Hugging Face's logo

aryaadhi
/
Phi-Medical-QA-LoRA

PEFT
TensorBoard
Safetensors
Generated from Trainer
Model card Files Metrics Community
aryaadhi commited on
Commit
f41184f
•
1 Parent(s): e6cf476

End of training

Browse files
Files changed (2) hide show
  1. README.md +43 -19
  2. runs/Jul20_14-40-55_cmle-training-13669393038079472133/events.out.tfevents.1721486456.cmle-training-13669393038079472133 +2 -2
README.md CHANGED
@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  This model is a fine-tuned version of [microsoft/phi-1_5](https://huggingface.co/microsoft/phi-1_5) on the None dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 1.9509
20
 
21
  ## Model description
22
 
@@ -41,29 +41,53 @@ The following hyperparameters were used during training:
41
  - seed: 42
42
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
43
  - lr_scheduler_type: cosine
44
- - num_epochs: 3
45
 
46
  ### Training results
47
 
48
  | Training Loss | Epoch | Step | Validation Loss |
49
  |:-------------:|:------:|:----:|:---------------:|
50
- | 3.1312 | 0.1684 | 100 | 2.7874 |
51
- | 2.6198 | 0.3367 | 200 | 2.5280 |
52
- | 2.4484 | 0.5051 | 300 | 2.3807 |
53
- | 2.3069 | 0.6734 | 400 | 2.2949 |
54
- | 2.2195 | 0.8418 | 500 | 2.2151 |
55
- | 2.2088 | 1.0101 | 600 | 2.1664 |
56
- | 2.0887 | 1.1785 | 700 | 2.1273 |
57
- | 2.0318 | 1.3468 | 800 | 2.0870 |
58
- | 2.0239 | 1.5152 | 900 | 2.0519 |
59
- | 1.9709 | 1.6835 | 1000 | 2.0285 |
60
- | 1.9697 | 1.8519 | 1100 | 2.0116 |
61
- | 1.9416 | 2.0202 | 1200 | 1.9881 |
62
- | 1.8815 | 2.1886 | 1300 | 1.9745 |
63
- | 1.8816 | 2.3569 | 1400 | 1.9661 |
64
- | 1.8641 | 2.5253 | 1500 | 1.9577 |
65
- | 1.843 | 2.6936 | 1600 | 1.9526 |
66
- | 1.8638 | 2.8620 | 1700 | 1.9509 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
67
 
68
 
69
  ### Framework versions
 
16
 
17
  This model is a fine-tuned version of [microsoft/phi-1_5](https://huggingface.co/microsoft/phi-1_5) on the None dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 1.7010
20
 
21
  ## Model description
22
 
 
41
  - seed: 42
42
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
43
  - lr_scheduler_type: cosine
44
+ - num_epochs: 7
45
 
46
  ### Training results
47
 
48
  | Training Loss | Epoch | Step | Validation Loss |
49
  |:-------------:|:------:|:----:|:---------------:|
50
+ | 3.1348 | 0.1684 | 100 | 2.7621 |
51
+ | 2.6302 | 0.3367 | 200 | 2.5110 |
52
+ | 2.4441 | 0.5051 | 300 | 2.3619 |
53
+ | 2.3325 | 0.6734 | 400 | 2.2485 |
54
+ | 2.2262 | 0.8418 | 500 | 2.1819 |
55
+ | 2.1809 | 1.0101 | 600 | 2.1318 |
56
+ | 2.0646 | 1.1785 | 700 | 2.0802 |
57
+ | 2.0665 | 1.3468 | 800 | 2.0541 |
58
+ | 2.0072 | 1.5152 | 900 | 2.0057 |
59
+ | 1.9954 | 1.6835 | 1000 | 1.9762 |
60
+ | 1.9577 | 1.8519 | 1100 | 1.9554 |
61
+ | 1.9263 | 2.0202 | 1200 | 1.9256 |
62
+ | 1.8635 | 2.1886 | 1300 | 1.9027 |
63
+ | 1.855 | 2.3569 | 1400 | 1.8871 |
64
+ | 1.8258 | 2.5253 | 1500 | 1.8750 |
65
+ | 1.8269 | 2.6936 | 1600 | 1.8555 |
66
+ | 1.8194 | 2.8620 | 1700 | 1.8415 |
67
+ | 1.775 | 3.0303 | 1800 | 1.8257 |
68
+ | 1.7379 | 3.1987 | 1900 | 1.8175 |
69
+ | 1.7384 | 3.3670 | 2000 | 1.8052 |
70
+ | 1.74 | 3.5354 | 2100 | 1.7943 |
71
+ | 1.7275 | 3.7037 | 2200 | 1.7778 |
72
+ | 1.6903 | 3.8721 | 2300 | 1.7680 |
73
+ | 1.6908 | 4.0404 | 2400 | 1.7594 |
74
+ | 1.6663 | 4.2088 | 2500 | 1.7559 |
75
+ | 1.6312 | 4.3771 | 2600 | 1.7457 |
76
+ | 1.6412 | 4.5455 | 2700 | 1.7395 |
77
+ | 1.6392 | 4.7138 | 2800 | 1.7327 |
78
+ | 1.6237 | 4.8822 | 2900 | 1.7260 |
79
+ | 1.6138 | 5.0505 | 3000 | 1.7244 |
80
+ | 1.5858 | 5.2189 | 3100 | 1.7205 |
81
+ | 1.6005 | 5.3872 | 3200 | 1.7163 |
82
+ | 1.5662 | 5.5556 | 3300 | 1.7120 |
83
+ | 1.5888 | 5.7239 | 3400 | 1.7075 |
84
+ | 1.5802 | 5.8923 | 3500 | 1.7068 |
85
+ | 1.5659 | 6.0606 | 3600 | 1.7038 |
86
+ | 1.5526 | 6.2290 | 3700 | 1.7039 |
87
+ | 1.54 | 6.3973 | 3800 | 1.7024 |
88
+ | 1.5653 | 6.5657 | 3900 | 1.7018 |
89
+ | 1.545 | 6.7340 | 4000 | 1.7012 |
90
+ | 1.5455 | 6.9024 | 4100 | 1.7010 |
91
 
92
 
93
  ### Framework versions
runs/Jul20_14-40-55_cmle-training-13669393038079472133/events.out.tfevents.1721486456.cmle-training-13669393038079472133 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8f68006f1cea6b68ceca330c140b4a4ade45359ef81b089ce9f56fc07601c008
3
- size 24864
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:453972f9e8d4b2e51513b6885e05cea8510256e88ed8d2940d58ba25d0892206
3
+ size 25489