AttributeError: can't set attribute

[jatinarora96]

Hi, I'm getting this error at the start of training for all the models. Below is the exact error:
`---> 28 ai.train(data, batch_size=32, num_steps=50000, generate_every=5000, save_every=5000)

/miniconda3/envs/test_env/lib/python3.8/site-packages/aitextgen/aitextgen.py in train(self, train_data, output_dir, fp16, fp16_opt_level, n_gpu, tpu_cores, max_grad_norm, gradient_accumulation_steps, seed, learning_rate, weight_decay, adam_epsilon, warmup_steps, num_steps, save_every, generate_every, n_generate, loggers, batch_size, num_workers, benchmark, avg_loss_smoothing, save_gdrive, run_id, progress_bar_refresh_rate, freeze_layers, num_layers_freeze, use_deepspeed, **kwargs)
673 # force single-GPU on Windows
674 if platform.system() == "Windows" and is_gpu_used and n_gpu != 1:
--> 675 logger.warning(
676 "Windows does not support multi-GPU training. Setting to 1 GPU."
677 )

/miniconda3/envs/test_env/lib/python3.8/site-packages/aitextgen/train.py in init(self, model, dataset, hparams, tokenizer)
19 def init(self, model, dataset, hparams, tokenizer):
20 super(ATGTransformer, self).init()
---> 21 self.model, self.dataset, self.hparams, self.tokenizer = (
22 model,
23 dataset,

/miniconda3/envs/test_env/lib/python3.8/site-packages/torch/nn/modules/module.py in setattr(self, name, value)
1311 buffers[name] = value
1312 else:
-> 1313 super().setattr(name, value)
1314
1315 def delattr(self, name):

AttributeError: can't set attribute`
Can someone help in resolving this issue, Thanks in advance

[calumrodger]

I was hitting the same error, having reverted to version 0.5.1 following a previous error which affected generation too, after reading the comments of @mirrorsonthewall and @rmc135 on the discussion thread for the initial issue here: #200

Happily I was able to resolve both issues by applying the fix by @llimllib on PR #202 - as highlighted by @carmocca on the same issue thread mentioned above

to reiterate @carmocca's recommendation: if you're using the colab notebook provided, change the first line !pip install aitextgen to !pip install git+https://github.com/llimllib/aitextgen@fix_tpu_available to install from @llimllib's branch

hopefully that will resolve the problem for you too, and I expect will resolve the issue for good once the PR is merged with the main branch (or until the next dependency update anyway!)