I was able to complete this notebook in kaggle:
I’m trying to run it locally to compare the speeds. I’m erroring out at the following step:
learn = vision_learner(dls, ‘resnet26d’, metrics=error_rate, path=‘.’).to_fp16()
Error:
NameError Traceback (most recent call last) Cell In[16], line 1 ----> 1 learn = vision_learner(dls, ‘resnet26d’, metrics=error_rate, path=‘.’).to_fp16() 2 #learn = vision_learner(dls, ‘levit_384’, metrics=error_rate, path=‘.’).to_fp16() File ~/anaconda3/envs/fastai/lib/python3.11/site-packages/fastai/vision/learner.py:232, in vision_learner(dls, arch, normalize, n_out, pretrained, weights, loss_func, opt_func, lr, splitter, cbs, metrics, path, model_dir, wd, wd_bn_bias, train_bn, moms, cut, init, custom_head, concat_pool, pool, lin_ftrs, ps, first_bn, bn_final, lin_first, y_range, **kwargs) 230 n_in = kwargs[‘n_in’] if ‘n_in’ in kwargs else 3 231 if isinstance(arch, str): → 232 model,cfg = create_timm_model(arch, n_out, default_split, pretrained, **model_args) 233 if normalize: _timm_norm(dls, cfg, pretrained, n_in) 234 else: File ~/anaconda3/envs/fastai/lib/python3.11/site-packages/fastai/vision/learner.py:191, in create_timm_model(arch, n_out, cut, pretrained, n_in, init, custom_head, concat_pool, pool, lin_ftrs, ps, first_bn, bn_final, lin_first, y_range, **kwargs) 188 def create_timm_model(arch, n_out, cut=None, pretrained=True, n_in=3, init=nn.in
NameError: name ‘timm’ is not defined