) or (gain < float(min_gain_pred)):
|
| 463 |
return {
|
| 464 |
-
"decision": "
|
| 465 |
"pred_keep": round(pred_keep, 2),
|
|
|
|
|
|
|
|
|
|
|
|
|
| 466 |
}
|
| 467 |
|
| 468 |
return {
|
| 469 |
-
"decision": "
|
| 470 |
"pred_keep": round(pred_keep, 2),
|
| 471 |
-
"pred_best": round(best_pred, 2),
|
| 472 |
-
"curr_rpm": round(curr_rpm, 0),
|
| 473 |
-
"best_rpm": round(best_rpm, 0),
|
| 474 |
-
"gain": round(gain, 2),
|
| 475 |
}
|
| 476 |
|
| 477 |
# =========================================================
|
| 478 |
-
# UI
|
| 479 |
# =========================================================
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 480 |
def run_predict(
|
| 481 |
deleted_genes,
|
| 482 |
added_genes,
|
|
@@ -487,22 +516,44 @@ def run_predict(
|
|
| 487 |
salts_sel,
|
| 488 |
add_sel,
|
| 489 |
medium_table,
|
| 490 |
-
|
| 491 |
-
t2, b2, od2, rpm2, feed2, cc2,
|
| 492 |
-
t3, b3, od3, rpm3, feed3, cc3,
|
| 493 |
):
|
| 494 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 495 |
|
| 496 |
-
#
|
| 497 |
-
|
| 498 |
|
| 499 |
-
|
| 500 |
-
|
| 501 |
-
|
| 502 |
-
|
| 503 |
-
|
|
|
|
|
|
|
|
|
|
| 504 |
|
| 505 |
-
|
|
|
|
|
|
|
|
|
|
| 506 |
timepoints=timepoints,
|
| 507 |
gene_vec=gene_vec,
|
| 508 |
medium_vec=medium_vec,
|
|
@@ -515,34 +566,35 @@ def run_predict(
|
|
| 515 |
min_gain_pred=float(min_gain_pred),
|
| 516 |
)
|
| 517 |
|
| 518 |
-
if out
|
| 519 |
-
return out
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 520 |
|
| 521 |
-
if out
|
| 522 |
return (
|
| 523 |
-
"RPM λ³κ²½μ κΆμ₯
|
| 524 |
-
f"- λ€μ μμ BDO μμΈ‘ (νμ¬ RPM μ μ§): {out['pred_keep']:.2f}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 525 |
)
|
| 526 |
|
| 527 |
-
# CHANGE_RPM
|
| 528 |
return (
|
| 529 |
-
|
| 530 |
-
f"- λ€μ μμ BDO μμΈ‘ (νμ¬
|
| 531 |
-
f"- λ€μ μμ BDO μμΈ‘ (μΆμ² RPM μ μ©): {out['pred_best']:.2f} g/L\n"
|
| 532 |
-
f"- μΆμ² RPM: {out['curr_rpm']:.0f} β {out['best_rpm']:.0f} (Ξ={out['best_rpm']-out['curr_rpm']:+.0f})\n"
|
| 533 |
-
f"- μμ μ¦κ°λ: +{out['gain']:.2f} g/L"
|
| 534 |
)
|
| 535 |
|
| 536 |
-
# =========================================================
|
| 537 |
-
# UI wiring helpers
|
| 538 |
-
# =========================================================
|
| 539 |
-
def on_any_medium_select_change(carbon_sel, salts_sel, add_sel, medium_table):
|
| 540 |
-
return merge_selected_to_table(medium_table, carbon_sel, salts_sel, add_sel)
|
| 541 |
-
|
| 542 |
-
def on_preset_change(preset_label_selected):
|
| 543 |
-
carbon_sel, salts_sel, add_sel, df = apply_preset(preset_label_selected)
|
| 544 |
-
return carbon_sel, salts_sel, add_sel, df
|
| 545 |
-
|
| 546 |
# =========================================================
|
| 547 |
# UI
|
| 548 |
# =========================================================
|
|
@@ -557,7 +609,7 @@ with gr.Blocks() as demo:
|
|
| 557 |
with gr.Row():
|
| 558 |
min_gain_pred = gr.Number(
|
| 559 |
value=DEFAULT_MIN_GAIN_PRED,
|
| 560 |
-
label="
|
| 561 |
)
|
| 562 |
|
| 563 |
gr.Markdown("### Strain (genes)")
|
|
@@ -575,77 +627,52 @@ with gr.Blocks() as demo:
|
|
| 575 |
|
| 576 |
gr.Markdown("### Medium components (by category)")
|
| 577 |
with gr.Row():
|
| 578 |
-
carbon_sel = gr.Dropdown(
|
| 579 |
-
|
| 580 |
-
|
| 581 |
-
label="Carbon source",
|
| 582 |
-
value=[]
|
| 583 |
-
)
|
| 584 |
-
salts_sel = gr.Dropdown(
|
| 585 |
-
choices=DISP_SALTS,
|
| 586 |
-
multiselect=True,
|
| 587 |
-
label="Salts",
|
| 588 |
-
value=[]
|
| 589 |
-
)
|
| 590 |
-
add_sel = gr.Dropdown(
|
| 591 |
-
choices=DISP_ADD,
|
| 592 |
-
multiselect=True,
|
| 593 |
-
label="Additives",
|
| 594 |
-
value=[]
|
| 595 |
-
)
|
| 596 |
|
| 597 |
medium_table = gr.Dataframe(
|
| 598 |
-
headers=["component", "value"],
|
| 599 |
datatype=["str", "number"],
|
| 600 |
row_count=(0, "dynamic"),
|
| 601 |
col_count=(2, "fixed"),
|
| 602 |
interactive=True,
|
| 603 |
-
label="Enter values for selected components"
|
| 604 |
)
|
| 605 |
|
| 606 |
-
# preset -> sets category selections + table values
|
| 607 |
preset_name.change(
|
| 608 |
fn=on_preset_change,
|
| 609 |
inputs=[preset_name],
|
| 610 |
outputs=[carbon_sel, salts_sel, add_sel, medium_table],
|
| 611 |
)
|
| 612 |
|
| 613 |
-
|
| 614 |
-
|
| 615 |
-
|
| 616 |
-
|
| 617 |
-
|
| 618 |
-
|
| 619 |
-
|
| 620 |
-
|
| 621 |
-
|
| 622 |
-
|
| 623 |
-
|
| 624 |
-
|
| 625 |
-
|
| 626 |
-
|
| 627 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 628 |
)
|
| 629 |
|
| 630 |
-
gr.
|
| 631 |
-
|
| 632 |
-
def tp_row(title):
|
| 633 |
-
gr.Markdown(f"**{title}**")
|
| 634 |
-
with gr.Row():
|
| 635 |
-
t = gr.Number(label="Time (h)", value=0.0)
|
| 636 |
-
b = gr.Number(label="BDO (g/L)", value=0.0)
|
| 637 |
-
od = gr.Number(label="OD", value=0.0)
|
| 638 |
-
rpm = gr.Number(label="rpm", value=0.0)
|
| 639 |
-
feed = gr.Number(label="Feed", value=0.0)
|
| 640 |
-
cc = gr.Number(label="CarbonContents", value=0.0)
|
| 641 |
-
return t, b, od, rpm, feed, cc
|
| 642 |
-
|
| 643 |
-
t1, b1, od1, rpm1, feed1, cc1 = tp_row("Timepoint 1 (oldest)")
|
| 644 |
-
t2, b2, od2, rpm2, feed2, cc2 = tp_row("Timepoint 2")
|
| 645 |
-
t3, b3, od3, rpm3, feed3, cc3 = tp_row("Timepoint 3 (latest)")
|
| 646 |
|
| 647 |
btn = gr.Button("Predict & Suggest")
|
| 648 |
-
out_box = gr.Textbox(label="Output", lines=
|
| 649 |
|
| 650 |
btn.click(
|
| 651 |
fn=run_predict,
|
|
@@ -655,9 +682,7 @@ with gr.Blocks() as demo:
|
|
| 655 |
preset_name,
|
| 656 |
carbon_sel, salts_sel, add_sel,
|
| 657 |
medium_table,
|
| 658 |
-
|
| 659 |
-
t2, b2, od2, rpm2, feed2, cc2,
|
| 660 |
-
t3, b3, od3, rpm3, feed3, cc3,
|
| 661 |
],
|
| 662 |
outputs=[out_box],
|
| 663 |
)
|
|
|
|
| 1 |
import os
|
|
|
|
| 2 |
import numpy as np
|
| 3 |
import pandas as pd
|
| 4 |
import gradio as gr
|
|
|
|
| 6 |
import tensorflow as tf
|
| 7 |
|
| 8 |
# =========================================================
|
| 9 |
+
# Settings / Constants
|
| 10 |
# =========================================================
|
| 11 |
BEST_L = 3
|
| 12 |
CC_COL = "CarbonContents"
|
| 13 |
|
| 14 |
+
# Eligibility rule for running recommendation
|
| 15 |
+
START_HOURS = 12.0
|
| 16 |
+
START_BDO = 20.0
|
| 17 |
+
|
| 18 |
+
# Feed logic
|
| 19 |
+
CC_THRESHOLD = 20.0
|
| 20 |
+
FEED_VALUE = 50.0 # unit: g/L (assumption for UI display)
|
| 21 |
+
|
| 22 |
+
# RPM search (ONLY Β±200 around current rpm)
|
| 23 |
RPM_RANGE = (150, 650)
|
| 24 |
+
RPM_DELTAS = (-200, -150, -100, -50, 50, 100, 150, 200)
|
| 25 |
FORBID_BOUNDARY = True
|
| 26 |
|
| 27 |
DEFAULT_MIN_GAIN_PRED = 0.5
|
| 28 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
# =========================================================
|
| 30 |
# Gene set (must match training)
|
| 31 |
# =========================================================
|
|
|
|
| 68 |
medium_cols_carbon = list(Carbon_source_dict.keys())
|
| 69 |
medium_cols_salts = list(Base_dict.keys()) # salts = base
|
| 70 |
medium_cols_add = list(Additional_dict.keys()) # additives
|
|
|
|
| 71 |
medium_cols = medium_cols_carbon + medium_cols_salts + medium_cols_add
|
| 72 |
|
| 73 |
+
# =========================================================
|
| 74 |
+
# Units (UI display)
|
| 75 |
+
# =========================================================
|
| 76 |
+
UNIT_MEDIUM_DEFAULT = "g/L"
|
| 77 |
+
UNIT_TIME = "h"
|
| 78 |
+
UNIT_BDO = "g/L"
|
| 79 |
+
UNIT_OD = "AU"
|
| 80 |
+
UNIT_RPM = "rpm"
|
| 81 |
+
UNIT_FEED = "g/L"
|
| 82 |
+
UNIT_CC = "g/L"
|
| 83 |
+
|
| 84 |
# =========================================================
|
| 85 |
# Medium display name mapping (hide "_i" in UI)
|
| 86 |
# =========================================================
|
|
|
|
| 116 |
model = tf.keras.models.load_model(MODEL_PATH, compile=False)
|
| 117 |
|
| 118 |
# =========================================================
|
| 119 |
+
# Presets (Top-5 combos from CSV if exists, else fallback)
|
|
|
|
| 120 |
# =========================================================
|
| 121 |
def _safe_read_csv(path: str):
|
| 122 |
try:
|
|
|
|
| 127 |
return None
|
| 128 |
|
| 129 |
def build_top5_presets_from_csv(csv_path: str, top_k: int = 5):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 130 |
df = _safe_read_csv(csv_path)
|
| 131 |
+
if df is None or "BatchNo" not in df.columns:
|
|
|
|
|
|
|
|
|
|
| 132 |
return {}
|
| 133 |
|
|
|
|
| 134 |
for c in medium_cols:
|
| 135 |
if c not in df.columns:
|
| 136 |
df[c] = 0.0
|
|
|
|
| 145 |
else:
|
| 146 |
first_rows = df.groupby("BatchNo", as_index=False).first()
|
| 147 |
|
|
|
|
| 148 |
first_rows[medium_cols] = first_rows[medium_cols].apply(pd.to_numeric, errors="coerce").fillna(0.0)
|
| 149 |
|
|
|
|
| 150 |
def signature(row):
|
| 151 |
nz = [c for c in medium_cols if float(row[c]) != 0.0]
|
| 152 |
nz.sort()
|
| 153 |
return "|".join(nz)
|
| 154 |
|
| 155 |
first_rows["_sig"] = first_rows.apply(signature, axis=1)
|
|
|
|
| 156 |
first_rows = first_rows[first_rows["_sig"].astype(str).str.len() > 0].copy()
|
| 157 |
if len(first_rows) == 0:
|
| 158 |
return {}
|
| 159 |
|
| 160 |
+
top_sigs = first_rows["_sig"].value_counts().head(top_k).index.tolist()
|
|
|
|
| 161 |
|
| 162 |
presets = {}
|
| 163 |
for i, sig in enumerate(top_sigs, start=1):
|
|
|
|
| 167 |
for c in comps:
|
| 168 |
vals[c] = float(np.median(sub[c].to_numpy(dtype=float)))
|
| 169 |
presets[f"TopCombo{i}"] = vals
|
|
|
|
| 170 |
return presets
|
| 171 |
|
| 172 |
def fallback_example_presets():
|
|
|
|
|
|
|
| 173 |
ex = {
|
| 174 |
"Example1": {"Glucose_i": 1.0, "YE_i": 10.0, "K2HPO4_i": 3.0, "KH2PO4_i": 1.0, "MgSO4Β·7H2O_i": 0.5},
|
| 175 |
"Example2": {"Glucose_i": 1.0, "YE_i": 5.0, "NaCl_i": 1.0, "K2HPO4_i": 3.0, "MgSO4Β·7H2O_i": 0.5},
|
|
|
|
| 177 |
"Example4": {"Glucose_i": 1.0, "YE_i": 10.0, "K2HPO4_i": 3.0, "MgSO4Β·7H2O_i": 0.5, "CSL_i": 5.0},
|
| 178 |
"Example5": {"Glucose_i": 1.0, "YE_i": 10.0, "K2HPO4_i": 3.0, "MgSO4Β·7H2O_i": 0.5, "IPTG_i": 0.1},
|
| 179 |
}
|
|
|
|
| 180 |
cleaned = {}
|
| 181 |
for k, d in ex.items():
|
| 182 |
dd = {}
|
|
|
|
| 186 |
cleaned[k] = dd
|
| 187 |
return cleaned
|
| 188 |
|
|
|
|
| 189 |
PRESETS = build_top5_presets_from_csv("BDO_Data_cleaned_with_Feed_deltat.csv", top_k=5)
|
| 190 |
if not PRESETS:
|
| 191 |
PRESETS = fallback_example_presets()
|
|
|
|
| 197 |
items = []
|
| 198 |
for internal, v in d.items():
|
| 199 |
disp = INTERNAL_TO_DISP.get(internal, internal)
|
| 200 |
+
fv = float(v)
|
|
|
|
|
|
|
|
|
|
| 201 |
items.append((disp, fv))
|
| 202 |
items = sorted(items, key=lambda x: x[0])
|
| 203 |
+
inside = "; ".join([f"{n}={val:g} {UNIT_MEDIUM_DEFAULT}" for n, val in items])
|
| 204 |
return f"{preset_key} ({inside})"
|
| 205 |
|
| 206 |
PRESET_LABELS = {preset_label(k): k for k in PRESETS.keys()}
|
|
|
|
| 210 |
preset_key = PRESET_LABELS.get(preset_label_selected, "")
|
| 211 |
d = PRESETS.get(preset_key, {})
|
| 212 |
if not isinstance(d, dict) or len(d) == 0:
|
| 213 |
+
return [], [], [], pd.DataFrame(columns=["component", f"value ({UNIT_MEDIUM_DEFAULT})"])
|
| 214 |
|
|
|
|
| 215 |
carbon_sel, salts_sel, add_sel = [], [], []
|
| 216 |
rows = []
|
|
|
|
| 217 |
for internal, v in d.items():
|
| 218 |
disp = INTERNAL_TO_DISP.get(internal, internal)
|
| 219 |
+
fv = float(v)
|
|
|
|
|
|
|
|
|
|
| 220 |
rows.append([disp, fv])
|
| 221 |
|
| 222 |
if internal in medium_cols_carbon:
|
|
|
|
| 226 |
elif internal in medium_cols_add:
|
| 227 |
add_sel.append(disp)
|
| 228 |
|
| 229 |
+
df = pd.DataFrame(rows, columns=["component", f"value ({UNIT_MEDIUM_DEFAULT})"])
|
| 230 |
+
return sorted(set(carbon_sel)), sorted(set(salts_sel)), sorted(set(add_sel)), df
|
|
|
|
|
|
|
|
|
|
| 231 |
|
| 232 |
# =========================================================
|
| 233 |
# Gene embedding helpers
|
|
|
|
| 251 |
for g in deleted:
|
| 252 |
if g in g2i:
|
| 253 |
emb[g2i[g]] = -1.0
|
|
|
|
| 254 |
for g in added:
|
| 255 |
if g in g2i:
|
| 256 |
emb[g2i[g]] += 1.0
|
|
|
|
| 257 |
for item in replaced:
|
| 258 |
if "->" in item:
|
| 259 |
old, new = [x.strip() for x in item.split("->", 1)]
|
|
|
|
| 264 |
else:
|
| 265 |
if item in g2i:
|
| 266 |
emb[g2i[item]] = -1.0
|
|
|
|
| 267 |
return emb.tolist()
|
| 268 |
|
| 269 |
# =========================================================
|
|
|
|
| 275 |
return vec.tolist()
|
| 276 |
|
| 277 |
dff = df.copy()
|
| 278 |
+
value_col = f"value ({UNIT_MEDIUM_DEFAULT})"
|
| 279 |
+
if "component" not in dff.columns or value_col not in dff.columns:
|
| 280 |
return vec.tolist()
|
| 281 |
|
| 282 |
dff["component"] = dff["component"].astype(str)
|
| 283 |
+
dff[value_col] = pd.to_numeric(dff[value_col], errors="coerce").fillna(0.0)
|
| 284 |
|
| 285 |
m2i = {m: i for i, m in enumerate(medium_cols)}
|
|
|
|
| 286 |
for _, r in dff.iterrows():
|
| 287 |
disp = r["component"]
|
| 288 |
internal = DISP_TO_INTERNAL.get(disp, None)
|
| 289 |
if internal is None:
|
| 290 |
continue
|
| 291 |
+
vec[m2i[internal]] = float(r[value_col])
|
|
|
|
|
|
|
| 292 |
return vec.tolist()
|
| 293 |
|
| 294 |
def merge_selected_to_table(existing_df, carbon_sel, salts_sel, add_sel):
|
|
|
|
|
|
|
|
|
|
|
|
|
| 295 |
all_sel = []
|
| 296 |
for arr in [carbon_sel, salts_sel, add_sel]:
|
| 297 |
if arr:
|
| 298 |
all_sel.extend(list(arr))
|
| 299 |
all_sel = sorted(list(dict.fromkeys(all_sel)))
|
| 300 |
|
| 301 |
+
value_col = f"value ({UNIT_MEDIUM_DEFAULT})"
|
| 302 |
+
|
| 303 |
if existing_df is None or not isinstance(existing_df, pd.DataFrame) or len(existing_df) == 0:
|
| 304 |
+
return pd.DataFrame([[c, 0.0] for c in all_sel], columns=["component", value_col])
|
| 305 |
|
| 306 |
dff = existing_df.copy()
|
| 307 |
+
if "component" not in dff.columns or value_col not in dff.columns:
|
| 308 |
+
return pd.DataFrame([[c, 0.0] for c in all_sel], columns=["component", value_col])
|
| 309 |
|
| 310 |
dff["component"] = dff["component"].astype(str)
|
| 311 |
+
dff[value_col] = pd.to_numeric(dff[value_col], errors="coerce").fillna(0.0)
|
| 312 |
|
| 313 |
+
have = {c: float(v) for c, v in zip(dff["component"], dff[value_col])}
|
| 314 |
rows = [[c, have.get(c, 0.0)] for c in all_sel]
|
| 315 |
+
return pd.DataFrame(rows, columns=["component", value_col])
|
| 316 |
|
| 317 |
# =========================================================
|
| 318 |
+
# Inference helpers
|
| 319 |
# =========================================================
|
| 320 |
def _prepare_window(win_raw: np.ndarray, L: int) -> np.ndarray:
|
| 321 |
if win_raw.shape[0] < L:
|
|
|
|
| 343 |
for c in required:
|
| 344 |
df[c] = pd.to_numeric(df[c], errors="coerce").fillna(0.0)
|
| 345 |
|
| 346 |
+
# sort by Time (important when user adds more points)
|
| 347 |
+
df = df.sort_values("Time").reset_index(drop=True)
|
| 348 |
+
|
| 349 |
t = df["Time"].to_numpy(dtype=float)
|
| 350 |
delta_t = np.zeros_like(t, dtype=float)
|
| 351 |
if len(t) >= 2:
|
|
|
|
| 370 |
rows.append([row_dict[c] for c in feature_cols])
|
| 371 |
return np.asarray(rows, dtype=np.float32)
|
| 372 |
|
| 373 |
+
def predict_with_optional_feed_and_rpm(
|
| 374 |
timepoints: list,
|
| 375 |
gene_vec: list,
|
| 376 |
medium_vec: list,
|
|
|
|
| 379 |
feature_cols: list,
|
| 380 |
gene_cols: list,
|
| 381 |
medium_cols: list,
|
| 382 |
+
seq_len: int,
|
| 383 |
+
min_gain_pred: float,
|
| 384 |
):
|
| 385 |
df_last = pd.DataFrame(timepoints).copy()
|
| 386 |
+
for c in ["Time", "BDO", "Feed", "CarbonContents", "rpm"]:
|
| 387 |
+
df_last[c] = pd.to_numeric(df_last[c], errors="coerce").fillna(0.0)
|
| 388 |
+
df_last = df_last.sort_values("Time").reset_index(drop=True)
|
| 389 |
+
|
| 390 |
t_last = float(df_last.iloc[-1]["Time"])
|
| 391 |
bdo_last = float(df_last.iloc[-1]["BDO"])
|
| 392 |
|
|
|
|
| 395 |
"decision": "NO_ACTION",
|
| 396 |
"reason": (
|
| 397 |
f"μΆμ² λμ ꡬκ°μ΄ μλλλ€.\n"
|
| 398 |
+
f"- νμ¬ Time={t_last:.1f} {UNIT_TIME}, BDO={bdo_last:.2f} {UNIT_BDO}\n"
|
| 399 |
+
f"- μΆμ²μ Time β₯ {START_HOURS:.1f} {UNIT_TIME} AND BDO β₯ {START_BDO:.1f} {UNIT_BDO} μμλ§ μνν©λλ€."
|
| 400 |
),
|
| 401 |
}
|
| 402 |
|
|
|
|
| 413 |
X_keep = _safe_transform(scaler, win, feature_cols)
|
| 414 |
pred_keep = float(model.predict(X_keep[None, :, :], verbose=0)[0, 0])
|
| 415 |
|
| 416 |
+
cc_last = float(df_last.iloc[-1]["CarbonContents"])
|
| 417 |
+
feed_last = float(df_last.iloc[-1]["Feed"])
|
| 418 |
+
feed_feat_idx = feature_cols.index("Feed")
|
| 419 |
+
|
| 420 |
+
if (cc_last < CC_THRESHOLD) and (abs(feed_last) < 1e-9):
|
| 421 |
+
win_feed = win.copy()
|
| 422 |
+
win_feed[-1, feed_feat_idx] = float(FEED_VALUE)
|
| 423 |
+
X_feed = _safe_transform(scaler, win_feed, feature_cols)
|
| 424 |
+
pred_feed = float(model.predict(X_feed[None, :, :], verbose=0)[0, 0])
|
| 425 |
+
|
| 426 |
+
gain_feed = float(pred_feed - pred_keep)
|
| 427 |
+
if gain_feed >= float(min_gain_pred):
|
| 428 |
+
return {
|
| 429 |
+
"decision": "FEED",
|
| 430 |
+
"pred_keep": round(pred_keep, 2),
|
| 431 |
+
"pred_after": round(pred_feed, 2),
|
| 432 |
+
"gain": round(gain_feed, 2),
|
| 433 |
+
"feed_from": round(feed_last, 2),
|
| 434 |
+
"feed_to": round(FEED_VALUE, 2),
|
| 435 |
+
}
|
| 436 |
+
|
| 437 |
+
curr_rpm = float(df_last.iloc[-1]["rpm"])
|
| 438 |
rpm_feat_idx = feature_cols.index("rpm")
|
| 439 |
|
| 440 |
best_pred = pred_keep
|
|
|
|
| 458 |
best_pred = pred
|
| 459 |
best_rpm = float(rpm_cand)
|
| 460 |
|
| 461 |
+
gain_rpm = float(best_pred - pred_keep)
|
| 462 |
+
if (abs(best_rpm - curr_rpm) > 1e-6) and (gain_rpm >= float(min_gain_pred)):
|
|
|
|
| 463 |
return {
|
| 464 |
+
"decision": "CHANGE_RPM",
|
| 465 |
"pred_keep": round(pred_keep, 2),
|
| 466 |
+
"pred_after": round(best_pred, 2),
|
| 467 |
+
"gain": round(gain_rpm, 2),
|
| 468 |
+
"rpm_from": round(curr_rpm, 0),
|
| 469 |
+
"rpm_to": round(best_rpm, 0),
|
| 470 |
}
|
| 471 |
|
| 472 |
return {
|
| 473 |
+
"decision": "KEEP",
|
| 474 |
"pred_keep": round(pred_keep, 2),
|
|
|
|
|
|
|
|
|
|
|
|
|
| 475 |
}
|
| 476 |
|
| 477 |
# =========================================================
|
| 478 |
+
# UI callbacks
|
| 479 |
# =========================================================
|
| 480 |
+
def on_any_medium_select_change(carbon_sel, salts_sel, add_sel, medium_table):
|
| 481 |
+
return merge_selected_to_table(medium_table, carbon_sel, salts_sel, add_sel)
|
| 482 |
+
|
| 483 |
+
def on_preset_change(preset_label_selected):
|
| 484 |
+
carbon_sel, salts_sel, add_sel, df = apply_preset(preset_label_selected)
|
| 485 |
+
return carbon_sel, salts_sel, add_sel, df
|
| 486 |
+
|
| 487 |
+
def init_timepoints_df():
|
| 488 |
+
# μ΅μ 3κ° row μ 곡
|
| 489 |
+
cols = [f"Time ({UNIT_TIME})",
|
| 490 |
+
f"BDO ({UNIT_BDO})",
|
| 491 |
+
f"OD ({UNIT_OD})",
|
| 492 |
+
f"rpm ({UNIT_RPM})",
|
| 493 |
+
f"Feed ({UNIT_FEED})",
|
| 494 |
+
f"CarbonContents ({UNIT_CC})"]
|
| 495 |
+
data = [
|
| 496 |
+
[0.0, 0.0, 0.0, 0.0, 0.0, 0.0],
|
| 497 |
+
[0.0, 0.0, 0.0, 0.0, 0.0, 0.0],
|
| 498 |
+
[0.0, 0.0, 0.0, 0.0, 0.0, 0.0],
|
| 499 |
+
]
|
| 500 |
+
return pd.DataFrame(data, columns=cols)
|
| 501 |
+
|
| 502 |
+
def add_timepoint_row(df):
|
| 503 |
+
if df is None or not isinstance(df, pd.DataFrame):
|
| 504 |
+
df = init_timepoints_df()
|
| 505 |
+
df2 = df.copy()
|
| 506 |
+
df2.loc[len(df2)] = [0.0, 0.0, 0.0, 0.0, 0.0, 0.0]
|
| 507 |
+
return df2
|
| 508 |
+
|
| 509 |
def run_predict(
|
| 510 |
deleted_genes,
|
| 511 |
added_genes,
|
|
|
|
| 516 |
salts_sel,
|
| 517 |
add_sel,
|
| 518 |
medium_table,
|
| 519 |
+
timepoints_df,
|
|
|
|
|
|
|
| 520 |
):
|
| 521 |
+
# validate timepoints
|
| 522 |
+
if timepoints_df is None or not isinstance(timepoints_df, pd.DataFrame) or len(timepoints_df) < BEST_L:
|
| 523 |
+
return f"Timepointsλ μ΅μ {BEST_L}κ° μ΄μ μ
λ ₯ν΄μΌ ν©λλ€."
|
| 524 |
+
|
| 525 |
+
df = timepoints_df.copy()
|
| 526 |
+
|
| 527 |
+
# column mapping from UI to internal keys
|
| 528 |
+
col_map = {
|
| 529 |
+
f"Time ({UNIT_TIME})": "Time",
|
| 530 |
+
f"BDO ({UNIT_BDO})": "BDO",
|
| 531 |
+
f"OD ({UNIT_OD})": "OD",
|
| 532 |
+
f"rpm ({UNIT_RPM})": "rpm",
|
| 533 |
+
f"Feed ({UNIT_FEED})": "Feed",
|
| 534 |
+
f"CarbonContents ({UNIT_CC})": "CarbonContents",
|
| 535 |
+
}
|
| 536 |
+
for ui_c, in_c in col_map.items():
|
| 537 |
+
if ui_c not in df.columns:
|
| 538 |
+
return f"Timepoints νμ '{ui_c}' 컬λΌμ΄ μμ΅λλ€."
|
| 539 |
+
df[in_c] = pd.to_numeric(df[ui_c], errors="coerce").fillna(0.0)
|
| 540 |
|
| 541 |
+
# keep only internal columns
|
| 542 |
+
df = df[list(col_map.values())].copy()
|
| 543 |
|
| 544 |
+
# sort by Time and drop rows with all zeros (optional)
|
| 545 |
+
df = df.sort_values("Time").reset_index(drop=True)
|
| 546 |
+
|
| 547 |
+
# require at least BEST_L rows AFTER sorting
|
| 548 |
+
if len(df) < BEST_L:
|
| 549 |
+
return f"Timepointsλ μ΅μ {BEST_L}κ° μ΄μ μ
λ ₯ν΄μΌ ν©λλ€."
|
| 550 |
+
|
| 551 |
+
timepoints = df.to_dict(orient="records")
|
| 552 |
|
| 553 |
+
gene_vec = build_gene_embedding(deleted_genes, added_genes, replaced_genes)
|
| 554 |
+
medium_vec = build_medium_vector_from_df(medium_table)
|
| 555 |
+
|
| 556 |
+
out = predict_with_optional_feed_and_rpm(
|
| 557 |
timepoints=timepoints,
|
| 558 |
gene_vec=gene_vec,
|
| 559 |
medium_vec=medium_vec,
|
|
|
|
| 566 |
min_gain_pred=float(min_gain_pred),
|
| 567 |
)
|
| 568 |
|
| 569 |
+
if out["decision"] == "NO_ACTION":
|
| 570 |
+
return out["reason"]
|
| 571 |
+
|
| 572 |
+
if out["decision"] == "FEED":
|
| 573 |
+
return (
|
| 574 |
+
f"μμ κ°μ νμ΄ {float(min_gain_pred):.2f} {UNIT_BDO} μ΄μμ΄λΌ Feedingμ κΆμ₯ν©λλ€.\n"
|
| 575 |
+
f"- λ€μ μμ BDO μμΈ‘ (νμ¬ μ‘°κ±΄ μ μ§): {out['pred_keep']:.2f} {UNIT_BDO}\n"
|
| 576 |
+
f"- λ€μ μμ BDO μμΈ‘ (Feeding μ μ©): {out['pred_after']:.2f} {UNIT_BDO}\n"
|
| 577 |
+
f"- μΆμ² Feed: {out['feed_from']:.2f} β {out['feed_to']:.2f} ({UNIT_FEED})\n"
|
| 578 |
+
f"- μμ μ¦κ°λ: +{out['gain']:.2f} {UNIT_BDO}\n\n"
|
| 579 |
+
f"(Feeding 쑰건: CarbonContents < {CC_THRESHOLD:.1f} {UNIT_CC} AND Feed = 0)"
|
| 580 |
+
)
|
| 581 |
|
| 582 |
+
if out["decision"] == "CHANGE_RPM":
|
| 583 |
return (
|
| 584 |
+
f"μμ κ°μ νμ΄ {float(min_gain_pred):.2f} {UNIT_BDO} μ΄μμ΄λΌ RPM λ³κ²½μ κΆμ₯ν©λλ€.\n"
|
| 585 |
+
f"- λ€μ μμ BDO μμΈ‘ (νμ¬ RPM μ μ§): {out['pred_keep']:.2f} {UNIT_BDO}\n"
|
| 586 |
+
f"- λ€μ μμ BDO μμΈ‘ (μΆμ² RPM μ μ©): {out['pred_after']:.2f} {UNIT_BDO}\n"
|
| 587 |
+
f"- μΆμ² RPM: {out['rpm_from']:.0f} β {out['rpm_to']:.0f} (Ξ={out['rpm_to']-out['rpm_from']:+.0f} {UNIT_RPM})\n"
|
| 588 |
+
f"- μμ μ¦κ°λ: +{out['gain']:.2f} {UNIT_BDO}\n\n"
|
| 589 |
+
"μ£Όμ:\n"
|
| 590 |
+
"- RPM νμμ νμ¬ RPM κΈ°μ€ Β±200 λ²μμμλ§ μνλ©λλ€."
|
| 591 |
)
|
| 592 |
|
|
|
|
| 593 |
return (
|
| 594 |
+
"λ³κ²½μ κΆμ₯νμ§ μμ΅λλ€.\n"
|
| 595 |
+
f"- λ€μ μμ BDO μμΈ‘ (νμ¬ μ‘°κ±΄ μ μ§): {out['pred_keep']:.2f} {UNIT_BDO}"
|
|
|
|
|
|
|
|
|
|
| 596 |
)
|
| 597 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 598 |
# =========================================================
|
| 599 |
# UI
|
| 600 |
# =========================================================
|
|
|
|
| 609 |
with gr.Row():
|
| 610 |
min_gain_pred = gr.Number(
|
| 611 |
value=DEFAULT_MIN_GAIN_PRED,
|
| 612 |
+
label=f"μ΅μ μμ κ°μ ν ({UNIT_BDO})"
|
| 613 |
)
|
| 614 |
|
| 615 |
gr.Markdown("### Strain (genes)")
|
|
|
|
| 627 |
|
| 628 |
gr.Markdown("### Medium components (by category)")
|
| 629 |
with gr.Row():
|
| 630 |
+
carbon_sel = gr.Dropdown(choices=DISP_CARBON, multiselect=True, label="Carbon source", value=[])
|
| 631 |
+
salts_sel = gr.Dropdown(choices=DISP_SALTS, multiselect=True, label="Salts", value=[])
|
| 632 |
+
add_sel = gr.Dropdown(choices=DISP_ADD, multiselect=True, label="Additives", value=[])
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 633 |
|
| 634 |
medium_table = gr.Dataframe(
|
| 635 |
+
headers=["component", f"value ({UNIT_MEDIUM_DEFAULT})"],
|
| 636 |
datatype=["str", "number"],
|
| 637 |
row_count=(0, "dynamic"),
|
| 638 |
col_count=(2, "fixed"),
|
| 639 |
interactive=True,
|
| 640 |
+
label=f"Enter values for selected components (unit: {UNIT_MEDIUM_DEFAULT})"
|
| 641 |
)
|
| 642 |
|
|
|
|
| 643 |
preset_name.change(
|
| 644 |
fn=on_preset_change,
|
| 645 |
inputs=[preset_name],
|
| 646 |
outputs=[carbon_sel, salts_sel, add_sel, medium_table],
|
| 647 |
)
|
| 648 |
|
| 649 |
+
carbon_sel.change(fn=on_any_medium_select_change, inputs=[carbon_sel, salts_sel, add_sel, medium_table], outputs=[medium_table])
|
| 650 |
+
salts_sel.change(fn=on_any_medium_select_change, inputs=[carbon_sel, salts_sel, add_sel, medium_table], outputs=[medium_table])
|
| 651 |
+
add_sel.change(fn=on_any_medium_select_change, inputs=[carbon_sel, salts_sel, add_sel, medium_table], outputs=[medium_table])
|
| 652 |
+
|
| 653 |
+
gr.Markdown("### Timepoints")
|
| 654 |
+
timepoints_df = gr.Dataframe(
|
| 655 |
+
value=init_timepoints_df(),
|
| 656 |
+
headers=[
|
| 657 |
+
f"Time ({UNIT_TIME})",
|
| 658 |
+
f"BDO ({UNIT_BDO})",
|
| 659 |
+
f"OD ({UNIT_OD})",
|
| 660 |
+
f"rpm ({UNIT_RPM})",
|
| 661 |
+
f"Feed ({UNIT_FEED})",
|
| 662 |
+
f"CarbonContents ({UNIT_CC})",
|
| 663 |
+
],
|
| 664 |
+
datatype=["number", "number", "number", "number", "number", "number"],
|
| 665 |
+
row_count=(3, "dynamic"),
|
| 666 |
+
col_count=(6, "fixed"),
|
| 667 |
+
interactive=True,
|
| 668 |
+
label=f"Enter timepoints (need β₯ {BEST_L} rows; sorted by Time internally)"
|
| 669 |
)
|
| 670 |
|
| 671 |
+
add_btn = gr.Button("+ Add timepoint")
|
| 672 |
+
add_btn.click(fn=add_timepoint_row, inputs=[timepoints_df], outputs=[timepoints_df])
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 673 |
|
| 674 |
btn = gr.Button("Predict & Suggest")
|
| 675 |
+
out_box = gr.Textbox(label="Output", lines=12)
|
| 676 |
|
| 677 |
btn.click(
|
| 678 |
fn=run_predict,
|
|
|
|
| 682 |
preset_name,
|
| 683 |
carbon_sel, salts_sel, add_sel,
|
| 684 |
medium_table,
|
| 685 |
+
timepoints_df,
|
|
|
|
|
|
|
| 686 |
],
|
| 687 |
outputs=[out_box],
|
| 688 |
)
|