{"record":{"id":"7ffbe2ccca6a3391","repo":"microsoft/qlib","slug":"optimizer-is-not-supported-7ffbe2","errorCode":null,"errorMessage":"optimizer {} is not supported!","messagePattern":"optimizer (.+?) is not supported!","errorType":"exception","errorClass":"NotImplementedError","httpStatus":null,"severity":"error","filePath":"qlib/contrib/model/pytorch_lstm_ts.py","lineNumber":128,"sourceCode":"            )\n        )\n\n        if self.seed is not None:\n            np.random.seed(self.seed)\n            torch.manual_seed(self.seed)\n\n        self.LSTM_model = LSTMModel(\n            d_feat=self.d_feat,\n            hidden_size=self.hidden_size,\n            num_layers=self.num_layers,\n            dropout=self.dropout,\n        ).to(self.device)\n        if optimizer.lower() == \"adam\":\n            self.train_optimizer = optim.Adam(self.LSTM_model.parameters(), lr=self.lr)\n        elif optimizer.lower() == \"gd\":\n            self.train_optimizer = optim.SGD(self.LSTM_model.parameters(), lr=self.lr)\n        else:\n            raise NotImplementedError(\"optimizer {} is not supported!\".format(optimizer))\n\n        self.fitted = False\n        self.LSTM_model.to(self.device)\n\n    @property\n    def use_gpu(self):\n        return self.device != torch.device(\"cpu\")\n\n    def mse(self, pred, label, weight):\n        loss = weight * (pred - label) ** 2\n        return torch.mean(loss)\n\n    def loss_fn(self, pred, label, weight):\n        mask = ~torch.isnan(label)\n\n        if weight is None:\n            weight = torch.ones_like(label)\n","sourceCodeStart":110,"sourceCodeEnd":146,"githubUrl":"https://github.com/microsoft/qlib/blob/79633dd9506ea689e5400dea0197717b5b3d74b7/qlib/contrib/model/pytorch_lstm_ts.py#L110-L146","documentation":"The time-series LSTM model's constructor accepts only 'adam' (optim.Adam) and 'gd' (optim.SGD), both lr-only without weight decay. Any other optimizer string raises NotImplementedError('optimizer {} is not supported!') in __init__, right after the LSTM module is created and moved to device.","triggerScenarios":"LSTMModel(..., optimizer=opt) in qlib/contrib/model/pytorch_lstm_ts.py with opt.lower() not in {'adam','gd'} — e.g. 'sgd', 'adamw', 'adagrad'.","commonSituations":"Sharing optimizer settings across different contrib model classes; expecting the PyTorch optimizer name 'SGD' or 'adamw' to work; typo'd strings.","solutions":["Use optimizer='adam' or optimizer='gd'.","Need another optimizer? Subclass and after super().__init__() assign self.train_optimizer over self.LSTM_model.parameters().","Verify the string in your YAML/kwargs matches exactly (case-insensitive)."],"exampleFix":"# before\nmodel = LSTMModel(..., optimizer=\"adamw\")  # NotImplementedError\n\n# after\nmodel = LSTMModel(..., optimizer=\"adam\")\n# plain SGD:\nmodel = LSTMModel(..., optimizer=\"gd\")","handlingStrategy":"validation","validationCode":"assert optimizer.lower() in {\"adam\", \"gd\"}, \"TS LSTM supports only 'adam' and 'gd'\"\nmodel = LSTMModel(..., optimizer=optimizer)","typeGuard":"def is_supported_optimizer(name: str) -> bool:\n    return isinstance(name, str) and name.lower() in {\"adam\", \"gd\"}","tryCatchPattern":"try:\n    model = LSTMModel(..., optimizer=opt)\nexcept NotImplementedError as e:\n    raise ValueError(f\"{e} — use 'adam' or 'gd'\") from e","preventionTips":["Same rule as sibling models: 'adam' or 'gd' only.","Normalize optimizer names in one config helper across your project.","Validate before constructing to keep failure messages config-focused."],"tags":["pytorch","qlib","optimizer","lstm","config"],"backgroundTag":null,"analyzedSha":"79633dd9506ea689e5400dea0197717b5b3d74b7","analyzedAt":"2026-08-15T07:01:27.511Z","schemaVersion":2},"datasetVersion":"2026-08-15T17:31:12.345Z"}