Bearnardd commited on
Commit
9688d7a
1 Parent(s): cdad7d4

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +124 -0
README.md ADDED
@@ -0,0 +1,124 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ license: apache-2.0
5
+ library_name: timm
6
+ tags:
7
+ - mobile
8
+ - vison
9
+ - image-classification
10
+ datasets:
11
+ - imagenet-1k
12
+ metrics:
13
+ - accuracy
14
+ ---
15
+
16
+ # EfficientFormer-L1
17
+
18
+ ## Table of Contents
19
+ - [EfficientFormer-L1](#-model_id--defaultmymodelname-true)
20
+ - [Table of Contents](#table-of-contents)
21
+ - [Model Details](#model-details)
22
+ - [How to Get Started with the Model](#how-to-get-started-with-the-model)
23
+ - [Uses](#uses)
24
+ - [Direct Use](#direct-use)
25
+ - [Downstream Use](#downstream-use)
26
+ - [Misuse and Out-of-scope Use](#misuse-and-out-of-scope-use)
27
+ - [Limitations and Biases](#limitations-and-biases)
28
+ - [Training](#training)
29
+ - [Training Data](#training-data)
30
+ - [Training Procedure](#training-procedure)
31
+ - [Evaluation Results](#evaluation-results)
32
+ - [Environmental Impact](#environmental-impact)
33
+ - [Citation Information](#citation-information)
34
+
35
+
36
+ <model_details>
37
+
38
+ ## Model Details
39
+
40
+ <!-- Give an overview of your model, the relevant research paper, who trained it, etc. -->
41
+
42
+ EfficientFormer-L1, developed by [Snap Research](https://github.com/snap-research), is one of three EfficientFormer models. The EfficientFormer models were released as part of an effort to prove that properly designed transformers can reach extremely low latency on mobile devices while maintaining high performance.
43
+
44
+ This checkpoint of EfficientFormer-L1 was trained for 1000 epochs.
45
+
46
+ - Developed by: Yanyu Li, Geng Yuan, Yang Wen, Eric Hu, Georgios Evangelidis, Sergey Tulyakov, Yanzhi Wang, Jian Ren
47
+ - Language(s): English
48
+ - License: This model is licensed under the apache-2.0 license
49
+ - Resources for more information:
50
+ - [Research Paper](https://arxiv.org/abs/2206.01191)
51
+ - [GitHub Repo](https://github.com/snap-research/EfficientFormer/)
52
+
53
+ </model_details>
54
+
55
+ <how_to_start>
56
+
57
+ ## How to Get Started with the Model
58
+
59
+ Use the code below to get started with the model.
60
+
61
+ ```python
62
+ # A nice code snippet here that describes how to use the model...
63
+ ```
64
+ </how_to_start>
65
+
66
+ <uses>
67
+
68
+ ## Uses
69
+
70
+ #### Direct Use
71
+
72
+ This model can be used for image classification and semantic segmentation. On mobile devices (the model was tested on iPhone 12), the CoreML checkpoints will perform these tasks with low latency.
73
+
74
+ <Limitations_and_Biases>
75
+
76
+ ## Limitations and Biases
77
+
78
+ Though most designs in EfficientFormer are general-purposed, e.g., dimension- consistent design and 4D block with CONV-BN fusion, the actual speed of EfficientFormer may vary on other platforms. For instance, if GeLU is not well supported while HardSwish is efficiently implemented on specific hardware and compiler, the operator may need to be modified accordingly. The proposed latency-driven slimming is simple and fast. However, better results may be achieved if search cost is not a concern and an enumeration-based brute search is performed.
79
+
80
+ Since the model was trained on Imagenet-1K, the [biases embedded in that dataset](https://huggingface.co/datasets/imagenet-1k#considerations-for-using-the-data) will be reflected in the EfficientFormer models.
81
+
82
+ </Limitations_and_Biases>
83
+
84
+ <Training>
85
+
86
+ ## Training
87
+
88
+ #### Training Data
89
+
90
+ This model was trained on ImageNet-1K.
91
+
92
+ See the [data card](https://huggingface.co/datasets/imagenet-1k) for additional information.
93
+
94
+ #### Training Procedure
95
+
96
+ * Parameters: 12.3 M
97
+ * GMACs: 1.3
98
+ * Train. Epochs: 1000
99
+
100
+ Trained on a cluster with NVIDIA A100 and V100 GPUs.
101
+
102
+ </Training>
103
+
104
+ <Eval_Results>
105
+
106
+ ## Evaluation Results
107
+
108
+ Top-1 Accuracy: 80.2% on ImageNet 10K
109
+
110
+ </Eval_Results>
111
+
112
+ <Cite>
113
+
114
+ ## Citation Information
115
+
116
+ ```bibtex
117
+ @article{li2022efficientformer,
118
+ title={EfficientFormer: Vision Transformers at MobileNet Speed},
119
+ author={Li, Yanyu and Yuan, Geng and Wen, Yang and Hu, Eric and Evangelidis, Georgios and Tulyakov, Sergey and Wang, Yanzhi and Ren, Jian},
120
+ journal={arXiv preprint arXiv:2206.01191},
121
+ year={2022}
122
+ }
123
+ ```
124
+ </Cite>