Dangal Details
Florence Rocle
Dangal is based on the true story of Mahavir Singh Phogat (played by Aamir Khan) and his two daughters, Geeta and Babita Phogat (played by Fatima Sana Shaikh and Sanya Malhotra, respectively). Mahavir was a wrestler (dangal in Hindi), who was never able to achieve his goal of bringing home a gold medal for his country of India. So when his daughters showed a natural talent for fighting, he became their coach and raised them as champion wrestlers. In 2010, Geeta won a gold medal at the Commonwealth Games, and in 2012 was the first female Indian wrestler to qualify for the Olympics. Babita, her younger sister, also won a gold medal at the 2014 Commonwealth Games.
All articles published by MDPI are made immediately available worldwide under an open access license. No special permission is required to reuse all or part of the article published by MDPI, including figures and tables. For articles published under an open access Creative Common CC BY license, any part of the article may be reused without permission provided that the original article is clearly cited. For more information, please refer to
Feature papers represent the most advanced research with significant potential for high impact in the field. A Feature Paper should be a substantial original Article that involves several techniques or approaches, provides an outlook for future research directions and describes possible research applications.
In this manuscript the authors evaluated various spectral datasets from soil samples to ascertain the need for spectral standardization. This study is a good first step towards the application of diffuse reflectance spectroscopy in analyzing vast amount of soil samples given the growing need to develop cost effective methods. These vast data sets across different continents can be analyzed by using clever machine learning approaches.
Comments: In this manuscript the authors evaluated various spectral datasets from soil samples to ascertain the need for spectral standardization. This study is a good first step towards the application of diffuse reflectance spectroscopy in analyzing vast amount of soil samples given the growing need to develop cost effective methods. These vast data sets across different continents can be analyzed by using clever machine learning approaches.
Response: We thank the reviewer for the comments. We did not see any comments from the reviewer that need to be addressed. We have revised the introduction and research design to improve the flow of manuscript.
This work presented the assessment of using USDA predictive models to identify correctness utilizing existing spectral libraries. All data showed in the manuscript were sufficient and the explanations seemed reasonable. The subject matter is appropriate and the quality of the presentation is adequate. Before it can be recommended to accept for publication in the Journal, a revision is needed addressed
Response: We thank the reviewer for the response. We have revised the conclusion section (particularly the first paragraph) to clarify that although good predictions were possible without calibration transfer, PDS was necessary to get unbiased predictions.
Comment 3: Although the manuscript title is short and simply, however it was difficult to understand clearly and directly. Suggest the title rewriting in order to reflect the text and meaning of the manuscript.
Response: We have updated the old citations with the new ones. However, in some instances, we have to keep the old references because some of the calibration transfer techniques originate from these papers.
The manuscript tackles an important subject of soil spectral standardization and calibration transfer. Contribution to new knowledge by the manuscript has efficiently spelt out, the Materials and Methods section is well-written, the Results and Discussion section is informative enough and the results have been well-discussed. However, some minor improvments need to be done in Abstract and Introduction before the publication of the manuscript.
- The Abstract is not well informative and more details need to be provided. For instance, Please clarify what the European test set is (Ln. 20). Which "soil properties" (Ln. 25)? Which spectral preprocessing tecniques did you use?
- Results have been provided and interpreted very well with a strong discussion. However, I do not know why the Tables and Figures have not been included in the text, but provided as suplemantary materials.
Comment 1: The Abstract is not well informative and more details need to be provided. For instance, Please clarify what the European test set is (Ln. 20). Which "soil properties" (Ln. 25)? Which spectral preprocessing tecniques did you use?
Response: We thank the reviewer for the comments. We have revised the abstract to include the description on European test set and also provided the list of soil properties used for prediction with and without calibration transfer.
Response: In the introduction, we have clearly provided information on different calibration transfer techniques and their limitation. There has been mixed results on the performance of predictive models with and without calibration transfer. Therefore, we try to address this issue by using the PDS techniques across samples of different origin to examine whether calibration transfer is necessary. We have revised the final paragraph of the introduction section to include the knowledge gap as provided below:
Comment 4: Results have been provided and interpreted very well with a strong discussion. However, I do not know why the Tables and Figures have not been included in the text, but provided as suplemantary materials.
Response: We appreciate the reviewer comments. We have provided figures relevant to the main findings of the study (Fig 2, 3, 4) as the text. The tables and figures in the supplementary materials mostly support the findings related to figs 2, 3 and 4.
The study compares predictions obtained by means of a secondary spectrometer, with and without PDS, with those obtained from primary spectrometer (i.e. that used in calibration). Two pre-treatment methods, three prediction algorithms and five target variables were considered. Two independent valiadation sets were used to test the results.
Presentation is generally clear, reference list and graphic material (included supplementary plots and tables) is adequate. I have detected only few unclear points requiring some fixing, which are listed below. Therefore, only a minor revision is required.
The reasoning at lines 46-48 is not very clear. I agree that impact of particle size increases at shorter wavelengths, but I do not understand why such issue should be more acute in the MIR (which has longer wavelengths) than in the NIR. The concept shoud be rephrased.
At lines 87-88 derivative pre-processing is suggested reducing noise method. It looks a bit odd. It is true that derivative remove baseline shift, but, in my experience, derivative spectra tend to be more noisy than original ones, even using smoothing windows. Such point should be addressed more clearly.
Another unclear point is at lines 186-187 (outlier detection). What exactly mean "to pick up the maximum of 1% of poorly performing samples". That at most 1% of the whole dataset is removed? Or that only 1% of poorly performing samples is removed? In the latter case there should be two thresholds: one for defining poorly performing samples and one for defining the effectively removed samples. I have also read Ref. [15], but it is not more clear. I suggest to rephrase this concept more clearly.
Comment 1: The reasoning at lines 46-48 is not very clear. I agree that impact of particle size increases at shorter wavelengths, but I do not understand why such issue should be more acute in the MIR (which has longer wavelengths) than in the NIR. The concept shoud be rephrased.
Response: Thank you for pointing out this issue. We have removed lines 46-48 and focused the study on differences in spectral responses due to environmental conditions and instrumental differences.
Comment 3: At lines 87-88 derivative pre-processing is suggested reducing noise method. It looks a bit odd. It is true that derivative remove baseline shift, but, in my experience, derivative spectra tend to be more noisy than original ones, even using smoothing windows. Such point should be addressed more clearly.