{"record":{"id":"6be3867dea3192f3","repo":"microsoft/qlib","slug":"model-is-not-fitted-yet-6be386","errorCode":null,"errorMessage":"model is not fitted yet!","messagePattern":"model is not fitted yet!","errorType":"validation","errorClass":"ValueError","httpStatus":null,"severity":"error","filePath":"qlib/contrib/model/pytorch_lstm_ts.py","lineNumber":278,"sourceCode":"                stop_steps = 0\n                best_epoch = step\n                best_param = copy.deepcopy(self.LSTM_model.state_dict())\n            else:\n                stop_steps += 1\n                if stop_steps >= self.early_stop:\n                    self.logger.info(\"early stop\")\n                    break\n\n        self.logger.info(\"best score: %.6lf @ %d\" % (best_score, best_epoch))\n        self.LSTM_model.load_state_dict(best_param)\n        torch.save(best_param, save_path)\n\n        if self.use_gpu:\n            torch.cuda.empty_cache()\n\n    def predict(self, dataset):\n        if not self.fitted:\n            raise ValueError(\"model is not fitted yet!\")\n\n        dl_test = dataset.prepare(\"test\", col_set=[\"feature\", \"label\"], data_key=DataHandlerLP.DK_I)\n        dl_test.config(fillna_type=\"ffill+bfill\")\n        test_loader = DataLoader(dl_test, batch_size=self.batch_size, num_workers=self.n_jobs)\n        self.LSTM_model.eval()\n        preds = []\n\n        for data in test_loader:\n            feature = data[:, :, 0:-1].to(self.device)\n\n            with torch.no_grad():\n                pred = self.LSTM_model(feature.float()).detach().cpu().numpy()\n\n            preds.append(pred)\n\n        return pd.Series(np.concatenate(preds), index=dl_test.get_index())\n\n","sourceCodeStart":260,"sourceCodeEnd":296,"githubUrl":"https://github.com/microsoft/qlib/blob/79633dd9506ea689e5400dea0197717b5b3d74b7/qlib/contrib/model/pytorch_lstm_ts.py#L260-L296","documentation":"The TS LSTM predict() checks self.fitted at entry; the flag flips True only at the very end of a successful fit(). Predicting beforehand raises ValueError('model is not fitted yet!') — no test data is prepared and no model eval occurs.","triggerScenarios":"model.predict(dataset) where fit() never completed: never called, interrupted, or failed on empty data / NaNs / device errors before setting fitted.","commonSituations":"Pipeline scripts proceeding to backtest after a training step errored; resuming a session with a fresh model object; notebook cell ordering mistakes.","solutions":["Complete model.fit(dataset, evals_result) first; check for the 'best score: ... @ epoch' log line as confirmation.","Fix any prior fit failure (its exception is the root cause; fitted remains False otherwise).","Restoring a checkpoint: model.LSTM_model.load_state_dict(torch.load(save_path)); model.fitted = True; then call predict."],"exampleFix":"# before\nmodel = LSTMModel(...)\nmodel.predict(dataset)  # ValueError: model is not fitted yet!\n\n# after\nmodel = LSTMModel(...)\nmodel.fit(dataset, evals_result)\nmodel.predict(dataset)","handlingStrategy":"validation","validationCode":"if not model.fitted:\n    raise RuntimeError(\"TS LSTM not fitted; run fit() first\")\nmodel.predict(dataset)","typeGuard":"def is_fitted(model) -> bool:\n    return bool(getattr(model, \"fitted\", False))","tryCatchPattern":"try:\n    model.predict(dataset)\nexcept ValueError as e:\n    if \"not fitted\" in str(e):\n        model.fit(dataset, evals_result)\n        model.predict(dataset)\n    else:\n        raise","preventionTips":["Check model.fitted before predict in all workflow code.","For checkpoint restore: load_state_dict then set fitted = True.","Halt pipelines when fit() fails so predict never executes."],"tags":["pytorch","qlib","model-lifecycle","lstm","validation"],"backgroundTag":null,"analyzedSha":"79633dd9506ea689e5400dea0197717b5b3d74b7","analyzedAt":"2026-08-15T07:01:27.511Z","schemaVersion":2},"datasetVersion":"2026-08-15T17:31:12.345Z"}